Search Results

Search found 16914 results on 677 pages for 'single threaded'.

Page 293/677 | < Previous Page | 289 290 291 292 293 294 295 296 297 298 299 300  | Next Page >

  • Can Microsoft Build Appliances?

    - by andrewbrust
    Billy Hollis, my Visual Studio Live! colleague and fellow Microsoft Regional Director said recently, and I am paraphrasing, that the computing world, especially on the consumer side, has shifted from one of building hardware and software that makes things possible to do, to building products and technologies that make things easy to do.  Billy crystalized things perfectly, as he often does. In this new world of “easy to do,” Apple has done very well and Microsoft has struggled.  In the old world, customers wanted a Swiss Army Knife, with the most gimmicks and gadgets possible.  In the new world, people want elegantly cutlery.  They may want cake cutters and utility knives too, but they don’t want one device that works for all three tasks.  People don’t want tools, they want utensils.  People don’t want machines.  They want appliances. Microsoft Appliances: They Do Exist Microsoft has built a few appliance-like devices.  I would say XBox 360 is an appliance,  It’s versatile, mind you, but it’s the kind of thing you plug in, turn on and use, as opposed to set-up, tune, and open up to upgrade the internals.  Windows Phone 7 is an appliance too.  It’s a true smartphone, unlike Windows Mobile which was a handheld computer with a radio stack.  Zune is an appliance too, and a nice one.  It hasn’t attained much traction in the market, but that’s probably because the seminal consumer computing appliance -- the iPod – got there so much more quickly. In the embedded world, Mediaroom, Microsoft’s set-top product for the cable industry (used by AT&T U-Verse and others) is an appliance.  So is Microsoft’s Sync technology, used in Ford automobiles.  Even on the enterprise side, Microsoft has an appliance: SQL Server Parallel Data Warehouse Edition (PDW) combines Microsoft software with select OEMs’ server, networking and storage hardware.  You buy the appliance units from the OEMs, plug them in, connect them and go. I would even say that Bing is an appliance.  Not in the hardware sense, mind you.  But from the software perspective, it’s a single-purpose product that you visit or run, use and then move on.  You don’t have to install it (except the iOS and Android native apps where it’s pretty straightforward), you don’t have to customize it, you don’t have to program it.  Basically, you just use it. Microsoft Appliances that Should Exist But Microsoft builds a bunch of things that are not appliances.  Media Center is not an appliance, and it most certainly should be.  Instead, it’s an app that runs on Windows 7.  It runs full-screen and you can use this configuration to conceal the fact that Windows is under it, but eventually something will cause you to abandon that masquerade (like Patch Tuesday). The next version of Windows Home Server won’t, in my opinion, be an appliance either.  Now that the Drive Extender technology is gone, and users can’t just add and remove drives into and from a single storage pool, the product is much more like a IT server and less like an appliance-premised one.  Much has been written about this decision by Microsoft.  I’ll just sum it up in one word: pity. Microsoft doesn’t have anything remotely appliance-like in the tablet category, either.  Until it does, it likely won’t have much market share in that space either.  And of course, the bulk of Microsoft’s product catalog on the business side is geared to enterprise machines and not personal appliances. Appliance DNA: They Gotta Have It. The consumerization of IT is real, because businesspeople are consumers too.  They appreciate the fit and finish of appliances at home, and they increasingly feel entitled to have it at work too.  Secure and reliable push email in a smartphone is necessary, but it isn’t enough.  People want great apps and a pleasurable user experience too.  The full Microsoft Office product is needed at work, but a PC with a keyboard and mouse, or maybe a touch screen that uses a stylus (or requires really small fingers), to run Office isn’t enough either.  People want a flawless touch experience available for the times they want to read and take quick notes.  Until Microsoft realizes this fully and internalizes it, it will suffer defeats in the consumer market and even setbacks in the business market.  Think about how slow the Office upgrade cycle is…now imagine if the next version of Office had a first-class alternate touch UI and consider the possible acceleration in adoption rates. Can Microsoft make the appliance switch?  Can the appliance mentality become pervasive at the company?  Can Microsoft hasten its release cycles dramatically and shed the “some assembly required” paradigm upon which many of its products are based?  Let’s face it, the chances that Microsoft won’t make this transition are significant. But there are also encouraging signs, and they should not be ignored.  The appliances we have already discussed, especially Xbox, Zune and Windows Phone 7, are the most obvious in this regard.  The fact that SQL Server has an appliance SKU now is a more subtle but perhaps also more significant outcome, because that product sits so smack in the middle of Microsoft’s enterprise stack.  Bing is encouraging too, especially given its integrated travel, maps and augmented reality capabilities.  As Bing gains market share, Microsoft has tangible proof that it can transform and win, even when everyone outside the company, and many within it, would bet otherwise. That Great Big Appliance in the Sky Perhaps the most promising (and evolving) proof points toward the appliance mentality, though, are Microsoft’s cloud offerings -- Azure and BPOS/Office 365.  While the cloud does not represent a physical appliance (quite the opposite in fact) its ability to make acquisition, deployment and use of technology simple for the user is absolutely an embodiment of the appliance mentality and spirit.  Azure is primarily a platform as a service offering; it doesn’t just provide infrastructure.  SQL Azure does likewise for databases.  And Office 365 does likewise for SharePoint, Exchange and Lync. You don’t administer, tune and manage servers; instead, you create databases or site collections or mailboxes and start using them. Upgrades come automatically, and it seems like releases will come more frequently.  Fault tolerance and content distribution is just there.  No muss.  No fuss.  You use these services; you don’t have to set them up and think about them.  That’s how appliances work.  To me, these signs point out that Microsoft has the full capability of transforming itself.  But there’s a lot of work ahead.  Microsoft may say they’re “all in” on the cloud, but the majority of the company is still oriented around its old products and models.  There needs to be a wholesale cultural transformation in Redmond.  It can happen, but product management, program management, the field and executive ranks must unify in the effort. So must partners, and even customers.  New leaders must rise up and Microsoft must be able to see itself as a winner.  If Microsoft does this, it could lock-in decades of new success, and be a standard business school case study for doing so.  If not, the company will have missed an opportunity, and may see its undoing.

    Read the article

  • The Application Architecture Domain

    - by Michael Glas
    I have been spending a lot of time thinking about Application Architecture in the context of EA. More specifically, as an Enterprise Architect, what do I need to consider when looking at/defining/designing the Application Architecture Domain?There are several definitions of Application Architecture. TOGAF says “The objective here [in Application Architecture] is to define the major kinds of application system necessary to process the data and support the business”. FEA says the Application Architecture “Defines the applications needed to manage the data and support the business functions”.I agree with these definitions. They reflect what the Application Architecture domain does. However, they need to be decomposed to be practical.I find it useful to define a set of views into the Application Architecture domain. These views reflect what an EA needs to consider when working with/in the Applications Architecture domain. These viewpoints are, at a high level:Capability View: This view reflects how applications alignment with business capabilities. It is a super set of the following views when viewed in aggregate. By looking at the Application Architecture domain in terms of the business capabilities it supports, you get a good perspective on how those applications are directly supporting the business.Technology View: The technology view reflects the underlying technology that makes up the applications. Based on the number of rationalization activities I have seen (more specifically application rationalization), the phrase “complexity equals cost” drives the importance of the technology view, especially when attempting to reduce that complexity through standardization type activities. Some of the technology components to be considered are: Software: The application itself as well as the software the application relies on to function (web servers, application servers). Infrastructure: The underlying hardware and network components required by the application and supporting application software. Development: How the application is created and maintained. This encompasses development components that are part of the application itself (i.e. customizable functions), as well as bolt on development through web services, API’s, etc. The maintenance process itself also falls under this view. Integration: The interfaces that the application provides for integration as well as the integrations to other applications and data sources the application requires to function. Type: Reflects the kind of application (mash-up, 3 tiered, etc). (Note: functional type [CRM, HCM, etc.] are reflected under the capability view). Organization View: Organizations are comprised of people and those people use applications to do their jobs. Trying to define the application architecture domain without taking the organization that will use/fund/change it into consideration is like trying to design a car without thinking about who will drive it (i.e. you may end up building a formula 1 car for a family of 5 that is really looking for a minivan). This view reflects the people aspect of the application. It includes: Ownership: Who ‘owns’ the application? This will usually reflect primary funding and utilization but not always. Funding: Who funds both the acquisition/creation as well as the on-going maintenance (funding to create/change/operate)? Change: Who can/does request changes to the application and what process to the follow? Utilization: Who uses the application, how often do they use it, and how do they use it? Support: Which organization is responsible for the on-going support of the application? Information View: Whether or not you subscribe to the view that “information drives the enterprise”, it is a fact that information is critical. The management, creation, and organization of that information are primary functions of enterprise applications. This view reflects how the applications are tied to information (or at a higher level – how the Application Architecture domain relates to the Information Architecture domain). It includes: Access: The application is the mechanism by which end users access information. This could be through a primary application (i.e. CRM application), or through an information access type application (a BI application as an example). Creation: Applications create data in order to provide information to end-users. (I.e. an application creates an order to be used by an end-user as part of the fulfillment process). Consumption: Describes the data required by applications to function (i.e. a product id is required by a purchasing application to create an order. Application Service View: Organizations today are striving to be more agile. As an EA, I need to provide an architecture that supports this agility. One of the primary ways to achieve the required agility in the application architecture domain is through the use of ‘services’ (think SOA, web services, etc.). Whether it is through building applications from the ground up utilizing services, service enabling an existing application, or buying applications that are already ‘service enabled’, compartmentalizing application functions for re-use helps enable flexibility in the use of those applications in support of the required business agility. The applications service view consists of: Services: Here, I refer to the generic definition of a service “a set of related software functionalities that can be reused for different purposes, together with the policies that should control its usage”. Functions: The activities within an application that are not available / applicable for re-use. This view is helpful when identifying duplication functions between applications that are not service enabled. Delivery Model View: It is hard to talk about EA today without hearing the terms ‘cloud’ or shared services.  Organizations are looking at the ways their applications are delivered for several reasons, to reduce cost (both CAPEX and OPEX), to improve agility (time to market as an example), etc.  From an EA perspective, where/how an application is deployed has impacts on the overall enterprise architecture. From integration concerns to SLA requirements to security and compliance issues, the Enterprise Architect needs to factor in how applications are delivered when designing the Enterprise Architecture. This view reflects how applications are delivered to end-users. The delivery model view consists of different types of delivery mechanisms/deployment options for applications: Traditional: Reflects non-cloud type delivery options. The most prevalent consists of an application running on dedicated hardware (usually specific to an environment) for a single consumer. Private Cloud: The application runs on infrastructure provisioned for exclusive use by a single organization comprising multiple consumers. Public Cloud: The application runs on infrastructure provisioned for open use by the general public. Hybrid: The application is deployed on two or more distinct cloud infrastructures (private, community, or public) that remain unique entities, but are bound together by standardized or proprietary technology that enables data and application portability. While by no means comprehensive, I find that applying these views to the application domain gives a good understanding of what an EA needs to consider when effecting changes to the Application Architecture domain.Finally, the application architecture domain is one of several architecture domains that an EA must consider when developing an overall Enterprise Architecture. The Oracle Enterprise Architecture Framework defines four Primary domains: Business Architecture, Application Architecture, Information Architecture, and Technology Architecture. Each domain links to the others either directly or indirectly at some point. Oracle links them at a high level as follows:Business Capabilities and/or Business Processes (Business Architecture), links to the Applications that enable the capability/process (Applications Architecture – COTS, Custom), links to the Information Assets managed/maintained by the Applications (Information Architecture), links to the technology infrastructure upon which all this runs (Technology Architecture - integration, security, BI/DW, DB infrastructure, deployment model). There are however, times when the EA needs to narrow focus to a particular domain for some period of time. These views help me to do just that.

    Read the article

  • Premature-Optimization and Performance Anxiety

    - by James Michael Hare
    While writing my post analyzing the new .NET 4 ConcurrentDictionary class (here), I fell into one of the classic blunders that I myself always love to warn about.  After analyzing the differences of time between a Dictionary with locking versus the new ConcurrentDictionary class, I noted that the ConcurrentDictionary was faster with read-heavy multi-threaded operations.  Then, I made the classic blunder of thinking that because the original Dictionary with locking was faster for those write-heavy uses, it was the best choice for those types of tasks.  In short, I fell into the premature-optimization anti-pattern. Basically, the premature-optimization anti-pattern is when a developer is coding very early for a perceived (whether rightly-or-wrongly) performance gain and sacrificing good design and maintainability in the process.  At best, the performance gains are usually negligible and at worst, can either negatively impact performance, or can degrade maintainability so much that time to market suffers or the code becomes very fragile due to the complexity. Keep in mind the distinction above.  I'm not talking about valid performance decisions.  There are decisions one should make when designing and writing an application that are valid performance decisions.  Examples of this are knowing the best data structures for a given situation (Dictionary versus List, for example) and choosing performance algorithms (linear search vs. binary search).  But these in my mind are macro optimizations.  The error is not in deciding to use a better data structure or algorithm, the anti-pattern as stated above is when you attempt to over-optimize early on in such a way that it sacrifices maintainability. In my case, I was actually considering trading the safety and maintainability gains of the ConcurrentDictionary (no locking required) for a slight performance gain by using the Dictionary with locking.  This would have been a mistake as I would be trading maintainability (ConcurrentDictionary requires no locking which helps readability) and safety (ConcurrentDictionary is safe for iteration even while being modified and you don't risk the developer locking incorrectly) -- and I fell for it even when I knew to watch out for it.  I think in my case, and it may be true for others as well, a large part of it was due to the time I was trained as a developer.  I began college in in the 90s when C and C++ was king and hardware speed and memory were still relatively priceless commodities and not to be squandered.  In those days, using a long instead of a short could waste precious resources, and as such, we were taught to try to minimize space and favor performance.  This is why in many cases such early code-bases were very hard to maintain.  I don't know how many times I heard back then to avoid too many function calls because of the overhead -- and in fact just last year I heard a new hire in the company where I work declare that she didn't want to refactor a long method because of function call overhead.  Now back then, that may have been a valid concern, but with today's modern hardware even if you're calling a trivial method in an extremely tight loop (which chances are the JIT compiler would optimize anyway) the results of removing method calls to speed up performance are negligible for the great majority of applications.  Now, obviously, there are those coding applications where speed is absolutely king (for example drivers, computer games, operating systems) where such sacrifices may be made.  But I would strongly advice against such optimization because of it's cost.  Many folks that are performing an optimization think it's always a win-win.  That they're simply adding speed to the application, what could possibly be wrong with that?  What they don't realize is the cost of their choice.  For every piece of straight-forward code that you obfuscate with performance enhancements, you risk the introduction of bugs in the long term technical debt of the application.  It will become so fragile over time that maintenance will become a nightmare.  I've seen such applications in places I have worked.  There are times I've seen applications where the designer was so obsessed with performance that they even designed their own memory management system for their application to try to squeeze out every ounce of performance.  Unfortunately, the application stability often suffers as a result and it is very difficult for anyone other than the original designer to maintain. I've even seen this recently where I heard a C++ developer bemoaning that in VS2010 the iterators are about twice as slow as they used to be because Microsoft added range checking (probably as part of the 0x standard implementation).  To me this was almost a joke.  Twice as slow sounds bad, but it almost never as bad as you think -- especially if you're gaining safety.  The only time twice is really that much slower is when once was too slow to begin with.  Think about it.  2 minutes is slow as a response time because 1 minute is slow.  But if an iterator takes 1 microsecond to move one position and a new, safer iterator takes 2 microseconds, this is trivial!  The only way you'd ever really notice this would be in iterating a collection just for the sake of iterating (i.e. no other operations).  To my mind, the added safety makes the extra time worth it. Always favor safety and maintainability when you can.  I know it can be a hard habit to break, especially if you started out your career early or in a language such as C where they are very performance conscious.  But in reality, these type of micro-optimizations only end up hurting you in the long run. Remember the two laws of optimization.  I'm not sure where I first heard these, but they are so true: For beginners: Do not optimize. For experts: Do not optimize yet. This is so true.  If you're a beginner, resist the urge to optimize at all costs.  And if you are an expert, delay that decision.  As long as you have chosen the right data structures and algorithms for your task, your performance will probably be more than sufficient.  Chances are it will be network, database, or disk hits that will be your slow-down, not your code.  As they say, 98% of your code's bottleneck is in 2% of your code so premature-optimization may add maintenance and safety debt that won't have any measurable impact.  Instead, code for maintainability and safety, and then, and only then, when you find a true bottleneck, then you should go back and optimize further.

    Read the article

  • The Business of Winning Innovation: An Exclusive Blog Series

    - by Kerrie Foy
    "The Business of Winning Innovation” is a series of articles authored by Oracle Agile PLM experts on what it takes to make innovation a successful and lucrative competitive advantage. Our customers have proven Agile PLM applications to be enormously flexible and comprehensive, so we’ve launched this article series to showcase some of the most fascinating, value-packed use cases. In this article by Keith Colonna, we kick-off the series by taking a look at the science side of innovation within the Consumer Products industry and how PLM can help companies innovate faster, cheaper, smarter. This article will review how innovation has become the lifeline for growth within consumer products companies and how certain companies are “winning” by creating a competitive advantage for themselves by taking a more enterprise-wide,systematic approach to “innovation”.   Managing the Science of Innovation within the Consumer Products Industry By: Keith Colonna, Value Chain Solution Manager, Oracle The consumer products (CP) industry is very mature and competitive. Most companies within this industry have saturated North America (NA) with their products thus maximizing their NA growth potential. Future growth is expected to come from either expansion outside of North America and/or by way of new ideas and products. Innovation plays an integral role in both of these strategies, whether you’re innovating business processes or the products themselves, and may cause several challenges for the typical CP company, Becoming more innovative is both an art and a science. Most CP companies are very good at the art of coming up with new innovative ideas, but many struggle with perfecting the science aspect that involves the best practice processes that help companies quickly turn ideas into sellable products and services. Symptoms and Causes of Business Pain Struggles associated with the science of innovation show up in a variety of ways, like: · Establishing and storing innovative product ideas and data · Funneling these ideas to the chosen few · Time to market cycle time and on-time launch rates · Success rates, or how often the best idea gets chosen · Imperfect decision making (i.e. the ability to kill projects that are not projected to be winners) · Achieving financial goals · Return on R&D investment · Communicating internally and externally as more outsource partners are added globally · Knowing your new product pipeline and project status These challenges (and others) can be consolidated into three root causes: A lack of visibility Poor data with limited access The inability to truly collaborate enterprise-wide throughout your extended value chain Choose the Right Remedy Product Lifecycle Management (PLM) solutions are uniquely designed to help companies solve these types challenges and their root causes. However, PLM solutions can vary widely in terms of configurability, functionality, time-to-value, etc. Business leaders should evaluate PLM solution in terms of their own business drivers and long-term vision to determine the right fit. Many of these solutions are point solutions that can help you cure only one or two business pains in the short term. Others have been designed to serve other industries with different needs. Then there are those solutions that demo well but are owned by companies that are either unable or unwilling to continuously improve their solution to stay abreast of the ever changing needs of the CP industry to grow through innovation. What the Right PLM Solution Should Do for You Based on more than twenty years working in the CP industry, I recommend investing in a single solution that can help you solve all of the issues associated with the science of innovation in a totally integrated fashion. By integration I mean the (1) integration of the all of the processes associated with the development, maintenance and delivery of your product data, and (2) the integration, or harmonization of this product data with other downstream sources, like ERP, product catalogues and the GS1 Global Data Synchronization Network (or GDSN, which is now a CP industry requirement for doing business with most retailers). The right PLM solution should help you: Increase Revenue. A best practice PLM solution should help a company grow its revenues by consolidating product development cycle-time and helping companies get new and improved products to market sooner. PLM should also eliminate many of the root causes for a product being returned, refused and/or reclaimed (which takes away from top-line growth) by creating an enterprise-wide, collaborative, workflow-driven environment. Reduce Costs. A strong PLM solution should help shave many unnecessary costs that companies typically take for granted. Rationalizing SKU’s, components (ingredients and packaging) and suppliers is a major opportunity at most companies that PLM should help address. A natural outcome of this rationalization is lower direct material spend and a reduction of inventory. Another cost cutting opportunity comes with PLM when it helps companies avoid certain costs associated with process inefficiencies that lead to scrap, rework, excess and obsolete inventory, poor end of life administration, higher cost of quality and regulatory and increased expediting. Mitigate Risk. Risks are the hardest to quantify but can be the most costly to a company. Food safety, recalls, line shutdowns, customer dissatisfaction and, worst of all, the potential tarnishing of your brands are a few of the debilitating risks that CP companies deal with on a daily basis. These risks are so uniquely severe that they require an enterprise PLM solution specifically designed for the CP industry that safeguards product information and processes while still allowing the art of innovation to flourish. Many CP companies have already created a winning advantage by leveraging a single, best practice PLM solution to establish an enterprise-wide, systematic approach to innovation. Oracle’s Answer for the Consumer Products Industry Oracle is dedicated to solving the growth and innovation challenges facing the CP industry. Oracle’s Agile Product Lifecycle Management for Process solution was originally developed with and for CP companies and is driven by a specialized development staff solely focused on maintaining and continuously improving the solution per the latest industry requirements. Agile PLM for Process helps CP companies handle all of the processes associated with managing the science of the innovation process, including: specification management, new product development/project and portfolio management, formulation optimization, supplier management, and quality and regulatory compliance to name a few. And as I mentioned earlier, integration is absolutely critical. Many Oracle CP customers, both with Oracle ERP systems and non-Oracle ERP systems, report benefits from Oracle’s Agile PLM for Process. In future articles we will explain in greater detail how both existing Oracle customers (like Gallo, Smuckers, Land-O-Lakes and Starbucks) and new Oracle customers (like ConAgra, Tyson, McDonalds and Heinz) have all realized the benefits of Agile PLM for Process and its integration to their ERP systems. More to Come Stay tuned for more articles in our blog series “The Business of Winning Innovation.” While we will also feature articles focused on other industries, look forward to more on how Agile PLM for Process addresses innovation challenges facing the CP industry. Additional topics include: Innovation Data Management (IDM), New Product Development (NPD), Product Quality Management (PQM), Menu Management,Private Label Management, and more! . Watch this video for more info about Agile PLM for Process

    Read the article

  • Getting your bearings and defining the project objective

    - by johndoucette
    I wrote this two years ago and thought it was worth posting… Some may think this is a daunting task and some may even say “what a waste of time” and want to open MS Project and start typing out tasks because someone asked for an estimate and a task list. Hell, maybe you even use Excel and pump out a spreadsheet with some real scientific formula for guessing how long it will take to code a bunch of classes. However, this short exercise will provide the basis for the entire project, whether small or large and be a great friend when communicating to anyone on your team or even your client. I call this the Project Brief. If you find yourself going beyond a single page, then you must decompose the sections and summarize your findings so there is a complete and clear picture of the project you are working on in a relatively short statement. Here is a great quote from the PMBOK (Project Management Body of Knowledge) relative to what a project is;   A project is a temporary endeavor undertaken to create a unique product, service or result. With this in mind, the project brief should encompass the entirety (objective) of the endeavor in its explanation and what it will take (goals) to create the product, service or result (deliverables). Normally the process of identifying the project objective is done during the first stage of a project called the Project Kickoff, but you can perform this very important step anytime to help you get a bearing. There are many more parts to helping a project stay on course, but this is usually the foundation where it can be grounded on. Through a series of 3 exercises, you should be able to come up with the objective, goals and deliverables on your project. Follow these steps, and in no time (about &frac12; hour), you will have the foundation of your project plan. (See examples below) Exercise 1 – Objectives Begin with the end in mind. Think about your project in business terms with a couple things to help you understand the objective; Reference the business benefit in terms of cost, speed and / or quality, Provide a higher level of what the outcome will look like (future sense) It should be non-measurable, that’s what the goals are all about The output should be a single paragraph with three sentences and take 10 minutes to write. *Typically, agreement must be reached on the objectives of the project before you would proceed to the next steps of the project. Exercise 2 – Goals A project goal is a statement that answers questions about who, what, why, where and when. A good project goal statement; Answers the five “W” questions for the project Is measurable in each of its parts Is published and agreed on by all the owners This helps the Project Manager receive confirmation on defining the project target. Using the established project objective done in the first exercise, think about the things it will take to get the job done. Think about tangible activities which are the top level tasks in a typical Work Breakdown Structure (WBS). The overall goal statement plus all the deliverables (next exercise) can be seen as the project team’s contract with the project owners. Write 3 - 5 goals in about 10 minutes. You should not write the words “Who, what, why, where and when, but merely be able to answer the questions when you read a goal. Exercise 3 – Deliverables Every project creates some type of output and these outputs are called deliverables. There are two classes of deliverables; Internal – produced for project team members to meet their goals External – produced for project owners to meet their expectations The list you enter here provides a checklist for the team’s delivery and/or is a statement of all the expectations of the project owners. Here are some typical project deliverables; Product and product documentation End product/system Requirements/feature documents Installation guides Demo/prototype System design documents User guides/help files Plans Project plan Training plan Conversion/installation/delivery plan Test plans Documentation plan Communication plan Reports and general documentation Progress reports System acceptance tests Outstanding bug list Procedures Risk and issue logs Project history Deliverables should go with each of the goals. Have 3-5 deliverables for each goal. When you are done, you will have established a great foundation for the clarity of your project. This exercise can take some time, but with practice, you should be able to whip this one out in 10 minutes as well, especially if you are intimate with an ongoing project. Samples  Objective [Client] is implementing a series of MOSS sites to support external public (Internet), internal employee (Intranet) and an external secure (password protected Internet) applications. This project will focus on the public-facing web site and will provide [Client] with architectural recommendations based on the current design being done by their design partner [Partner] and the internal Content Team. In addition, it will provide [Client] with a development plan and confidence they need to deploy a world class public Internet website. Goals 1.  [Consultant] will provide technical guidance and set project team expectations for the implementation of the MOSS Internet site based on provided features/functions within three weeks. 2.  [Consultant] will understand phase 2 secure password-protected Internet site design and provide recommendations.   Deliverables 1.1  Public Internet (unsecure) Architectural Recommendation Plan 1.2  Physical Site construction Work Breakdown Structure and plan (Time, cost and resources needed) 2.1  Two Factor authentication recommendation document   Objective [Client] is currently using an application developed by [Consultant] many years ago called "XXX". This application, although functional, does not meet their new updated business requirements and contains a few defects which [Client] has developed work-around processes. [Client] would like to have a "new and improved" system to support their membership management needs by expanding membership and subscription capabilities, provide accounting integration with internal (GL) and external (VeriSign) systems, and implement hooks to the current CRM solution. This effort will take place through a series of phases, beginning with envisioning. Goals 1. Through discussions with users, [Consultant] will discover current issues/bugs which need to be resolved which must meet the current functionality requirements within three weeks. 2. [Consultant] will gather requirements from the users about what is "needed" vs. "what they have" for enhancements and provide a high level document supporting their needs. 3. [Consultant] will meet with the team members through a series of meetings and help define the overall project plan to deliver a new and improved solution. Deliverables 1.1 Prioritized list of Current application issues/bugs that need to be resolved 1.2 Provide a resolution plan on the issues/bugs identified in the current application 1.3 Risk Assessment Document 2.1 Deliver a Requirements Document showing high-level [Client] needs for the new XXX application. · New feature functionality not in the application today · Existing functionality that will remain in the new functionality 2.2 Reporting Requirements Document 3.1 A Project Plan showing the deliverables and cost for the next (second) phase of this project. 3.2 A Statement of Work for the next (second) phase of this project. 3.3 An Estimate of any work that would need to follow the second phase.

    Read the article

  • tile_static, tile_barrier, and tiled matrix multiplication with C++ AMP

    - by Daniel Moth
    We ended the previous post with a mechanical transformation of the C++ AMP matrix multiplication example to the tiled model and in the process introduced tiled_index and tiled_grid. This is part 2. tile_static memory You all know that in regular CPU code, static variables have the same value regardless of which thread accesses the static variable. This is in contrast with non-static local variables, where each thread has its own copy. Back to C++ AMP, the same rules apply and each thread has its own value for local variables in your lambda, whereas all threads see the same global memory, which is the data they have access to via the array and array_view. In addition, on an accelerator like the GPU, there is a programmable cache, a third kind of memory type if you'd like to think of it that way (some call it shared memory, others call it scratchpad memory). Variables stored in that memory share the same value for every thread in the same tile. So, when you use the tiled model, you can have variables where each thread in the same tile sees the same value for that variable, that threads from other tiles do not. The new storage class for local variables introduced for this purpose is called tile_static. You can only use tile_static in restrict(direct3d) functions, and only when explicitly using the tiled model. What this looks like in code should be no surprise, but here is a snippet to confirm your mental image, using a good old regular C array // each tile of threads has its own copy of locA, // shared among the threads of the tile tile_static float locA[16][16]; Note that tile_static variables are scoped and have the lifetime of the tile, and they cannot have constructors or destructors. tile_barrier In amp.h one of the types introduced is tile_barrier. You cannot construct this object yourself (although if you had one, you could use a copy constructor to create another one). So how do you get one of these? You get it, from a tiled_index object. Beyond the 4 properties returning index objects, tiled_index has another property, barrier, that returns a tile_barrier object. The tile_barrier class exposes a single member, the method wait. 15: // Given a tiled_index object named t_idx 16: t_idx.barrier.wait(); 17: // more code …in the code above, all threads in the tile will reach line 16 before a single one progresses to line 17. Note that all threads must be able to reach the barrier, i.e. if you had branchy code in such a way which meant that there is a chance that not all threads could reach line 16, then the code above would be illegal. Tiled Matrix Multiplication Example – part 2 So now that we added to our understanding the concepts of tile_static and tile_barrier, let me obfuscate rewrite the matrix multiplication code so that it takes advantage of tiling. Before you start reading this, I suggest you get a cup of your favorite non-alcoholic beverage to enjoy while you try to fully understand the code. 01: void MatrixMultiplyTiled(vector<float>& vC, const vector<float>& vA, const vector<float>& vB, int M, int N, int W) 02: { 03: static const int TS = 16; 04: array_view<const float,2> a(M, W, vA); 05: array_view<const float,2> b(W, N, vB); 06: array_view<writeonly<float>,2> c(M,N,vC); 07: parallel_for_each(c.grid.tile< TS, TS >(), 08: [=] (tiled_index< TS, TS> t_idx) restrict(direct3d) 09: { 10: int row = t_idx.local[0]; int col = t_idx.local[1]; 11: float sum = 0.0f; 12: for (int i = 0; i < W; i += TS) { 13: tile_static float locA[TS][TS], locB[TS][TS]; 14: locA[row][col] = a(t_idx.global[0], col + i); 15: locB[row][col] = b(row + i, t_idx.global[1]); 16: t_idx.barrier.wait(); 17: for (int k = 0; k < TS; k++) 18: sum += locA[row][k] * locB[k][col]; 19: t_idx.barrier.wait(); 20: } 21: c[t_idx.global] = sum; 22: }); 23: } Notice that all the code up to line 9 is the same as per the changes we made in part 1 of tiling introduction. If you squint, the body of the lambda itself preserves the original algorithm on lines 10, 11, and 17, 18, and 21. The difference being that those lines use new indexing and the tile_static arrays; the tile_static arrays are declared and initialized on the brand new lines 13-15. On those lines we copy from the global memory represented by the array_view objects (a and b), to the tile_static vanilla arrays (locA and locB) – we are copying enough to fit a tile. Because in the code that follows on line 18 we expect the data for this tile to be in the tile_static storage, we need to synchronize the threads within each tile with a barrier, which we do on line 16 (to avoid accessing uninitialized memory on line 18). We also need to synchronize the threads within a tile on line 19, again to avoid the race between lines 14, 15 (retrieving the next set of data for each tile and overwriting the previous set) and line 18 (not being done processing the previous set of data). Luckily, as part of the awesome C++ AMP debugger in Visual Studio there is an option that helps you find such races, but that is a story for another blog post another time. May I suggest reading the next section, and then coming back to re-read and walk through this code with pen and paper to really grok what is going on, if you haven't already? Cool. Why would I introduce this tiling complexity into my code? Funny you should ask that, I was just about to tell you. There is only one reason we tiled our extent, had to deal with finding a good tile size, ensure the number of threads we schedule are correctly divisible with the tile size, had to use a tiled_index instead of a normal index, and had to understand tile_barrier and to figure out where we need to use it, and double the size of our lambda in terms of lines of code: the reason is to be able to use tile_static memory. Why do we want to use tile_static memory? Because accessing tile_static memory is around 10 times faster than accessing the global memory on an accelerator like the GPU, e.g. in the code above, if you can get 150GB/second accessing data from the array_view a, you can get 1500GB/second accessing the tile_static array locA. And since by definition you are dealing with really large data sets, the savings really pay off. We have seen tiled implementations being twice as fast as their non-tiled counterparts. Now, some algorithms will not have performance benefits from tiling (and in fact may deteriorate), e.g. algorithms that require you to go only once to global memory will not benefit from tiling, since with tiling you already have to fetch the data once from global memory! Other algorithms may benefit, but you may decide that you are happy with your code being 150 times faster than the serial-version you had, and you do not need to invest to make it 250 times faster. Also algorithms with more than 3 dimensions, which C++ AMP supports in the non-tiled model, cannot be tiled. Also note that in future releases, we may invest in making the non-tiled model, which already uses tiling under the covers, go the extra step and use tile_static memory on your behalf, but it is obviously way to early to commit to anything like that, and we certainly don't do any of that today. Comments about this post by Daniel Moth welcome at the original blog.

    Read the article

  • Oracle Database 12 c New Partition Maintenance Features by Gwen Lazenby

    - by hamsun
    One of my favourite new features in Oracle Database 12c is the ability to perform partition maintenance operations on multiple partitions. This means we can now add, drop, truncate and merge multiple partitions in one operation, and can split a single partition into more than two partitions also in just one command. This would certainly have made my life slightly easier had it been available when I administered a data warehouse at Oracle 9i. To demonstrate this new functionality and syntax, I am going to create two tables, ORDERS and ORDERS_ITEMS which have a parent-child relationship. ORDERS is to be partitioned using range partitioning on the ORDER_DATE column, and ORDER_ITEMS is going to partitioned using reference partitioning and its foreign key relationship with the ORDERS table. This form of partitioning was a new feature in 11g and means that any partition maintenance operations performed on the ORDERS table will also take place on the ORDER_ITEMS table as well. First create the ORDERS table - SQL CREATE TABLE orders ( order_id NUMBER(12), order_date TIMESTAMP, order_mode VARCHAR2(8), customer_id NUMBER(6), order_status NUMBER(2), order_total NUMBER(8,2), sales_rep_id NUMBER(6), promotion_id NUMBER(6), CONSTRAINT orders_pk PRIMARY KEY(order_id) ) PARTITION BY RANGE(order_date) (PARTITION Q1_2007 VALUES LESS THAN (TO_DATE('01-APR-2007','DD-MON-YYYY')), PARTITION Q2_2007 VALUES LESS THAN (TO_DATE('01-JUL-2007','DD-MON-YYYY')), PARTITION Q3_2007 VALUES LESS THAN (TO_DATE('01-OCT-2007','DD-MON-YYYY')), PARTITION Q4_2007 VALUES LESS THAN (TO_DATE('01-JAN-2008','DD-MON-YYYY')) ); Table created. Now the ORDER_ITEMS table SQL CREATE TABLE order_items ( order_id NUMBER(12) NOT NULL, line_item_id NUMBER(3) NOT NULL, product_id NUMBER(6) NOT NULL, unit_price NUMBER(8,2), quantity NUMBER(8), CONSTRAINT order_items_fk FOREIGN KEY(order_id) REFERENCES orders(order_id) on delete cascade) PARTITION BY REFERENCE(order_items_fk) tablespace example; Table created. Now look at DBA_TAB_PARTITIONS to get details of what partitions we have in the two tables – SQL select table_name,partition_name, partition_position position, high_value from dba_tab_partitions where table_owner='SH' and table_name like 'ORDER_%' order by partition_position, table_name; TABLE_NAME PARTITION_NAME POSITION HIGH_VALUE -------------- --------------- -------- ------------------------- ORDERS Q1_2007 1 TIMESTAMP' 2007-04-01 00:00:00' ORDER_ITEMS Q1_2007 1 ORDERS Q2_2007 2 TIMESTAMP' 2007-07-01 00:00:00' ORDER_ITEMS Q2_2007 2 ORDERS Q3_2007 3 TIMESTAMP' 2007-10-01 00:00:00' ORDER_ITEMS Q3_2007 3 ORDERS Q4_2007 4 TIMESTAMP' 2008-01-01 00:00:00' ORDER_ITEMS Q4_2007 4 Just as an aside it is also now possible in 12c to use interval partitioning on reference partitioned tables. In 11g it was not possible to combine these two new partitioning features. For our first example of the new 12cfunctionality, let us add all the partitions necessary for 2008 to the tables using one command. Notice that the partition specification part of the add command is identical in format to the partition specification part of the create command as shown above - SQL alter table orders add PARTITION Q1_2008 VALUES LESS THAN (TO_DATE('01-APR-2008','DD-MON-YYYY')), PARTITION Q2_2008 VALUES LESS THAN (TO_DATE('01-JUL-2008','DD-MON-YYYY')), PARTITION Q3_2008 VALUES LESS THAN (TO_DATE('01-OCT-2008','DD-MON-YYYY')), PARTITION Q4_2008 VALUES LESS THAN (TO_DATE('01-JAN-2009','DD-MON-YYYY')); Table altered. Now look at DBA_TAB_PARTITIONS and we can see that the 4 new partitions have been added to both tables – SQL select table_name,partition_name, partition_position position, high_value from dba_tab_partitions where table_owner='SH' and table_name like 'ORDER_%' order by partition_position, table_name; TABLE_NAME PARTITION_NAME POSITION HIGH_VALUE -------------- --------------- -------- ------------------------- ORDERS Q1_2007 1 TIMESTAMP' 2007-04-01 00:00:00' ORDER_ITEMS Q1_2007 1 ORDERS Q2_2007 2 TIMESTAMP' 2007-07-01 00:00:00' ORDER_ITEMS Q2_2007 2 ORDERS Q3_2007 3 TIMESTAMP' 2007-10-01 00:00:00' ORDER_ITEMS Q3_2007 3 ORDERS Q4_2007 4 TIMESTAMP' 2008-01-01 00:00:00' ORDER_ITEMS Q4_2007 4 ORDERS Q1_2008 5 TIMESTAMP' 2008-04-01 00:00:00' ORDER_ITEMS Q1_2008 5 ORDERS Q2_2008 6 TIMESTAMP' 2008-07-01 00:00:00' ORDER_ITEM Q2_2008 6 ORDERS Q3_2008 7 TIMESTAMP' 2008-10-01 00:00:00' ORDER_ITEMS Q3_2008 7 ORDERS Q4_2008 8 TIMESTAMP' 2009-01-01 00:00:00' ORDER_ITEMS Q4_2008 8 Next, we can drop or truncate multiple partitions by giving a comma separated list in the alter table command. Note the use of the plural ‘partitions’ in the command as opposed to the singular ‘partition’ prior to 12c– SQL alter table orders drop partitions Q3_2008,Q2_2008,Q1_2008; Table altered. Now look at DBA_TAB_PARTITIONS and we can see that the 3 partitions have been dropped in both the two tables – TABLE_NAME PARTITION_NAME POSITION HIGH_VALUE -------------- --------------- -------- ------------------------- ORDERS Q1_2007 1 TIMESTAMP' 2007-04-01 00:00:00' ORDER_ITEMS Q1_2007 1 ORDERS Q2_2007 2 TIMESTAMP' 2007-07-01 00:00:00' ORDER_ITEMS Q2_2007 2 ORDERS Q3_2007 3 TIMESTAMP' 2007-10-01 00:00:00' ORDER_ITEMS Q3_2007 3 ORDERS Q4_2007 4 TIMESTAMP' 2008-01-01 00:00:00' ORDER_ITEMS Q4_2007 4 ORDERS Q4_2008 5 TIMESTAMP' 2009-01-01 00:00:00' ORDER_ITEMS Q4_2008 5 Now let us merge all the 2007 partitions together to form one single partition – SQL alter table orders merge partitions Q1_2005, Q2_2005, Q3_2005, Q4_2005 into partition Y_2007; Table altered. TABLE_NAME PARTITION_NAME POSITION HIGH_VALUE -------------- --------------- -------- ------------------------- ORDERS Y_2007 1 TIMESTAMP' 2008-01-01 00:00:00' ORDER_ITEMS Y_2007 1 ORDERS Q4_2008 2 TIMESTAMP' 2009-01-01 00:00:00' ORDER_ITEMS Q4_2008 2 Splitting partitions is a slightly more involved. In the case of range partitioning one of the new partitions must have no high value defined, and in list partitioning one of the new partitions must have no list of values defined. I call these partitions the ‘everything else’ partitions, and will contain any rows contained in the original partition that are not contained in the any of the other new partitions. For example, let us split the Y_2007 partition back into 4 quarterly partitions – SQL alter table orders split partition Y_2007 into (PARTITION Q1_2007 VALUES LESS THAN (TO_DATE('01-APR-2007','DD-MON-YYYY')), PARTITION Q2_2007 VALUES LESS THAN (TO_DATE('01-JUL-2007','DD-MON-YYYY')), PARTITION Q3_2007 VALUES LESS THAN (TO_DATE('01-OCT-2007','DD-MON-YYYY')), PARTITION Q4_2007); Now look at DBA_TAB_PARTITIONS to get details of the new partitions – TABLE_NAME PARTITION_NAME POSITION HIGH_VALUE -------------- --------------- -------- ------------------------- ORDERS Q1_2007 1 TIMESTAMP' 2007-04-01 00:00:00' ORDER_ITEMS Q1_2007 1 ORDERS Q2_2007 2 TIMESTAMP' 2007-07-01 00:00:00' ORDER_ITEMS Q2_2007 2 ORDERS Q3_2007 3 TIMESTAMP' 2007-10-01 00:00:00' ORDER_ITEMS Q3_2007 3 ORDERS Q4_2007 4 TIMESTAMP' 2008-01-01 00:00:00' ORDER_ITEMS Q4_2007 4 ORDERS Q4_2008 5 TIMESTAMP' 2009-01-01 00:00:00' ORDER_ITEMS Q4_2008 5 Partition Q4_2007 has a high value equal to the high value of the original Y_2007 partition, and so has inherited its upper boundary from the partition that was split. As for a list partitioning example let look at the following another table, SALES_PAR_LIST, which has 2 partitions, Americas and Europe and a partitioning key of country_name. SQL select table_name,partition_name, high_value from dba_tab_partitions where table_owner='SH' and table_name = 'SALES_PAR_LIST'; TABLE_NAME PARTITION_NAME HIGH_VALUE -------------- --------------- ----------------------------- SALES_PAR_LIST AMERICAS 'Argentina', 'Canada', 'Peru', 'USA', 'Honduras', 'Brazil', 'Nicaragua' SALES_PAR_LIST EUROPE 'France', 'Spain', 'Ireland', 'Germany', 'Belgium', 'Portugal', 'Denmark' Now split the Americas partition into 3 partitions – SQL alter table sales_par_list split partition americas into (partition south_america values ('Argentina','Peru','Brazil'), partition north_america values('Canada','USA'), partition central_america); Table altered. Note that no list of values was given for the ‘Central America’ partition. However it should have inherited any values in the original ‘Americas’ partition that were not assigned to either the ‘North America’ or ‘South America’ partitions. We can confirm this by looking at the DBA_TAB_PARTITIONS view. SQL select table_name,partition_name, high_value from dba_tab_partitions where table_owner='SH' and table_name = 'SALES_PAR_LIST'; TABLE_NAME PARTITION_NAME HIGH_VALUE --------------- --------------- -------------------------------- SALES_PAR_LIST SOUTH_AMERICA 'Argentina', 'Peru', 'Brazil' SALES_PAR_LIST NORTH_AMERICA 'Canada', 'USA' SALES_PAR_LIST CENTRAL_AMERICA 'Honduras', 'Nicaragua' SALES_PAR_LIST EUROPE 'France', 'Spain', 'Ireland', 'Germany', 'Belgium', 'Portugal', 'Denmark' In conclusion, I hope that DBA’s whose work involves maintaining partitions will find the operations a bit more straight forward to carry out once they have upgraded to Oracle Database 12c. Gwen Lazenby is a Principal Training Consultant at Oracle. She is part of Oracle University's Core Technology delivery team based in the UK, teaching Database Administration and Linux courses. Her specialist topics include using Oracle Partitioning and Parallelism in Data Warehouse environments, as well as Oracle Spatial and RMAN.

    Read the article

  • Guide to MySQL & NoSQL, Webinar Q&A

    - by Mat Keep
    0 0 1 959 5469 Homework 45 12 6416 14.0 Normal 0 false false false EN-US JA X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:12.0pt; font-family:Cambria; mso-ascii-font-family:Cambria; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Cambria; mso-hansi-theme-font:minor-latin; mso-ansi-language:EN-US;} Yesterday we ran a webinar discussing the demands of next generation web services and how blending the best of relational and NoSQL technologies enables developers and architects to deliver the agility, performance and availability needed to be successful. Attendees posted a number of great questions to the MySQL developers, serving to provide additional insights into areas like auto-sharding and cross-shard JOINs, replication, performance, client libraries, etc. So I thought it would be useful to post those below, for the benefit of those unable to attend the webinar. Before getting to the Q&A, there are a couple of other resources that maybe useful to those looking at NoSQL capabilities within MySQL: - On-Demand webinar (coming soon!) - Slides used during the webinar - Guide to MySQL and NoSQL whitepaper  - MySQL Cluster demo, including NoSQL interfaces, auto-sharing, high availability, etc.  So here is the Q&A from the event  Q. Where does MySQL Cluster fit in to the CAP theorem? A. MySQL Cluster is flexible. A single Cluster will prefer consistency over availability in the presence of network partitions. A pair of Clusters can be configured to prefer availability over consistency. A full explanation can be found on the MySQL Cluster & CAP Theorem blog post.  Q. Can you configure the number of replicas? (the slide used a replication factor of 1) Yes. A cluster is configured by an .ini file. The option NoOfReplicas sets the number of originals and replicas: 1 = no data redundancy, 2 = one copy etc. Usually there's no benefit in setting it >2. Q. Interestingly most (if not all) of the NoSQL databases recommend having 3 copies of data (the replication factor).    Yes, with configurable quorum based Reads and writes. MySQL Cluster does not need a quorum of replicas online to provide service. Systems that require a quorum need > 2 replicas to be able to tolerate a single failure. Additionally, many NoSQL systems take liberal inspiration from the original GFS paper which described a 3 replica configuration. MySQL Cluster avoids the need for a quorum by using a lightweight arbitrator. You can configure more than 2 replicas, but this is a tradeoff between incrementally improved availability, and linearly increased cost. Q. Can you have cross node group JOINS? Wouldn't that run into the risk of flooding the network? MySQL Cluster 7.2 supports cross nodegroup joins. A full cross-join can require a large amount of data transfer, which may bottleneck on network bandwidth. However, for more selective joins, typically seen with OLTP and light analytic applications, cross node-group joins give a great performance boost and network bandwidth saving over having the MySQL Server perform the join. Q. Are the details of the benchmark available anywhere? According to my calculations it results in approx. 350k ops/sec per processor which is the largest number I've seen lately The details are linked from Mikael Ronstrom's blog The benchmark uses a benchmarking tool we call flexAsynch which runs parallel asynchronous transactions. It involved 100 byte reads, of 25 columns each. Regarding the per-processor ops/s, MySQL Cluster is particularly efficient in terms of throughput/node. It uses lock-free minimal copy message passing internally, and maximizes ID cache reuse. Note also that these are in-memory tables, there is no need to read anything from disk. Q. Is access control (like table) planned to be supported for NoSQL access mode? Currently we have not seen much need for full SQL-like access control (which has always been overkill for web apps and telco apps). So we have no plans, though especially with memcached it is certainly possible to turn-on connection-level access control. But specifically table level controls are not planned. Q. How is the performance of memcached APi with MySQL against memcached+MySQL or any other Object Cache like Ecache with MySQL DB? With the memcache API we generally see a memcached response in less than 1 ms. and a small cluster with one memcached server can handle tens of thousands of operations per second. Q. Can .NET can access MemcachedAPI? Yes, just use a .Net memcache client such as the enyim or BeIT memcache libraries. Q. Is the row level locking applicable when you update a column through memcached API? An update that comes through memcached uses a row lock and then releases it immediately. Memcached operations like "INCREMENT" are actually pushed down to the data nodes. In most cases the locks are not even held long enough for a network round trip. Q. Has anyone published an example using something like PHP? I am assuming that you just use the PHP memcached extension to hook into the memcached API. Is that correct? Not that I'm aware of but absolutely you can use it with php or any of the other drivers Q. For beginner we need more examples. Take a look here for a fully worked example Q. Can I access MySQL using Cobol (Open Cobol) or C and if so where can I find the coding libraries etc? A. There is a cobol implementation that works well with MySQL, but I do not think it is Open Cobol. Also there is a MySQL C client library that is a standard part of every mysql distribution Q. Is there a place to go to find help when testing and/implementing the NoSQL access? If using Cluster then you can use the [email protected] alias or post on the MySQL Cluster forum Q. Are there any white papers on this?  Yes - there is more detail in the MySQL Guide to NoSQL whitepaper If you have further questions, please don’t hesitate to use the comments below!

    Read the article

  • SQL University: What and why of database refactoring

    - by Mladen Prajdic
    This is a post for a great idea called SQL University started by Jorge Segarra also famously known as SqlChicken on Twitter. It’s a collection of blog posts on different database related topics contributed by several smart people all over the world. So this week is mine and we’ll be talking about database testing and refactoring. In 3 posts we’ll cover: SQLU part 1 - What and why of database testing SQLU part 2 - What and why of database refactoring SQLU part 3 - Tools of the trade This is a second part of the series and in it we’ll take a look at what database refactoring is and why do it. Why refactor a database To know why refactor we first have to know what refactoring actually is. Code refactoring is a process where we change module internals in a way that does not change that module’s input/output behavior. For successful refactoring there is one crucial thing we absolutely must have: Tests. Automated unit tests are the only guarantee we have that we haven’t broken the input/output behavior before refactoring. If you haven’t go back ad read my post on the matter. Then start writing them. Next thing you need is a code module. Those are views, UDFs and stored procedures. By having direct table access we can kiss fast and sweet refactoring good bye. One more point to have a database abstraction layer. And no, ORM’s don’t fall into that category. But also know that refactoring is NOT adding new functionality to your code. Many have fallen into this trap. Don’t be one of them and resist the lure of the dark side. And it’s a strong lure. We developers in general love to add new stuff to our code, but hate fixing our own mistakes or changing existing code for no apparent reason. To be a good refactorer one needs discipline and focus. Now we know that refactoring is all about changing inner workings of existing code. This can be due to performance optimizations, changing internal code workflows or some other reason. This is a typical black box scenario to the outside world. If we upgrade the car engine it still has to drive on the road (preferably faster) and not fly (no matter how cool that would be). Also be aware that white box tests will break when we refactor. What to refactor in a database Refactoring databases doesn’t happen that often but when it does it can include a lot of stuff. Let us look at a few common cases. Adding or removing database schema objects Adding, removing or changing table columns in any way, adding constraints, keys, etc… All of these can be counted as internal changes not visible to the data consumer. But each of these carries a potential input/output behavior change. Dropping a column can result in views not working anymore or stored procedure logic crashing. Adding a unique constraint shows duplicated data that shouldn’t exist. Foreign keys break a truncate table command executed from an application that runs once a month. All these scenarios are very real and can happen. With the proper database abstraction layer fully covered with black box tests we can make sure something like that does not happen (hopefully at all). Changing physical structures Physical structures include heaps, indexes and partitions. We can pretty much add or remove those without changing the data returned by the database. But the performance can be affected. So here we use our performance tests. We do have them, right? Just by adding a single index we can achieve orders of magnitude performance improvement. Won’t that make users happy? But what if that index causes our write operations to crawl to a stop. again we have to test this. There are a lot of things to think about and have tests for. Without tests we can’t do successful refactoring! Fixing bad code We all have some bad code in our systems. We usually refer to that code as code smell as they violate good coding practices. Examples of such code smells are SQL injection, use of SELECT *, scalar UDFs or cursors, etc… Each of those is huge code smell and can result in major code changes. Take SELECT * from example. If we remove a column from a table the client using that SELECT * statement won’t have a clue about that until it runs. Then it will gracefully crash and burn. Not to mention the widely unknown SELECT * view refresh problem that Tomas LaRock (@SQLRockstar on Twitter) and Colin Stasiuk (@BenchmarkIT on Twitter) talk about in detail. Go read about it, it’s informative. Refactoring this includes replacing the * with column names and most likely change to application using the database. Breaking apart huge stored procedures Have you ever seen seen a stored procedure that was 2000 lines long? I have. It’s not pretty. It hurts the eyes and sucks the will to live the next 10 minutes. They are a maintenance nightmare and turn into things no one dares to touch. I’m willing to bet that 100% of time they don’t have a single test on them. Large stored procedures (and functions) are a clear sign that they contain business logic. General opinion on good database coding practices says that business logic has no business in the database. That’s the applications part. Refactoring such behemoths requires writing lots of edge case tests for the stored procedure input/output behavior and then start to refactor it. First we split the logic inside into smaller parts like new stored procedures and UDFs. Those then get called from the master stored procedure. Once we’ve successfully modularized the database code it’s best to transfer that logic into the applications consuming it. This only leaves the stored procedure with common data manipulation logic. Of course this isn’t always possible so having a plethora of performance and behavior unit tests is absolutely necessary to confirm we’ve actually improved the codebase in some way.   Refactoring is not a popular chore amongst developers or managers. The former don’t like fixing old code, the latter can’t see the financial benefit. Remember how we talked about being lousy at estimating future costs in the previous post? But there comes a time when it must be done. Hopefully I’ve given you some ideas how to get started. In the last post of the series we’ll take a look at the tools to use and an example of testing and refactoring.

    Read the article

  • How-to delete a tree node using the context menu

    - by frank.nimphius
    Hierarchical trees in Oracle ADF make use of View Accessors, which means that only the top level node needs to be exposed as a View Object instance on the ADF Business Components Data Model. This also means that only the top level node has a representation in the PageDef file as a tree binding and iterator binding reference. Detail nodes are accessed through tree rule definitions that use the accessor mentioned above (or nested collections in the case of POJO or EJB business services). The tree component is configured for single node selection, which however can be declaratively changed for users to press the ctrl key and selecting multiple nodes. In the following, I explain how to create a context menu on the tree for users to delete the selected tree nodes. For this, the context menu item will access a managed bean, which then determines the selected node(s), the internal ADF node bindings and the rows they represent. As mentioned, the ADF Business Components Data Model only needs to expose the top level node data sources, which in this example is an instance of the Locations View Object. For the tree to work, you need to have associations defined between entities, which usually is done for you by Oracle JDeveloper if the database tables have foreign keys defined Note: As a general hint of best practices and to simplify your life: Make sure your database schema is well defined and designed before starting your development project. Don't treat the database as something organic that grows and changes with the requirements as you proceed in your project. Business service refactoring in response to database changes is possible, but should be treated as an exception, not the rule. Good database design is a necessity – even for application developers – and nothing evil. To create the tree component, expand the Data Controls panel and drag the View Object collection to the view. From the context menu, select the tree component entry and continue with defining the tree rules that make up the hierarchical structure. As you see, when pressing the green plus icon  in the Edit Tree Binding  dialog, the data structure, Locations -  Departments – Employees in my sample, shows without you having created a View Object instance for each of the nodes in the ADF Business Components Data Model. After you configured the tree structure in the Edit Tree Binding dialog, you press OK and the tree is created. Select the tree in the page editor and open the Structure Window (ctrl+shift+S). In the Structure window, expand the tree node to access the conextMenu facet. Use the right mouse button to insert a Popup  into the facet. Repeat the same steps to insert a Menu and a Menu Item into the Popup you created. The Menu item text should be changed to something meaningful like "Delete". Note that the custom menu item later is added to the context menu together with the default context menu options like expand and expand all. To define the action that is executed when the menu item is clicked on, you select the Action Listener property in the Property Inspector and click the arrow icon followed by the Edit menu option. Create or select a managed bean and define a method name for the action handler. Next, select the tree component and browse to its binding property in the Property Inspector. Again, use the arrow icon | Edit option to create a component binding in the same managed bean that has the action listener defined. The tree handle is used in the action listener code, which is shown below: public void onTreeNodeDelete(ActionEvent actionEvent) {   //access the tree from the JSF component reference created   //using the af:tree "binding" property. The "binding" property   //creates a pair of set/get methods to access the RichTree instance   RichTree tree = this.getTreeHandler();   //get the list of selected row keys   RowKeySet rks = tree.getSelectedRowKeys();   //access the iterator to loop over selected nodes   Iterator rksIterator = rks.iterator();          //The CollectionModel represents the tree model and is   //accessed from the tree "value" property   CollectionModel model = (CollectionModel) tree.getValue();   //The CollectionModel is a wrapper for the ADF tree binding   //class, which is JUCtrlHierBinding   JUCtrlHierBinding treeBinding =                  (JUCtrlHierBinding) model.getWrappedData();          //loop over the selected nodes and delete the rows they   //represent   while(rksIterator.hasNext()){     List nodeKey = (List) rksIterator.next();     //find the ADF node binding using the node key     JUCtrlHierNodeBinding node =                       treeBinding.findNodeByKeyPath(nodeKey);     //delete the row.     Row rw = node.getRow();       rw.remove();   }          //only refresh the tree if tree nodes have been selected   if(rks.size() > 0){     AdfFacesContext adfFacesContext =                          AdfFacesContext.getCurrentInstance();     adfFacesContext.addPartialTarget(tree);   } } Note: To enable multi node selection for a tree, select the tree and change the row selection setting from "single" to "multiple". Note: a fully pictured version of this post will become available at the end of the month in a PDF summary on ADF Code Corner : http://www.oracle.com/technetwork/developer-tools/adf/learnmore/index-101235.html 

    Read the article

  • Instructions on how to configure a WebLogic Cluster and use it with Oracle Http Server

    - by Laurent Goldsztejn
    On October 17th I delivered a webcast on WebLogic Clustering that included a demo with Apache as the proxy server.  I realized that many steps are needed to set up the configuration I used during the demo.  The purpose of this article is to go through these steps to show how quickly and easily one can define a new cluster and then proxy requests via an Oracle Http Server (OHS). The domain configuration wizard offers the option to create a cluster.  The administration console or WLST, the Weblogic scripting tool can also be used to define a new cluster.  It can be created at any time but the servers that will participate in it cannot be in a running state. Cluster Creation using the configuration wizard Network and architecture requirements need to be considered while choosing between unicast and multicast. Multicast Vs. Unicast with WebLogic Clustering is of great help to make the best decision between the two messaging modes.  In addition, Configure Cluster offers details on each single field displayed above. After this initial configuration page, individual servers could be assigned to this newly created cluster although servers can be added later to the cluster.  What is not recommended is for the Admin server to participate in a cluster as the main purpose of the Admin server is to perform the bulk of the processing for the domain.  Servers need to stop before being assigned to a cluster.  There is also no minimum number of servers that have to participate in the cluster. At this point the configuration should be done and the cluster created successfully.  This can easily be verified from the console. Each clustered managed server can be launched to join the cluster.   At startup the following messages should be logged for each clustered managed server: <Notice> <WeblogicServer> <BEA-000365> <Server state changed to STARTING> <Notice> <Cluster> <BEA-000197> <Listening for announcements from cluster using messaging_mode cluster messaging> <Notice> <Cluster> <BEA-000133> <Waiting to synchronize with other running members of cluster_name>  It's time to try sending requests to the cluster and we will do this with the help of Oracle Http Server to play the role of a proxy server to demonstrate load balancing.  Proxy Server configuration  The first step is to download Weblogic Server Web Server Plugin that will enhance the web server by handling requests aimed at being sent to the Weblogic cluster.  For our test Oracle Http Server (OHS) will be used.  However plug-ins are also available for Apache Http server, Microsoft Internet Information Server (IIS), Oracle iPlanet Webserver or even WebLogic Server with the HttpClusterServlet. Once OHS is installed on the system, the configuration file, mod_wl_ohs.conf, will need to be altered to include Weblogic proxy specifics. First of all, add the following directive to instruct Apache to load the Weblogic shared object module extracted from the plugins file just downloaded. LoadModule weblogic_module modules/mod_wl_ohs.so and then create an IfModule directive to encapsulate the following location block so that proxy will be enabled by path (each request including /wls will be directed directly to the WebLogic Cluster).  You could also proxy requests by MIME type using MatchExpression in the Location block. <IfModule weblogic_module> <Location /wls>    SetHandler weblogic-handler    PathTrim /wls    WebLogicCluster MS1_URL:port,MS2_URL:port    Debug ON    WLLogFile        c:/tmp/global_proxy.log     WLTempDir        "c:/myTemp"    DebugConfigInfo  On </Location> </IfModule> SetHandler specifies the handler for the plug-in module  PathTrim will instruct the plug-in to trim /w ls from the URL before forwarding the request to the cluster. The list of WebLogic Servers defined in WeblogicCluster could contain a mixed set of clustered and single servers.  However, the dynamic list returned for this parameter will only contain valid clustered servers and may contain more servers if not all clustered servers are listed in WeblogicCluster. Testing proxy and load balancing It's time to start OHS web server which should at this point be configured correctly to proxy requests to the clustered servers.  By default round-robin is the load balancing strategy set by WebLogic. Testing the load balancing can be easily done by disabling cookies on your browser given that a request containing a cookie attempts to connect to the primary server. If that attempt fails, the plug-in attempts to make a connection to the next available server in the list in a round-robin fashion.  With cookies enabled, you could use two different browsers to test the load balancing with a JSP page that contains the following: <%@ page contentType="text/html; charset=iso-8859-1" language="java"  %>  <%  String path = request.getContextPath();   String getProtocol=request.getScheme();   String getDomain=request.getServerName();   String getPort=Integer.toString(request.getLocalPort());   String getPath = getProtocol+"://"+getDomain+":"+getPort+path+"/"; %> <html> <body> Receiving Server <%=getPath%> </body> </html>  Assuming that you name the JSP page Test.jsp and the webapp that contains it TestApp, your browsers should open the following URL: http://localhost/wls/TestApp/Test.jsp  Each browser should connect to a different clustered server and this simple JSP should confirm that.  The webapp that contains the JSP needs to be deployed to the cluster. You can also verify that the load is correctly balanced by looking at the proxy log file.  Each request generates a set of log entries that starts with : timestamp ================New Request: Each request is associated with a primary server and a secondary server if one is available.  For our test request, the following entries should appear in the log as well:Using Uri /wls/TestApp/Test.jsp After trimming path: '/TestApp/Test.jsp' The final request string is '/TestApp/Test.jsp' If an exception occurs, it should also be logged in the proxy log file with the prefix:timestamp *******Exception type   WeblogicBridgeConfig DebugConfigInfo enables runtime statistics and the production of configuration information.  For security purposes, this parameter should be turned off in production. http://webserver_host:port/path/xyz.jsp?__WebLogicBridgeConfig will display a proxy bridge page detailing the plugin configuration followed by runtime statistics which could help in diagnosing issues along with the analyzing of the proxy log file.  In our example the url would be: http://localhost/wls/TestApp/Test.jsp?__WebLogicBridgeConfig  Here is how the top section of the screen can look like: The bottom part of the page contains runtime statistics, here is a snippet of it (unrelated with the previous JSP example).   This entire plugin configuration should be very similar with other web servers, what varies is the name of the proxy server configuration file. So, as you can see, it only takes a few minutes to configure a Weblogic cluster and get servers to join it. 

    Read the article

  • 10 tape technology features that make you go hmm.

    - by Karoly Vegh
    A week ago an Oracle/StorageTek Tape Specialist, Christian Vanden Balck, visited Vienna, and agreed to visit customers to do techtalks and update them about the technology boom going around tape. I had the privilege to attend some of his sessions and noted the information and features that took the customers by surprise and made them think. Allow me to share the top 10: I. StorageTek as a brand: StorageTek is one of he strongest names in the Tape field. The brand itself was valued so much by customers that even after Sun Microsystems acquiring StorageTek and the Oracle acquiring Sun the brand lives on with all the Oracle tapelibraries are officially branded StorageTek.See http://www.oracle.com/us/products/servers-storage/storage/tape-storage/overview/index.html II. Disk information density limitations: Disk technology struggles with information density. You haven't seen the disk sizes exploding lately, have you? That's partly because there are physical limits on a disk platter. The size is given, the number of platters is limited, they just can't grow, and are running out of physical area to write to. Now, in a T10000C tape cartridge we have over 1000m long tape. There you go, you have got your physical space and don't need to stuff all that data crammed together. You can write in a reliable pattern, and have space to grow too. III. Oracle has a market share of 62% worldwide in recording head manufacturing. That's right. If you are running LTO drives, with a good chance you rely on StorageTek production. That's two out of three LTO recording heads produced worldwide.  IV. You can store 1 Exabyte data in a single tape library. Yes, an Exabyte. That is 1000 Petabytes. Or, a million Terabytes. A thousand million GigaBytes. You can store that in a stacked StorageTek SL8500 tapelibrary. In one SL8500 you can put 10.000 T10000C cartridges, that store 10TB data (compressed). You can stack 10 of these SL8500s together. Boom. 1000.000 TB.(n.b.: stacking means interconnecting the libraries. Yes, cartridges are moved between the stacked libraries automatically.)  V. EMC: 'Tape doesn't suck after all. We moved on.': Do you remember the infamous 'Tape sucks, move on' Datadomain slogan? Of course they had to put it that way, having only had disk products. But here's a fun fact: on the EMCWorld 2012 there was a major presence of a Tape-tech company - EMC, in a sudden burst of sanity is embracing tape again. VI. The miraculous T10000C: Oracle StorageTek has developed an enterprise-grade tapedrive and cartridge, the T10000C. With awesome numbers: The Cartridge: Native 5TB capacity, 10TB with compression Over a kilometer long tape within the cartridge. And it's locked when unmounted, no rattling of your data.  Replaced the metalparticles datalayer with BaFe (bariumferrite) - metalparticles lose around 7% of magnetism within 30 days. BaFe does not. Yes we employ solid-state physicists doing R&D on demagnetisation in our labs. Can be partitioned, storage tiering within the cartridge!  The Drive: 2GB Cache Encryption implemented in HW - no performance hit 252 MB/s native sustained data rate, beats disk technology by far. Not to mention peak throughput.  Leading the tape while never touching the data side of it, protecting your data physically too Data integritiy checking (CRC recalculation) on tape within the drive without having to read it back to the server reordering data from tape-order, delivering it back in application-order  writing 32 tracks at once, reading them back for CRC check at once VII. You only use 20% of your data on a regular basis. The rest 80% is just lying around for years. On continuously spinning disks. Doubly consuming energy (power+cooling), blocking diskstorage capacity. There is a solution called SAM (Storage Archive Manager) that provides you a filesystem unifying disk and tape, moving data on-demand and for clients transparently between the different storage tiers. You can share these filesystems with NFS or CIFS for clients, and enjoy the low TCO of tape. Tapes don't spin. They sit quietly in their slots, storing 10TB data, using no energy, producing no heat, automounted when a client accesses their data.See: http://www.oracle.com/us/products/servers-storage/storage/storage-software/storage-archive-manager/overview/index.html VIII. HW supported for three decades: Did you know that the original PowderHorn library was released in '87 and has been only discontinued in 2010? That is over two decades of supported operation. Tape libraries are - just like the data carrying on tapecartridges - built for longevity. Oh, and the T10000C cartridge has 30-year archival life for long-term retention.  IX. Tape is easy to manage: Have you heard of Tape Storage Analytics? It is a central graphical tool to summarize, monitor, analyze dataflow, health and performance of drives and libraries, see: http://www.oracle.com/us/products/servers-storage/storage/tape-storage/tape-analytics/overview/index.html X. The next generation: The T10000B drives were able to reuse the T10000A cartridges and write on them even more data. On the same cartridges. We call this investment protection, and this is very important for Oracle for the future too. We usually support two generations of cartridges together. The current drive is a T10000C. (...I know I promised to enlist 10, but I got still two more I really want to mention. Allow me to work around the problem: ) X++. The TallBots, the robots moving around the cartridges in the StorageTek library from tapeslots to the drives are cableless. Cables, belts, chains running to moving parts in a library cause maintenance downtimes. So StorageTek eliminated them. The TallBots get power, commands, even firmwareupgrades through the rails they are running on. Also, the TallBots don't just hook'n'pull the tapes out of their slots, they actually grip'n'lift them out. No friction, no scratches, no zillion little plastic particles floating around in the library, in the drives, on your data. (X++)++: Tape beats SSDs and Disks. In terms of throughput (252 MB/s), in terms of TCO: disks cause around 290x more power and cooling, in terms of capacity: 10TB on a single media and soon more.  So... do you need to store large amounts of data? Are you legally bound to archive it for dozens of years? Would you benefit from automatic storage tiering? Have you got large mediachunks to be streamed at times? Have you got power and cooling issues in the growing datacenters? Do you find EMC's 180° turn of tape attitude interesting, but appreciate it at the same time? With all that, you aren't alone. The most data on this planet is stored on tape. Tape is coming. Big time.

    Read the article

  • External File Upload Optimizations for Windows Azure

    - by rgillen
    [Cross posted from here: http://rob.gillenfamily.net/post/External-File-Upload-Optimizations-for-Windows-Azure.aspx] I’m wrapping up a bit of the work we’ve been doing on data movement optimizations for cloud computing and the latest set of data yielded some interesting points I thought I’d share. The work done here is not really rocket science but may, in some ways, be slightly counter-intuitive and therefore seemed worthy of posting. Summary: for those who don’t like to read detailed posts or don’t have time, the synopsis is that if you are uploading data to Azure, block your data (even down to 1MB) and upload in parallel. Set your block size based on your source file size, but if you must choose a fixed value, use 1MB. Following the above will result in significant performance gains… upwards of 10x-24x and a reduction in overall file transfer time of upwards of 90% (eg, uploading a 1GB file averaged 46.37 minutes prior to optimizations and averaged 1.86 minutes afterwards). Detail: For those of you who want more detail, or think that the claims at the end of the preceding paragraph are over-reaching, what follows is information and code supporting these claims. As the title would indicate, these tests were run from our research facility pointing to the Azure cloud (specifically US North Central as it is physically closest to us) and do not represent intra-cloud results… we have performed intra-cloud tests and the overall results are similar in notion but the data rates are significantly different as well as the tipping points for the various block sizes… this will be detailed separately). We started by building a very simple console application that would loop through a directory and upload each file to Azure storage. This application used the shipping storage client library from the 1.1 version of the azure tools. The only real variation from the client library is that we added code to collect and record the duration (in ms) and size (in bytes) for each file transferred. The code is available here. We then created a directory that had a collection of files for the following sizes: 2KB, 32KB, 64KB, 128KB, 512KB, 1MB, 5MB, 10MB, 25MB, 50MB, 100MB, 250MB, 500MB, 750MB, and 1GB (50 files for each size listed). These files contained randomly-generated binary data and do not benefit from compression (a separate discussion topic). Our file generation tool is available here. The baseline was established by running the application described above against the directory containing all of the data files. This application uploads the files in a random order so as to avoid transferring all of the files of a given size sequentially and thereby spreading the affects of periodic Internet delays across the collection of results.  We then ran some scripts to split the resulting data and generate some reports. The raw data collected for our non-optimized tests is available via the links in the Related Resources section at the bottom of this post. For each file size, we calculated the average upload time (and standard deviation) and the average transfer rate (and standard deviation). As you likely are aware, transferring data across the Internet is susceptible to many transient delays which can cause anomalies in the resulting data. It is for this reason that we randomized the order of source file processing as well as executed the tests 50x for each file size. We expect that these steps will yield a sufficiently balanced set of results. Once the baseline was collected and analyzed, we updated the test harness application with some methods to split the source file into user-defined block sizes and then to upload those blocks in parallel (using the PutBlock() method of Azure storage). The parallelization was handled by simply relying on the Parallel Extensions to .NET to provide a Parallel.For loop (see linked source for specific implementation details in Program.cs, line 173 and following… less than 100 lines total). Once all of the blocks were uploaded, we called PutBlockList() to assemble/commit the file in Azure storage. For each block transferred, the MD5 was calculated and sent ensuring that the bits that arrived matched was was intended. The timer for the blocked/parallelized transfer method wraps the entire process (source file splitting, block transfer, MD5 validation, file committal). A diagram of the process is as follows: We then tested the affects of blocking & parallelizing the transfers by running the updated application against the same source set and did a parameter sweep on the block size including 256KB, 512KB, 1MB, 2MB, and 4MB (our assumption was that anything lower than 256KB wasn’t worth the trouble and 4MB is the maximum size of a block supported by Azure). The raw data for the parallel tests is available via the links in the Related Resources section at the bottom of this post. This data was processed and then compared against the single-threaded / non-optimized transfer numbers and the results were encouraging. The Excel version of the results is available here. Two semi-obvious points need to be made prior to reviewing the data. The first is that if the block size is larger than the source file size you will end up with a “negative optimization” due to the overhead of attempting to block and parallelize. The second is that as the files get smaller, the clock-time cost of blocking and parallelizing (overhead) is more apparent and can tend towards negative optimizations. For this reason (and is supported in the raw data provided in the linked worksheet) the charts and dialog below ignore source file sizes less than 1MB. (click chart for full size image) The chart above illustrates some interesting points about the results: When the block size is smaller than the source file, performance increases but as the block size approaches and then passes the source file size, you see decreasing benefit to the point of negative gains (see the values for the 1MB file size) For some of the moderately-sized source files, small blocks (256KB) are best As the size of the source file gets larger (see values for 50MB and up), the smallest block size is not the most efficient (presumably due, at least in part, to the increased number of blocks, increased number of individual transfer requests, and reassembly/committal costs). Once you pass the 250MB source file size, the difference in rate for 1MB to 4MB blocks is more-or-less constant The 1MB block size gives the best average improvement (~16x) but the optimal approach would be to vary the block size based on the size of the source file.    (click chart for full size image) The above is another view of the same data as the prior chart just with the axis changed (x-axis represents file size and plotted data shows improvement by block size). It again highlights the fact that the 1MB block size is probably the best overall size but highlights the benefits of some of the other block sizes at different source file sizes. This last chart shows the change in total duration of the file uploads based on different block sizes for the source file sizes. Nothing really new here other than this view of the data highlights the negative affects of poorly choosing a block size for smaller files.   Summary What we have found so far is that blocking your file uploads and uploading them in parallel results in significant performance improvements. Further, utilizing extension methods and the Task Parallel Library (.NET 4.0) make short work of altering the shipping client library to provide this functionality while minimizing the amount of change to existing applications that might be using the client library for other interactions.   Related Resources Source code for upload test application Source code for random file generator ODatas feed of raw data from non-optimized transfer tests Experiment Metadata Experiment Datasets 2KB Uploads 32KB Uploads 64KB Uploads 128KB Uploads 256KB Uploads 512KB Uploads 1MB Uploads 5MB Uploads 10MB Uploads 25MB Uploads 50MB Uploads 100MB Uploads 250MB Uploads 500MB Uploads 750MB Uploads 1GB Uploads Raw Data OData feeds of raw data from blocked/parallelized transfer tests Experiment Metadata Experiment Datasets Raw Data 256KB Blocks 512KB Blocks 1MB Blocks 2MB Blocks 4MB Blocks Excel worksheet showing summarizations and comparisons

    Read the article

  • RIF PRD: Presentation syntax issues

    - by Charles Young
    Over Christmas I got to play a bit with the W3C RIF PRD and came across a few issues which I thought I would record for posterity. Specifically, I was working on a grammar for the presentation syntax using a GLR grammar parser tool (I was using the current CTP of ‘M’ (MGrammer) and Intellipad – I do so hope the MS guys don’t kill off M and Intellipad now they have dropped the other parts of SQL Server Modelling). I realise that the presentation syntax is non-normative and that any issues with it do not therefore compromise the standard. However, presentation syntax is useful in its own right, and it would be great to iron out any issues in a future revision of the standard. The main issues are actually not to do with the grammar at all, but rather with the ‘running example’ in the RIF PRD recommendation. I started with the code provided in Example 9.1. There are several discrepancies when compared with the EBNF rules documented in the standard. Broadly the problems can be categorised as follows: ·      Parenthesis mismatch – the wrong number of parentheses are used in various places. For example, in GoldRule, the RHS of the rule (the ‘Then’) is nested in the LHS (‘the If’). In NewCustomerAndWidgetRule, the RHS is orphaned from the LHS. Together with additional incorrect parenthesis, this leads to orphanage of UnknownStatusRule from the entire Document. ·      Invalid use of parenthesis in ‘Forall’ constructs. Parenthesis should not be used to enclose formulae. Removal of the invalid parenthesis gave me a feeling of inconsistency when comparing formulae in Forall to formulae in If. The use of parenthesis is not actually inconsistent in these two context, but in an If construct it ‘feels’ as if you are enclosing formulae in parenthesis in a LISP-like fashion. In reality, the parenthesis is simply being used to group subordinate syntax elements. The fact that an If construct can contain only a single formula as an immediate child adds to this feeling of inconsistency. ·      Invalid representation of compact URIs (CURIEs) in the context of Frame productions. In several places the URIs are not qualified with a namespace prefix (‘ex1:’). This conflicts with the definition of CURIEs in the RIF Datatypes and Built-Ins 1.0 document. Here are the productions: CURIE          ::= PNAME_LN                  | PNAME_NS PNAME_LN       ::= PNAME_NS PN_LOCAL PNAME_NS       ::= PN_PREFIX? ':' PN_LOCAL       ::= ( PN_CHARS_U | [0-9] ) ((PN_CHARS|'.')* PN_CHARS)? PN_CHARS       ::= PN_CHARS_U                  | '-' | [0-9] | #x00B7                  | [#x0300-#x036F] | [#x203F-#x2040] PN_CHARS_U     ::= PN_CHARS_BASE                  | '_' PN_CHARS_BASE ::= [A-Z] | [a-z] | [#x00C0-#x00D6] | [#x00D8-#x00F6]                  | [#x00F8-#x02FF] | [#x0370-#x037D] | [#x037F-#x1FFF]                  | [#x200C-#x200D] | [#x2070-#x218F] | [#x2C00-#x2FEF]                  | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD]                  | [#x10000-#xEFFFF] PN_PREFIX      ::= PN_CHARS_BASE ((PN_CHARS|'.')* PN_CHARS)? The more I look at CURIEs, the more my head hurts! The RIF specification allows prefixes and colons without local names, which surprised me. However, the CURIE Syntax 1.0 working group note specifically states that this form is supported…and then promptly provides a syntactic definition that seems to preclude it! However, on (much) deeper inspection, it appears that ‘ex1:’ (for example) is allowed, but would really represent a ‘fragment’ of the ‘reference’, rather than a prefix! Ouch! This is so completely ambiguous that it surely calls into question the whole CURIE specification.   In any case, RIF does not allow local names without a prefix. ·      Missing ‘External’ specifiers for built-in functions and predicates.  The EBNF specification enforces this for terms within frames, but does not appear to enforce (what I believe is) the correct use of External on built-in predicates. In any case, the running example only specifies ‘External’ once on the predicate in UnknownStatusRule. External() is required in several other places. ·      The List used on the LHS of UnknownStatusRule is comma-delimited. This is not supported by the EBNF definition. Similarly, the argument list of pred:list-contains is illegally comma-delimited. ·      Unnecessary use of conjunction around a single formula in DiscountRule. This is strictly legal in the EBNF, but redundant.   All the above issues concern the presentation syntax used in the running example. There are a few minor issues with the grammar itself. Note that Michael Kiefer stated in his paper “Rule Interchange Format: The Framework” that: “The presentation syntax of RIF … is an abstract syntax and, as such, it omits certain details that might be important for unambiguous parsing.” ·      The grammar cannot differentiate unambiguously between strategies and priorities on groups. A processor is forced to resolve this by detecting the use of IRIs and integers. This could easily be fixed in the grammar.   ·      The grammar cannot unambiguously parse the ‘->’ operator in frames. Specifically, ‘-’ characters are allowed in PN_LOCAL names and hence a parser cannot determine if ‘status->’ is (‘status’ ‘->’) or (‘status-’ ‘>’).   One way to fix this is to amend the PN_LOCAL production as follows: PN_LOCAL ::= ( PN_CHARS_U | [0-9] ) ((PN_CHARS|'.')* ((PN_CHARS)-('-')))? However, unilaterally changing the definition of this production, which is defined in the SPARQL Query Language for RDF specification, makes me uncomfortable. ·      I assume that the presentation syntax is case-sensitive. I couldn’t find this stated anywhere in the documentation, but function/predicate names do appear to be documented as being case-sensitive. ·      The EBNF does not specify whitespace handling. A couple of productions (RULE and ACTION_BLOCK) are crafted to enforce the use of whitespace. This is not necessary. It seems inconsistent with the rest of the specification and can cause parsing issues. In addition, the Const production exhibits whitespaces issues. The intention may have been to disallow the use of whitespace around ‘^^’, but any direct implementation of the EBNF will probably allow whitespace between ‘^^’ and the SYMSPACE. Of course, I am being a little nit-picking about all this. On the whole, the EBNF translated very smoothly and directly to ‘M’ (MGrammar) and proved to be fairly complete. I have encountered far worse issues when translating other EBNF specifications into usable grammars.   I can’t imagine there would be any difficulty in implementing the same grammar in Antlr, COCO/R, gppg, XText, Bison, etc. A general observation, which repeats a point made above, is that the use of parenthesis in the presentation syntax can feel inconsistent and un-intuitive.   It isn’t actually inconsistent, but I think the presentation syntax could be improved by adopting braces, rather than parenthesis, to delimit subordinate syntax elements in a similar way to so many programming languages. The familiarity of braces would communicate the structure of the syntax more clearly to people like me.  If braces were adopted, parentheses could be retained around ‘var (frame | ‘new()’) constructs in action blocks. This use of parenthesis feels very LISP-like, and I think that this is my issue. It’s as if the presentation syntax represents the deformed love-child of LISP and C. In some places (specifically, action blocks), parenthesis is used in a LISP-like fashion. In other places it is used like braces in C. I find this quite confusing. Here is a corrected version of the running example (Example 9.1) in compliant presentation syntax: Document(    Prefix( ex1 <http://example.com/2009/prd2> )    (* ex1:CheckoutRuleset *)  Group rif:forwardChaining (     (* ex1:GoldRule *)    Group 10 (      Forall ?customer such that And(?customer # ex1:Customer                                     ?customer[ex1:status->"Silver"])        (Forall ?shoppingCart such that ?customer[ex1:shoppingCart->?shoppingCart]           (If Exists ?value (And(?shoppingCart[ex1:value->?value]                                  External(pred:numeric-greater-than-or-equal(?value 2000))))            Then Do(Modify(?customer[ex1:status->"Gold"])))))      (* ex1:DiscountRule *)    Group (      Forall ?customer such that ?customer # ex1:Customer        (If Or( ?customer[ex1:status->"Silver"]                ?customer[ex1:status->"Gold"])         Then Do ((?s ?customer[ex1:shoppingCart-> ?s])                  (?v ?s[ex1:value->?v])                  Modify(?s [ex1:value->External(func:numeric-multiply (?v 0.95))]))))      (* ex1:NewCustomerAndWidgetRule *)    Group (      Forall ?customer such that And(?customer # ex1:Customer                                     ?customer[ex1:status->"New"] )        (If Exists ?shoppingCart ?item                   (And(?customer[ex1:shoppingCart->?shoppingCart]                        ?shoppingCart[ex1:containsItem->?item]                        ?item # ex1:Widget ) )         Then Do( (?s ?customer[ex1:shoppingCart->?s])                  (?val ?s[ex1:value->?val])                  (?voucher ?customer[ex1:voucher->?voucher])                  Retract(?customer[ex1:voucher->?voucher])                  Retract(?voucher)                  Modify(?s[ex1:value->External(func:numeric-multiply(?val 0.90))]))))      (* ex1:UnknownStatusRule *)    Group (      Forall ?customer such that ?customer # ex1:Customer        (If Not(Exists ?status                       (And(?customer[ex1:status->?status]                            External(pred:list-contains(List("New" "Bronze" "Silver" "Gold") ?status)) )))         Then Do( Execute(act:print(External(func:concat("New customer: " ?customer))))                  Assert(?customer[ex1:status->"New"]))))  ) )   I hope that helps someone out there :-)

    Read the article

  • How to set Grub to automatically load Xen kernel

    - by Cerin
    How do you configure Grub to automatically use the Xen kernel under Ubuntu 11.10? No matter what I do, it loads the first menuentry. The only way I can get it to load Xen is to manually select the kernel, which I can't do if I have to reboot the server remotely, or there's a power failure and the machine automatically boots up when power's restored, etc. It's driving me nuts. In my /boot/grub/grub.cfg, the Xen kernel is at index 4 (i.e. it's the 5th menuentry). So I've tried: Setting GRUB_DEFAULT=4, and running sudo update-grub Setting GRUB_DEFAULT=saved and GRUB_SAVEDEFAULT=true, and running sudo update-grub Setting GRUB_DEFAULT="Ubuntu GNU/Linux, with Xen 4.1-amd64 and Linux 3.0.0-16-server", and running sudo update-grub None of these work. It continues to load the first menuentry, which is "Ubuntu, with Linux 3.0.0-16-server". Below is my current /boot/grub/grub.cfg. What am I doing wrong? # # DO NOT EDIT THIS FILE # # It is automatically generated by grub-mkconfig using templates # from /etc/grub.d and settings from /etc/default/grub # ### BEGIN /etc/grub.d/00_header ### if [ -s $prefix/grubenv ]; then set have_grubenv=true load_env fi set default="Ubuntu GNU/Linux, with Xen 4.1-amd64 and Linux 3.0.0-16-server" if [ "${prev_saved_entry}" ]; then set saved_entry="${prev_saved_entry}" save_env saved_entry set prev_saved_entry= save_env prev_saved_entry set boot_once=true fi function savedefault { if [ -z "${boot_once}" ]; then saved_entry="${chosen}" save_env saved_entry fi } function recordfail { set recordfail=1 if [ -n "${have_grubenv}" ]; then if [ -z "${boot_once}" ]; then save_env recordfail; fi; fi } function load_video { insmod vbe insmod vga insmod video_bochs insmod video_cirrus } insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac if loadfont /usr/share/grub/unicode.pf2 ; then set gfxmode=auto load_video insmod gfxterm insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac set locale_dir=($root)/boot/grub/locale set lang=en_US insmod gettext fi terminal_output gfxterm if [ "${recordfail}" = 1 ]; then set timeout=-1 else set timeout=2 fi ### END /etc/grub.d/00_header ### ### BEGIN /etc/grub.d/05_debian_theme ### set menu_color_normal=white/black set menu_color_highlight=black/light-gray if background_color 44,0,30; then clear fi ### END /etc/grub.d/05_debian_theme ### ### BEGIN /etc/grub.d/10_linux ### if [ ${recordfail} != 1 ]; then if [ -e ${prefix}/gfxblacklist.txt ]; then if hwmatch ${prefix}/gfxblacklist.txt 3; then if [ ${match} = 0 ]; then set linux_gfx_mode=keep else set linux_gfx_mode=text fi else set linux_gfx_mode=text fi else set linux_gfx_mode=keep fi else set linux_gfx_mode=text fi export linux_gfx_mode if [ "$linux_gfx_mode" != "text" ]; then load_video; fi menuentry 'Ubuntu, with Linux 3.0.0-16-server' --class ubuntu --class gnu-linux --class gnu --class os { recordfail set gfxpayload=$linux_gfx_mode insmod gzio insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac linux /boot/vmlinuz-3.0.0-16-server root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro initrd /boot/initrd.img-3.0.0-16-server } menuentry 'Ubuntu, with Linux 3.0.0-16-server (recovery mode)' --class ubuntu --class gnu-linux --class gnu --class os { recordfail insmod gzio insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac echo 'Loading Linux 3.0.0-16-server ...' linux /boot/vmlinuz-3.0.0-16-server root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro recovery nomodeset echo 'Loading initial ramdisk ...' initrd /boot/initrd.img-3.0.0-16-server } submenu "Previous Linux versions" { menuentry 'Ubuntu, with Linux 3.0.0-12-server' --class ubuntu --class gnu-linux --class gnu --class os { recordfail set gfxpayload=$linux_gfx_mode insmod gzio insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac linux /boot/vmlinuz-3.0.0-12-server root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro initrd /boot/initrd.img-3.0.0-12-server } menuentry 'Ubuntu, with Linux 3.0.0-12-server (recovery mode)' --class ubuntu --class gnu-linux --class gnu --class os { recordfail insmod gzio insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac echo 'Loading Linux 3.0.0-12-server ...' linux /boot/vmlinuz-3.0.0-12-server root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro recovery nomodeset echo 'Loading initial ramdisk ...' initrd /boot/initrd.img-3.0.0-12-server } } ### END /etc/grub.d/10_linux ### ### BEGIN /etc/grub.d/20_linux_xen ### submenu "Xen 4.1-amd64" { menuentry 'Ubuntu GNU/Linux, with Xen 4.1-amd64 and Linux 3.0.0-16-server' --class ubuntu --class gnu-linux --class gnu --class os --class xen { insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac echo 'Loading Xen 4.1-amd64 ...' multiboot /boot/xen-4.1-amd64.gz placeholder echo 'Loading Linux 3.0.0-16-server ...' module /boot/vmlinuz-3.0.0-16-server placeholder root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro echo 'Loading initial ramdisk ...' module /boot/initrd.img-3.0.0-16-server } menuentry 'Ubuntu GNU/Linux, with Xen 4.1-amd64 and Linux 3.0.0-16-server (recovery mode)' --class ubuntu --class gnu-linux --class gnu --class os --class xen { insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac echo 'Loading Xen 4.1-amd64 ...' multiboot /boot/xen-4.1-amd64.gz placeholder echo 'Loading Linux 3.0.0-16-server ...' module /boot/vmlinuz-3.0.0-16-server placeholder root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro single echo 'Loading initial ramdisk ...' module /boot/initrd.img-3.0.0-16-server } menuentry 'Ubuntu GNU/Linux, with Xen 4.1-amd64 and Linux 3.0.0-12-server' --class ubuntu --class gnu-linux --class gnu --class os --class xen { insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac echo 'Loading Xen 4.1-amd64 ...' multiboot /boot/xen-4.1-amd64.gz placeholder echo 'Loading Linux 3.0.0-12-server ...' module /boot/vmlinuz-3.0.0-12-server placeholder root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro echo 'Loading initial ramdisk ...' module /boot/initrd.img-3.0.0-12-server } menuentry 'Ubuntu GNU/Linux, with Xen 4.1-amd64 and Linux 3.0.0-12-server (recovery mode)' --class ubuntu --class gnu-linux --class gnu --class os --class xen { insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac echo 'Loading Xen 4.1-amd64 ...' multiboot /boot/xen-4.1-amd64.gz placeholder echo 'Loading Linux 3.0.0-12-server ...' module /boot/vmlinuz-3.0.0-12-server placeholder root=UUID=d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac ro single echo 'Loading initial ramdisk ...' module /boot/initrd.img-3.0.0-12-server } } ### END /etc/grub.d/20_linux_xen ### ### BEGIN /etc/grub.d/20_memtest86+ ### menuentry "Memory test (memtest86+)" { insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac linux16 /boot/memtest86+.bin } menuentry "Memory test (memtest86+, serial console 115200)" { insmod raid insmod mdraid1x insmod part_msdos insmod part_msdos insmod ext2 set root='(mduuid/be73165bc31d6f5cd00d05036c7b964f)' search --no-floppy --fs-uuid --set=root d72bad3f-9ed7-44b9-b3d1-d7af9f62a8ac linux16 /boot/memtest86+.bin console=ttyS0,115200n8 } ### END /etc/grub.d/20_memtest86+ ### ### BEGIN /etc/grub.d/30_os-prober ### ### END /etc/grub.d/30_os-prober ### ### BEGIN /etc/grub.d/40_custom ### # This file provides an easy way to add custom menu entries. Simply type the # menu entries you want to add after this comment. Be careful not to change # the 'exec tail' line above. ### END /etc/grub.d/40_custom ### ### BEGIN /etc/grub.d/41_custom ### if [ -f $prefix/custom.cfg ]; then source $prefix/custom.cfg; fi ### END /etc/grub.d/41_custom ###

    Read the article

  • RiverTrail - JavaScript GPPGU Data Parallelism

    - by JoshReuben
    Where is WebCL ? The Khronos WebCL working group is working on a JavaScript binding to the OpenCL standard so that HTML 5 compliant browsers can host GPGPU web apps – e.g. for image processing or physics for WebGL games - http://www.khronos.org/webcl/ . While Nokia & Samsung have some protype WebCL APIs, Intel has one-upped them with a higher level of abstraction: RiverTrail. Intro to RiverTrail Intel Labs JavaScript RiverTrail provides GPU accelerated SIMD data-parallelism in web applications via a familiar JavaScript programming paradigm. It extends JavaScript with simple deterministic data-parallel constructs that are translated at runtime into a low-level hardware abstraction layer. With its high-level JS API, programmers do not have to learn a new language or explicitly manage threads, orchestrate shared data synchronization or scheduling. It has been proposed as a draft specification to ECMA a (known as ECMA strawman). RiverTrail runs in all popular browsers (except I.E. of course). To get started, download a prebuilt version https://github.com/downloads/RiverTrail/RiverTrail/rivertrail-0.17.xpi , install Intel's OpenCL SDK http://www.intel.com/go/opencl and try out the interactive River Trail shell http://rivertrail.github.com/interactive For a video overview, see  http://www.youtube.com/watch?v=jueg6zB5XaM . ParallelArray the ParallelArray type is the central component of this API & is a JS object that contains ordered collections of scalars – i.e. multidimensional uniform arrays. A shape property describes the dimensionality and size– e.g. a 2D RGBA image will have shape [height, width, 4]. ParallelArrays are immutable & fluent – they are manipulated by invoking methods on them which produce new ParallelArray objects. ParallelArray supports several constructors over arrays, functions & even the canvas. // Create an empty Parallel Array var pa = new ParallelArray(); // pa0 = <>   // Create a ParallelArray out of a nested JS array. // Note that the inner arrays are also ParallelArrays var pa = new ParallelArray([ [0,1], [2,3], [4,5] ]); // pa1 = <<0,1>, <2,3>, <4.5>>   // Create a two-dimensional ParallelArray with shape [3, 2] using the comprehension constructor var pa = new ParallelArray([3, 2], function(iv){return iv[0] * iv[1];}); // pa7 = <<0,0>, <0,1>, <0,2>>   // Create a ParallelArray from canvas.  This creates a PA with shape [w, h, 4], var pa = new ParallelArray(canvas); // pa8 = CanvasPixelArray   ParallelArray exposes fluent API functions that take an elemental JS function for data manipulation: map, combine, scan, filter, and scatter that return a new ParallelArray. Other functions are scalar - reduce  returns a scalar value & get returns the value located at a given index. The onus is on the developer to ensure that the elemental function does not defeat data parallelization optimization (avoid global var manipulation, recursion). For reduce & scan, order is not guaranteed - the onus is on the dev to provide an elemental function that is commutative and associative so that scan will be deterministic – E.g. Sum is associative, but Avg is not. map Applies a provided elemental function to each element of the source array and stores the result in the corresponding position in the result array. The map method is shape preserving & index free - can not inspect neighboring values. // Adding one to each element. var source = new ParallelArray([1,2,3,4,5]); var plusOne = source.map(function inc(v) {     return v+1; }); //<2,3,4,5,6> combine Combine is similar to map, except an index is provided. This allows elemental functions to access elements from the source array relative to the one at the current index position. While the map method operates on the outermost dimension only, combine, can choose how deep to traverse - it provides a depth argument to specify the number of dimensions it iterates over. The elemental function of combine accesses the source array & the current index within it - element is computed by calling the get method of the source ParallelArray object with index i as argument. It requires more code but is more expressive. var source = new ParallelArray([1,2,3,4,5]); var plusOne = source.combine(function inc(i) { return this.get(i)+1; }); reduce reduces the elements from an array to a single scalar result – e.g. Sum. // Calculate the sum of the elements var source = new ParallelArray([1,2,3,4,5]); var sum = source.reduce(function plus(a,b) { return a+b; }); scan Like reduce, but stores the intermediate results – return a ParallelArray whose ith elements is the results of using the elemental function to reduce the elements between 0 and I in the original ParallelArray. // do a partial sum var source = new ParallelArray([1,2,3,4,5]); var psum = source.scan(function plus(a,b) { return a+b; }); //<1, 3, 6, 10, 15> scatter a reordering function - specify for a certain source index where it should be stored in the result array. An optional conflict function can prevent an exception if two source values are assigned the same position of the result: var source = new ParallelArray([1,2,3,4,5]); var reorder = source.scatter([4,0,3,1,2]); // <2, 4, 5, 3, 1> // if there is a conflict use the max. use 33 as a default value. var reorder = source.scatter([4,0,3,4,2], 33, function max(a, b) {return a>b?a:b; }); //<2, 33, 5, 3, 4> filter // filter out values that are not even var source = new ParallelArray([1,2,3,4,5]); var even = source.filter(function even(iv) { return (this.get(iv) % 2) == 0; }); // <2,4> Flatten used to collapse the outer dimensions of an array into a single dimension. pa = new ParallelArray([ [1,2], [3,4] ]); // <<1,2>,<3,4>> pa.flatten(); // <1,2,3,4> Partition used to restore the original shape of the array. var pa = new ParallelArray([1,2,3,4]); // <1,2,3,4> pa.partition(2); // <<1,2>,<3,4>> Get return value found at the indices or undefined if no such value exists. var pa = new ParallelArray([0,1,2,3,4], [10,11,12,13,14], [20,21,22,23,24]) pa.get([1,1]); // 11 pa.get([1]); // <10,11,12,13,14>

    Read the article

  • Community Outreach - Where Should I Go

    - by Roger Brinkley
    A few days ago I was talking to person new to community development and they asked me what guidelines I used to determine the worthiness of a particular event. After our conversation was over I thought about it a little bit more and figured out there are three ways to determine if any event (be it conference, blog, podcast or other social medias) is worth doing: Transferability, Multiplication, and Impact. Transferability - Is what I have to say useful to the people that are going to hear it. For instance, consider a company that has product offering that can connect up using a number of languages like Scala, Grovey or Java. Sending a Scala expert to talk about Scala and the product is not transferable to a Java User Group, but a Java expert doing the same talk with a Java slant is. Similarly, talking about JavaFX to any Java User Group meeting in Brazil was pretty much a wasted effort until it was open sourced. Once it was open sourced it was well received. You can also look at transferability in relation to the subject matter that you're dealing with. How transferable is a presentation that I create. Can I, or a technical writer on the staff, turn it into some technical document. Could it be converted into some type of screen cast. If we have a regular podcast can we make a reference to the document, catch the high points or turn it into a interview. Is there a way of using this in the sales group. In other words is the document purely one dimensional or can it be re-purposed in other forms. Multiplication - On every trip I'm looking for 2 to 5 solid connections that I can make with developers. These are long term connections, because I know that once that relationship is established it will lead to another 2 - 5 from that connection and within a couple of years were talking about some 100 connections from just one developer. For instance, when I was working on JavaHelp in 2000 I hired a science teacher with a programming background. We've developed a very tight relationship over the year though we rarely see each other more than once a year. But at this JavaOne, one of his employees came up to me and said, "Richard (Rick Hard in Czech) told me to tell you that he couldn't make it to JavaOne this year but if I saw you to tell you hi". Another example is from my Mobile & Embedded days in Brasil. On our very first FISL trip about 5 years ago there were two university students that had created a project called "Marge". Marge was a Bluetooth framework that made connecting bluetooth devices easier. I invited them to a "Sun" dinner that evening. Originally they were planning on leaving that afternoon, but they changed their plans recognizing the opportunity. Their eyes were as big a saucers when they realized the level of engineers at the meeting. They went home started a JUG in Florianoplis that we've visited more than a couple of times. One of them went to work for Brazilian government lab like Berkley Labs, MIT Lab, John Hopkins Applied Physicas Labs or Lincoln Labs in the US. That presented us with an opportunity to show Embedded Java as a possibility for some of the work they were doing there. Impact - The final criteria is how life changing is what I'm going to say be to the individuals I'm reaching. A t-shirt is just a token, but when I reach down and tug at their developer hearts then I know I've succeeded. I'll never forget one time we flew all night to reach Joan Pasoa in Northern Brazil. We arrived at 2am went immediately to our hotel only to be woken up at 6 am to travel 2 hours by car to the presentation hall. When we arrived we were totally exhausted. Outside the facility there were 500 people lined up to hear 6 speakers for the day. That itself was uplifting.  I delivered one of my favorite talks on "I have passion". It was a talk on golf and embedded java development, "Find your passion". When we finished a couple of first year students came up to me and said how much my talk had inspired them. FISL is another great example. I had been about 4 years in a row. FISL is a very young group of developers so capturing their attention is important. Several of the students will come back 2 or 3 years later and ask me questions about research or jobs. And then there's Louis. Louis is one my favorite Brazilians. I can only describe him as a big Brazilian teddy bear. I see him every year at FISL. He works primarily in Java EE but he's attended every single one of my talks over the last 4 years. I can't tell you why, but he always greets me and gives me a hug. For some reason I've had a real impact. And of course when it comes to impact you don't just measure a presentation but every single interaction you have at an event. It's the hall way conversations, the booth conversations, but more importantly it's the conversations at dinner tables or in the cars when you're getting transported to an event. There's a good story that illustrates this. Last year in the spring I was traveling to Goiânia in Brazil. I've been there many times and leaders there no me well. One young man has picked me up at the airport on more than one occasion. We were going out to dinner one evening and he brought his girl friend along. One thing let to another and I eventually asked him, in front of her, "Why haven't you asked her to marry you?" There were all kinds of excuses and she just looked at him and smiled. When I came back in December for JavaOne he came and sought me. "I just want to tell you that I thought a lot about what you said, and I asked her to marry me. We're getting married next Spring." Sometimes just one presentation is all it takes to make an impact. Other times it takes years. Some impacts are directly related to the company and some are more personal in nature. It doesn't matter which it is because it's having the impact that matters.

    Read the article

  • DTracing TCP congestion control

    - by user12820842
    In a previous post, I showed how we can use DTrace to probe TCP receive and send window events. TCP receive and send windows are in effect both about flow-controlling how much data can be received - the receive window reflects how much data the local TCP is prepared to receive, while the send window simply reflects the size of the receive window of the peer TCP. Both then represent flow control as imposed by the receiver. However, consider that without the sender imposing flow control, and a slow link to a peer, TCP will simply fill up it's window with sent segments. Dealing with multiple TCP implementations filling their peer TCP's receive windows in this manner, busy intermediate routers may drop some of these segments, leading to timeout and retransmission, which may again lead to drops. This is termed congestion, and TCP has multiple congestion control strategies. We can see that in this example, we need to have some way of adjusting how much data we send depending on how quickly we receive acknowledgement - if we get ACKs quickly, we can safely send more segments, but if acknowledgements come slowly, we should proceed with more caution. More generally, we need to implement flow control on the send side also. Slow Start and Congestion Avoidance From RFC2581, let's examine the relevant variables: "The congestion window (cwnd) is a sender-side limit on the amount of data the sender can transmit into the network before receiving an acknowledgment (ACK). Another state variable, the slow start threshold (ssthresh), is used to determine whether the slow start or congestion avoidance algorithm is used to control data transmission" Slow start is used to probe the network's ability to handle transmission bursts both when a connection is first created and when retransmission timers fire. The latter case is important, as the fact that we have effectively lost TCP data acts as a motivator for re-probing how much data the network can handle from the sending TCP. The congestion window (cwnd) is initialized to a relatively small value, generally a low multiple of the sending maximum segment size. When slow start kicks in, we will only send that number of bytes before waiting for acknowledgement. When acknowledgements are received, the congestion window is increased in size until cwnd reaches the slow start threshold ssthresh value. For most congestion control algorithms the window increases exponentially under slow start, assuming we receive acknowledgements. We send 1 segment, receive an ACK, increase the cwnd by 1 MSS to 2*MSS, send 2 segments, receive 2 ACKs, increase the cwnd by 2*MSS to 4*MSS, send 4 segments etc. When the congestion window exceeds the slow start threshold, congestion avoidance is used instead of slow start. During congestion avoidance, the congestion window is generally updated by one MSS for each round-trip-time as opposed to each ACK, and so cwnd growth is linear instead of exponential (we may receive multiple ACKs within a single RTT). This continues until congestion is detected. If a retransmit timer fires, congestion is assumed and the ssthresh value is reset. It is reset to a fraction of the number of bytes outstanding (unacknowledged) in the network. At the same time the congestion window is reset to a single max segment size. Thus, we initiate slow start until we start receiving acknowledgements again, at which point we can eventually flip over to congestion avoidance when cwnd ssthresh. Congestion control algorithms differ most in how they handle the other indication of congestion - duplicate ACKs. A duplicate ACK is a strong indication that data has been lost, since they often come from a receiver explicitly asking for a retransmission. In some cases, a duplicate ACK may be generated at the receiver as a result of packets arriving out-of-order, so it is sensible to wait for multiple duplicate ACKs before assuming packet loss rather than out-of-order delivery. This is termed fast retransmit (i.e. retransmit without waiting for the retransmission timer to expire). Note that on Oracle Solaris 11, the congestion control method used can be customized. See here for more details. In general, 3 or more duplicate ACKs indicate packet loss and should trigger fast retransmit . It's best not to revert to slow start in this case, as the fact that the receiver knew it was missing data suggests it has received data with a higher sequence number, so we know traffic is still flowing. Falling back to slow start would be excessive therefore, so fast recovery is used instead. Observing slow start and congestion avoidance The following script counts TCP segments sent when under slow start (cwnd ssthresh). #!/usr/sbin/dtrace -s #pragma D option quiet tcp:::connect-request / start[args[1]-cs_cid] == 0/ { start[args[1]-cs_cid] = 1; } tcp:::send / start[args[1]-cs_cid] == 1 && args[3]-tcps_cwnd tcps_cwnd_ssthresh / { @c["Slow start", args[2]-ip_daddr, args[4]-tcp_dport] = count(); } tcp:::send / start[args[1]-cs_cid] == 1 && args[3]-tcps_cwnd args[3]-tcps_cwnd_ssthresh / { @c["Congestion avoidance", args[2]-ip_daddr, args[4]-tcp_dport] = count(); } As we can see the script only works on connections initiated since it is started (using the start[] associative array with the connection ID as index to set whether it's a new connection (start[cid] = 1). From there we simply differentiate send events where cwnd ssthresh (congestion avoidance). Here's the output taken when I accessed a YouTube video (where rport is 80) and from an FTP session where I put a large file onto a remote system. # dtrace -s tcp_slow_start.d ^C ALGORITHM RADDR RPORT #SEG Slow start 10.153.125.222 20 6 Slow start 138.3.237.7 80 14 Slow start 10.153.125.222 21 18 Congestion avoidance 10.153.125.222 20 1164 We see that in the case of the YouTube video, slow start was exclusively used. Most of the segments we sent in that case were likely ACKs. Compare this case - where 14 segments were sent using slow start - to the FTP case, where only 6 segments were sent before we switched to congestion avoidance for 1164 segments. In the case of the FTP session, the FTP data on port 20 was predominantly sent with congestion avoidance in operation, while the FTP session relied exclusively on slow start. For the default congestion control algorithm - "newreno" - on Solaris 11, slow start will increase the cwnd by 1 MSS for every acknowledgement received, and by 1 MSS for each RTT in congestion avoidance mode. Different pluggable congestion control algorithms operate slightly differently. For example "highspeed" will update the slow start cwnd by the number of bytes ACKed rather than the MSS. And to finish, here's a neat oneliner to visually display the distribution of congestion window values for all TCP connections to a given remote port using a quantization. In this example, only port 80 is in use and we see the majority of cwnd values for that port are in the 4096-8191 range. # dtrace -n 'tcp:::send { @q[args[4]-tcp_dport] = quantize(args[3]-tcps_cwnd); }' dtrace: description 'tcp:::send ' matched 10 probes ^C 80 value ------------- Distribution ------------- count -1 | 0 0 |@@@@@@ 5 1 | 0 2 | 0 4 | 0 8 | 0 16 | 0 32 | 0 64 | 0 128 | 0 256 | 0 512 | 0 1024 | 0 2048 |@@@@@@@@@ 8 4096 |@@@@@@@@@@@@@@@@@@@@@@@@@@ 23 8192 | 0

    Read the article

  • ApiChange Is Released!

    - by Alois Kraus
    I have been working on little tool to simplify my life and perhaps yours as developer as well. It is basically a command line tool that allows you to execute queries on your compiled .NET code base. The main purpose is to find out how big the impact of an api change would be if you changed this or that.  Now you can do high level operations like Diff public types for breaking changes. Who uses a method? Who uses a type? Who uses implements an interface? Who references me? What format has the binary  (32/64, Managed C++, Pure IL, Unmanaged)? Search for all event subscribers and unsubscribers. A unique feature is to check for event subscription imbalances. Forgotten event subscriptions are the 90% cause of managed memory leaks. It is done at a per class level. If one class does subscribe to one event more often than it does unsubscribe it is treated as possible event subscription imbalance. Another unique ability is to search for users of string literals which allows you to track users of a string constant which is not possible otherwise. For incremental builds the ShowRebuildTargets command can be used to identify the dependant targets that need a rebuild after you did compile one assembly. It has some heuristics in place to determine the impact of breaking changes and finds out which targets need to be recompiled as well. It has a ton of other features and a an API to access these things programmatically so you can build upon these simple queries create even better tools. Perhaps we get a Visual Studio plug in? You can download it from CodePlex here. It works via XCopy deployment. Simply let it run and check the command line help out. The best feature in my opinion is that the output of nearly all commands can be piped to Excel for further analysis. Since it does read also the pdbs it can show you the source file name and line number as well for all matches. The following picture shows the output of a –WhousesType query. The following command checks where type from BaseLibraryV1.dll are used inside DependantLibV1.dll. All matches are printed out with the reason and matching item along with file and line number. There is even a hyper link to the match which will open Visual Studio. ApiChange -whousestype "*" BaseLibraryV1.dll -in DependantLibV1.dll –excel The "*” is the actual query which means all types. The syntax is the same like in C# just that placeholders are allowed ;-). More info's can be found at the Codeplex Documentation.     The tool was developed in a TDD style manner which means that it is heavily tested and already used by a quite large user base inside the company I do work for. Luckily for you I got the permission to make it public so you take advantage of it. It is fully instrumented with tracing. If you find bugs simply add the –trace command line switch to find out what is failing and send me the output. How is it done? Your first guess might be that it uses reflection. Wrong. It is based on Mono Cecil a free IL parser with a fantastic API to access all internals of a managed assembly. The speed is awesome and to make it even faster I did make the tool heavily multi threaded. The query above did execute in 1.8s with the Excel output. On a rather slow machine I can analyze over 1500 assemblies in less than 40s with a very low memory consumption. The true power of Mono Cecil is that I can load an assembly like any other data file. I have no problems unloading a file but if I would have used reflection I would need to unload a whole AppDomain just to get rid of one assembly in my memory. Just to give you a glimpse how ApiChange.Api.dll can be used I show you one of the unit tests:           public void Can_Find_GenericMethodInvocations_With_Type_Parameters()         { // 1. Create an aggregator to collect our matches             UsageQueryAggregator agg = new UsageQueryAggregator();   // 2. This is the type we want to search for. Load it via the type query             var decimalType = TypeQuery.GetTypeByName(TestConstants.MscorlibAssembly, "System.Decimal");   // 3. register the type query which searches for uses of the Decimal type             new WhoUsesType(agg, decimalType);   // 4. Search for all users of the Decimal type in the DependandLibV1Assembly             agg.Analyze(TestConstants.DependandLibV1Assembly);   // Extract matches and assert             Assert.AreEqual(2, agg.MethodMatches.Count, "Method match count");             Assert.AreEqual("UseGenericMethod", agg.MethodMatches[0].Match.Name);             Assert.AreEqual("UseGenericMethod", agg.MethodMatches[1].Match.Name);         } Many thanks go from here to Jb Evian for the creation of Mono.Cecil. Without this fantastic piece of code it would have been much much harder. There are other options around like the Common Compiler Infrastructure  Metadata Api which should do the same thing but it was not a real option since the Microsoft reader did fail on even simple assemblies (at least in September 2009 this was the case). Besides this I found the CCI Apis much harder to use. The only real competitor was Reflector which does support many things but does not let me access his cool high level analyze commands. So I decided to dig into the IL specs and as a result you can query your compiled binaries from the command line or programmatically. The best thing is you try it out for yourself and give me some feedback what you miss. If you want to contribute or have a cool idea what should be added drop me a mail at A Kraus1@___No [email protected]. There is much more inside the tool I did not talk about it (yet).

    Read the article

  • SQL Server Developer Tools &ndash; Codename Juneau vs. Red-Gate SQL Source Control

    - by Ajarn Mark Caldwell
    So how do the new SQL Server Developer Tools (previously code-named Juneau) stack up against SQL Source Control?  Read on to find out. At the PASS Community Summit a couple of weeks ago, it was announced that the previously code-named Juneau software would be released under the name of SQL Server Developer Tools with the release of SQL Server 2012.  This replacement for Database Projects in Visual Studio (also known in a former life as Data Dude) has some great new features.  I won’t attempt to describe them all here, but I will applaud Microsoft for making major improvements.  One of my favorite changes is the way database elements are broken down.  Previously every little thing was in its own file.  For example, indexes were each in their own file.  I always hated that.  Now, SSDT uses a pattern similar to Red-Gate’s and puts the indexes and keys into the same file as the overall table definition. Of course there are really cool features to keep your database model in sync with the actual source scripts, and the rename refactoring feature is now touted as being more than just a search and replace, but rather a “semantic-aware” search and replace.  Funny, it reminds me of SQL Prompt’s Smart Rename feature.  But I’m not writing this just to criticize Microsoft and argue that they are late to the party with this feature set.  Instead, I do see it as a viable alternative for folks who want all of their source code to be version controlled, but there are a couple of key trade-offs that you need to know about when you choose which tool set to use. First, the basics Both tool sets integrate with a wide variety of source control systems including the most popular: Subversion, GIT, Vault, and Team Foundation Server.  Both tools have integrated functionality to produce objects to upgrade your target database when you are ready (DACPACs in SSDT, integration with SQL Compare for SQL Source Control).  If you regularly live in Visual Studio or the Business Intelligence Development Studio (BIDS) then SSDT will likely be comfortable for you.  Like BIDS, SSDT is a Visual Studio Project Type that comes with SQL Server, and if you don’t already have Visual Studio installed, it will install the shell for you.  If you already have Visual Studio 2010 installed, then it will just add this as an available project type.  On the other hand, if you regularly live in SQL Server Management Studio (SSMS) then you will really enjoy the SQL Source Control integration from within SSMS.  Both tool sets store their database model in script files.  In SSDT, these are on your file system like other source files; in SQL Source Control, these are stored in the folder structure in your source control system, and you can always GET them to your file system if you want to browse them directly. For me, the key differentiating factors are 1) a single, unified check-in, and 2) migration scripts.  How you value those two features will likely make your decision for you. Unified Check-In If you do a continuous-integration (CI) style of development that triggers an automated build with unit testing on every check-in of source code, and you use Visual Studio for the rest of your development, then you will want to really consider SSDT.  Because it is just another project in Visual Studio, it can be added to your existing Solution, and you can then do a complete, or unified single check-in of all changes whether they are application or database changes.  This is simply not possible with SQL Source Control because it is in a different development tool (SSMS instead of Visual Studio) and there is no way to do one unified check-in between the two.  You CAN do really fast back-to-back check-ins, but there is the possibility that the automated build that is triggered from the first check-in will cause your unit tests to fail and the CI tool to report that you broke the build.  Of course, the automated build that is triggered from the second check-in which contains the “other half” of your changes should pass and so the amount of time that the build was broken may be very, very short, but if that is very, very important to you, then SQL Source Control just won’t work; you’ll have to use SSDT. Refactoring and Migrations If you work on a mature system, or on a not-so-mature but also not-so-well-designed system, where you want to refactor the database schema as you go along, but you can’t have data suddenly disappearing from your target system, then you’ll probably want to go with SQL Source Control.  As I wrote previously, there are a number of changes which you can make to your database that the comparison tools (both from Microsoft and Red Gate) simply cannot handle without the possibility (or probability) of data loss.  Currently, SSDT only offers you the ability to inject PRE and POST custom deployment scripts.  There is no way to insert your own script in the middle to override the default behavior of the tool.  In version 3.0 of SQL Source Control (Early Access version now available) you have that ability to create your own custom migration script to take the place of the commands that the tool would have done, and ensure the preservation of your data.  Or, even if the default tool behavior would have worked, but you simply know a better way then you can take control and do things your way instead of theirs. You Decide In the environment I work in, our automated builds are not triggered off of check-ins, but off of the clock (currently once per night) and so there is no point at which the automated build and unit tests will be triggered without having both sides of the development effort already checked-in.  Therefore having a unified check-in, while handy, is not critical for us.  As for migration scripts, these are critically important to us.  We do a lot of new development on systems that have already been in production for years, and it is not uncommon for us to need to do a refactoring of the database.  Because of the maturity of the existing system, that often involves data migrations or other additional SQL tasks that the comparison tools just can’t detect on their own.  Therefore, the ability to create a custom migration script to override the tool’s default behavior is very important to us.  And so, you can see why we will continue to use Red Gate SQL Source Control for the foreseeable future.

    Read the article

  • D2K to OA Framework Transition

    - by PRajkumar
    What is the difference between D2K form and OA Framework? It is a very innocent but important question for someone that desires to make transition from D2K to OA Framework. I hope you have already read and implemented OA Framework Getting Started. I will re-visit my own experience of implementing HelloWorld program in "OA Framework". When I implemented HelloWorld a year ago, I had no clue as to what I was doing & why I was doing those steps. I merely copied the steps from Oracle Tutorial without understanding them. Hence in this blog, I will try to explain in simple manner the meaning of OA Framework HelloWorld Program and compare the steps to D2K form [where possible]. To keep things simple, only basics will be discussed. Following key Steps were needed for HelloWorld Step 1 Create a new Workspace and a new Project as dictated by Oracle's tutorial. When defining project, you will specify a default package, which in this case was oracle.apps.ak.hello This means the following: - ak is the short name of the Application in Oracle           [means fnd_applications.short_name] hello is the name of your project Step 2 Next, you will create a OA Page within hello project Think OA Page as the fmx file itself in D2K. I am saying so because this page gets attached to the form function. This page will be created within hello project, hence the package name oracle.apps.ak.hello.webui Note the webui, it is a convention to have page in webui, means this page represents the Web User Interface You will assign the default AM [OAApplicationModule]. Think of AM "Connection Manager" and "Transaction State Manager" for your page          I can't co-relate this to anything in D2k, as there is no concept of Connection Pooling and that D2k is not stateless. Reason being that as soon as you kick off a D2K Form, it connects to a single session of Oracle and sticks to that single Oracle database session. So is not the case in OAF, hence AM is needed. Step 3 You create Region within the Page. ·         Region is what will store your fields. Text input fields will be of type messageTextInput. Think of Canvas in D2K. You can have nested regions. Stacked Canvas in D2K comes the closest to this component of OA Framework Step 4 Add a button to one of the nested regions The itemStyle should be submitButton, in case you want the page to be submitted when this button is clicked There is no WHEN-BUTTON-PRESSED trigger in OAF. In Framework, you will add a controller java code to handle events like Form Submit button clicks. JDeveloper generates the default code for you. Primarily two functions [should I call methods] will be created processRequest [for UI Rendering Handling] and processFormRequest          Think of processRequest as WHEN-NEW-FORM-INSTANCE, though processRequest is very restrictive. Note What is the difference between processRequest and processFormRequest? These two methods are available in the Default Controller class that gets created. processFormRequest This method is commonly used to react/respond to the event that has taken place, for example click of a button. Some examples are if(oapagecontext.getParameter("Cancel") != null) (Do your processing for Cancellation/ Rollback) if(oapagecontext.getParameter("Submit") != null) (Do your validations and commit here) if(oapagecontext.getParameter("Update") != null) (Do your validations and commit here) In the above three examples, you could be calling oapagecontext.forwardImmediately to re-direct the page navigation to some other page if needed. processRequest In this method, usually page rendering related code is written. Effectively, each GUI component is a bean that gets initialised during processRequest. Those who are familiar with D2K forms, something like pre-query may be written in this method. Step 5 In the controller to access the value in field "HelloName" the command is String userContent = pageContext.getParameter("HelloName"); In D2k, we used :block.field. In OAFramework, at submission of page, all the field values get passed into to OAPageContext object. Use getParameter to access the field value To set the value of the field, use OAMessageTextInputBean field HelloName = (OAMessageTextInputBean)webBean.findChildRecursive("HelloName"); fieldHelloName.setText(pageContext,"Setting the default value" ); Note when setting field value in controller: Note 1. Do not set the value in processFormRequest Note 2. If the field comes from View Object, then do not use setText in controller Note 3. For control fields [that are not based on View Objects], you can use setText to assign values in processRequest method Lets take some notes to expand beyond the HelloWorld Project Note 1 In D2K-forms we sort of created a Window, attached to Canvas, and then fields within that Canvas. However in OA Framework, think of Page being fmx/Window, think of Region being a Canvas, and fields being within Regions. This is not a formal/accurate understanding of analogy between D2k and Framework, but is close to being logical. Note 2 In D2k, your Forms fmb file was compiled to fmx. It was fmx file that was deployed on mid-tier. In case of OAF, your OA Page is nothing but a XML file. We call this MDS [meta data]. Whatever name you give to "Page" in OAF, an XML file of the same name gets created. This xml file must then be loaded into database by using XML Importer command. Note 3 Apart from MDS XML file, almost everything else is merely deployed to your mid-tier. Usually this is underneath $JAVA_TOP/oracle/apps/../.. All java files will go underneath java top/oracle/apps/../.. etc. Note 4 When building tutorial, ignore the steps for setting "Attribute Sets". These are not mandatory. Oracle might just have developed their tutorials without including these. Think of these like Visual Attributes of D2K forms Note 5 Controller is where you will write any java code in OA Framework. You can create a Controller per Page or have a different Controller for each of the Regions with the same Page. Note 6 In the method processFormRequest of the Controller, you can access the values of the page by using notation pageContext.getParameter("<fieldname here>"). This method processFormRequest is executed when the OAF Screen/Page is submitted by click of a button. Note 7 Inside the controller, all the Database Related interactions for example interaction with View Objects happen via Application Module. But why so? Because Application Module Manages the transaction state of the Application. OAApplicationModuleImpl oaapplicationmoduleimpl = OAApplicationModuleImpl)oapagecontext.getApplicationModule(oawebbean); OADBTransaction oadbtransaction = OADBTransaction)oaapplicationmoduleimpl.getDBTransaction(); Note 8 In D2K, we have control block or a block based on database view. Similarly, in OA Framework, if the field does not have view Object attached, then it is like a control field. Hence in HelloWorld example, field HelloName is a control field [in D2K terminology]. A view Object can either be based on a view/table, synonym or on a SQL statement. Note 9 I wish to access the fields in multi record block that is based on view Object. Can I do this in Controller? Sure you can. To traverse through those records, do the below ·         Get the reference to the View Object using (OAViewObject)oapagecontext.getApplicationModule(oawebbean).findViewObject("VO Name Here") ·         Loop through the records in View Objects using count returned from oaviewobject.getFetchedRowCount() ·         For each record, fetch the value of the fields within the loop as oracle.jbo.Row row = oaviewobject.getRowAtRangeIndex(loop index here); (String)row.getAttribute("Column name of VO here ");

    Read the article

  • C#/.NET Little Wonders: Interlocked Read() and Exchange()

    - by James Michael Hare
    Once again, in this series of posts I look at the parts of the .NET Framework that may seem trivial, but can help improve your code by making it easier to write and maintain. The index of all my past little wonders posts can be found here. Last time we discussed the Interlocked class and its Add(), Increment(), and Decrement() methods which are all useful for updating a value atomically by adding (or subtracting).  However, this begs the question of how do we set and read those values atomically as well? Read() – Read a value atomically Let’s begin by examining the following code: 1: public class Incrementor 2: { 3: private long _value = 0; 4:  5: public long Value { get { return _value; } } 6:  7: public void Increment() 8: { 9: Interlocked.Increment(ref _value); 10: } 11: } 12:  It uses an interlocked increment, as we discuss in my previous post (here), so we know that the increment will be thread-safe.  But, to realize what’s potentially wrong we have to know a bit about how atomic reads are in 32 bit and 64 bit .NET environments. When you are dealing with an item smaller or equal to the system word size (such as an int on a 32 bit system or a long on a 64 bit system) then the read is generally atomic, because it can grab all of the bits needed at once.  However, when dealing with something larger than the system word size (reading a long on a 32 bit system for example), it cannot grab the whole value at once, which can lead to some problems since this read isn’t atomic. For example, this means that on a 32 bit system we may read one half of the long before another thread increments the value, and the other half of it after the increment.  To protect us from reading an invalid value in this manner, we can do an Interlocked.Read() to force the read to be atomic (of course, you’d want to make sure any writes or increments are atomic also): 1: public class Incrementor 2: { 3: private long _value = 0; 4:  5: public long Value 6: { 7: get { return Interlocked.Read(ref _value); } 8: } 9:  10: public void Increment() 11: { 12: Interlocked.Increment(ref _value); 13: } 14: } Now we are guaranteed that we will read the 64 bit value atomically on a 32 bit system, thus ensuring our thread safety (assuming all other reads, writes, increments, etc. are likewise protected).  Note that as stated before, and according to the MSDN (here), it isn’t strictly necessary to use Interlocked.Read() for reading 64 bit values on 64 bit systems, but for those still working in 32 bit environments, it comes in handy when dealing with long atomically. Exchange() – Exchanges two values atomically Exchange() lets us store a new value in the given location (the ref parameter) and return the old value as a result. So just as Read() allows us to read atomically, one use of Exchange() is to write values atomically.  For example, if we wanted to add a Reset() method to our Incrementor, we could do something like this: 1: public void Reset() 2: { 3: _value = 0; 4: } But the assignment wouldn’t be atomic on 32 bit systems, since the word size is 32 bits and the variable is a long (64 bits).  Thus our assignment could have only set half the value when a threaded read or increment happens, which would put us in a bad state. So instead, we could write Reset() like this: 1: public void Reset() 2: { 3: Interlocked.Exchange(ref _value, 0); 4: } And we’d be safe again on a 32 bit system. But this isn’t the only reason Exchange() is valuable.  The key comes in realizing that Exchange() doesn’t just set a new value, it returns the old as well in an atomic step.  Hence the name “exchange”: you are swapping the value to set with the stored value. So why would we want to do this?  Well, anytime you want to set a value and take action based on the previous value.  An example of this might be a scheme where you have several tasks, and during every so often, each of the tasks may nominate themselves to do some administrative chore.  Perhaps you don’t want to make this thread dedicated for whatever reason, but want to be robust enough to let any of the threads that isn’t currently occupied nominate itself for the job.  An easy and lightweight way to do this would be to have a long representing whether someone has acquired the “election” or not.  So a 0 would indicate no one has been elected and 1 would indicate someone has been elected. We could then base our nomination strategy as follows: every so often, a thread will attempt an Interlocked.Exchange() on the long and with a value of 1.  The first thread to do so will set it to a 1 and return back the old value of 0.  We can use this to show that they were the first to nominate and be chosen are thus “in charge”.  Anyone who nominates after that will attempt the same Exchange() but will get back a value of 1, which indicates that someone already had set it to a 1 before them, thus they are not elected. Then, the only other step we need take is to remember to release the election flag once the elected thread accomplishes its task, which we’d do by setting the value back to 0.  In this way, the next thread to nominate with Exchange() will get back the 0 letting them know they are the new elected nominee. Such code might look like this: 1: public class Nominator 2: { 3: private long _nomination = 0; 4: public bool Elect() 5: { 6: return Interlocked.Exchange(ref _nomination, 1) == 0; 7: } 8: public bool Release() 9: { 10: return Interlocked.Exchange(ref _nomination, 0) == 1; 11: } 12: } There’s many ways to do this, of course, but you get the idea.  Running 5 threads doing some “sleep” work might look like this: 1: var nominator = new Nominator(); 2: var random = new Random(); 3: Parallel.For(0, 5, i => 4: { 5:  6: for (int j = 0; j < _iterations; ++j) 7: { 8: if (nominator.Elect()) 9: { 10: // elected 11: Console.WriteLine("Elected nominee " + i); 12: Thread.Sleep(random.Next(100, 5000)); 13: nominator.Release(); 14: } 15: else 16: { 17: // not elected 18: Console.WriteLine("Did not elect nominee " + i); 19: } 20: // sleep before check again 21: Thread.Sleep(1000); 22: } 23: }); And would spit out results like: 1: Elected nominee 0 2: Did not elect nominee 2 3: Did not elect nominee 1 4: Did not elect nominee 4 5: Did not elect nominee 3 6: Did not elect nominee 3 7: Did not elect nominee 1 8: Did not elect nominee 2 9: Did not elect nominee 4 10: Elected nominee 3 11: Did not elect nominee 2 12: Did not elect nominee 1 13: Did not elect nominee 4 14: Elected nominee 0 15: Did not elect nominee 2 16: Did not elect nominee 4 17: ... Another nice thing about the Interlocked.Exchange() is it can be used to thread-safely set pretty much anything 64 bits or less in size including references, pointers (in unsafe mode), floats, doubles, etc.  Summary So, now we’ve seen two more things we can do with Interlocked: reading and exchanging a value atomically.  Read() and Exchange() are especially valuable for reading/writing 64 bit values atomically in a 32 bit system.  Exchange() has value even beyond simply atomic writes by using the Exchange() to your advantage, since it reads and set the value atomically, which allows you to do lightweight nomination systems. There’s still a few more goodies in the Interlocked class which we’ll explore next time! Technorati Tags: C#,CSharp,.NET,Little Wonders,Interlocked

    Read the article

  • Introduction to Human Workflow 11g

    - by agiovannetti
    Human Workflow is a component of SOA Suite just like BPEL, Mediator, Business Rules, etc. The Human Workflow component allows you to incorporate human intervention in a business process. You can use Human Workflow to create a business process that requires a manager to approve purchase orders greater than $10,000; or a business process that handles article reviews in which a group of reviewers need to vote/approve an article before it gets published. Human Workflow can handle the task assignment and routing as well as the generation of notifications to the participants. There are three common patterns or usages of Human Workflow: 1) Approval Scenarios: manage documents and other transactional data through approval chains . For example: approve expense report, vacation approval, hiring approval, etc. 2) Reviews by multiple users or groups: group collaboration and review of documents or proposals. For example, processing a sales quote which is subject to review by multiple people. 3) Case Management: workflows around work management or case management. For example, processing a service request. This could be routed to various people who all need to modify the task. It may also incorporate ad hoc routing which is unknown at design time. SOA 11g Human Workflow includes the following features: Assignment and routing of tasks to the correct users or groups. Deadlines, escalations, notifications, and other features required for ensuring the timely performance of a task. Presentation of tasks to end users through a variety of mechanisms, including a Worklist application. Organization, filtering, prioritization and other features required for end users to productively perform their tasks. Reports, reassignments, load balancing and other features required by supervisors and business owners to manage the performance of tasks. Human Workflow Architecture The Human Workflow component is divided into 3 modules: the service interface, the task definition and the client interface module. The Service Interface handles the interaction with BPEL and other components. The Client Interface handles the presentation of task data through clients like the Worklist application, portals and notification channels. The task definition module is in charge of managing the lifecycle of a task. Who should get the task assigned? What should happen next with the task? When must the task be completed? Should the task be escalated?, etc Stages and Participants When you create a Human Task you need to specify how the task is assigned and routed. The first step is to define the stages and participants. A stage is just a logical group. A participant can be a user, a group of users or an application role. The participants indicate the type of assignment and routing that will be performed. Stages can be sequential or in parallel. You can combine them to create any usage you require. See diagram below: Assignment and Routing There are different ways a task can be assigned and routed: Single Approver: task is assigned to a single user, group or role. For example, a vacation request is assigned to a manager. If the manager approves or rejects the request, the employee is notified with the decision. If the task is assigned to a group then once one of managers acts on it, the task is completed. Parallel : task is assigned to a set of people that must work in parallel. This is commonly used for voting. For example, a task gets approved once 50% of the participants approve it. You can also set it up to be a unanimous vote. Serial : participants must work in sequence. The most common scenario for this is management chain escalation. FYI (For Your Information) : task is assigned to participants who can view it, add comments and attachments, but can not modify or complete the task. Task Actions The following is the list of actions that can be performed on a task: Claim : if a task is assigned to a group or multiple users, then the task must be claimed first to be able to act on it. Escalate : if the participant is not able to complete a task, he/she can escalate it. The task is reassigned to his/her manager (up one level in a hierarchy). Pushback : the task is sent back to the previous assignee. Reassign :if the participant is a manager, he/she can delegate a task to his/her reports. Release : if a task is assigned to a group or multiple users, it can be released if the user who claimed the task cannot complete the task. Any of the other assignees can claim and complete the task. Request Information and Submit Information : use when the participant needs to supply more information or to request more information from the task creator or any of the previous assignees. Suspend and Resume :if a task is not relevant, it can be suspended. A suspension is indefinite. It does not expire until Resume is used to resume working on the task. Withdraw : if the creator of a task does not want to continue with it, for example, he wants to cancel a vacation request, he can withdraw the task. The business process determines what happens next. Renew : if a task is about to expire, the participant can renew it. The task expiration date is extended one week. Notifications Human Workflow provides a mechanism for sending notifications to participants to alert them of changes on a task. Notifications can be sent via email, telephone voice message, instant messaging (IM) or short message service (SMS). Notifications can be sent when the task status changes to any of the following: Assigned/renewed/delegated/reassigned/escalated Completed Error Expired Request Info Resume Suspended Added/Updated comments and/or attachments Updated Outcome Withdraw Other Actions (e.g. acquiring a task) Here is an example of an email notification: Worklist Application Oracle BPM Worklist application is the default user interface included in SOA Suite. It allows users to access and act on tasks that have been assigned to them. For example, from the Worklist application, a loan agent can review loan applications or a manager can approve employee vacation requests. Through the Worklist Application users can: Perform authorized actions on tasks, acquire and check out shared tasks, define personal to-do tasks and define subtasks. Filter tasks view based on various criteria. Work with standard work queues, such as high priority tasks, tasks due soon and so on. Work queues allow users to create a custom view to group a subset of tasks in the worklist, for example, high priority tasks, tasks due in 24 hours, expense approval tasks and more. Define custom work queues. Gain proxy access to part of another user's tasks. Define custom vacation rules and delegation rules. Enable group owners to define task dispatching rules for shared tasks. Collect a complete workflow history and audit trail. Use digital signatures for tasks. Run reports like Unattended tasks, Tasks productivity, etc. Here is a screenshoot of what the Worklist Application looks like. On the right hand side you can see the tasks that have been assigned to the user and the task's detail. References Introduction to SOA Suite 11g Human Workflow Webcast Note 1452937.2 Human Workflow Information Center Using the Human Workflow Service Component 11.1.1.6 Human Workflow Samples Human Workflow APIs Java Docs

    Read the article

  • Azure Task Scheduling Options

    - by charlie.mott
    Currently, the Azure PaaS does not offer a distributed\resilient task scheduling service.  If you do want to host a task scheduling product\solution off-premise (and ideally use Azure), what are your options? PaaS Option 1: Worker Roles Use a worker role to schedule and execute actions at specific time periods.  There are a few frameworks available to assist with this: http://azuretoolkit.codeplex.com https://github.com/Lokad/lokad-cloud/wiki/TaskScheduler http://blog.smarx.com/posts/building-a-task-scheduler-in-windows-azure - This addresses a slightly different set of requirements. It’s a more dynamic approach for queuing up tasks, but not repeatable tasks (e.g. daily). I found the Azure Toolkit option the most simple to implement.  Step 1 : Create a domain entity implementing IJob for each job to schedule.  In this sample, I asynchronously call a WCF service method. 1: namespace Acme.WorkerRole.Jobs 2: { 3: using AzureToolkit; 4: using ScheduledTasksService; 5: 6: public class UploadEmployeesJob : IJob 7: { 8: public void Run() 9: { 10: // Call Tasks Service 11: var client = new ScheduledTasksServiceClient("BasicHttpBinding_IScheduledTasksService"); 12: client.UploadEmployees(); 13: client.Close(); 14: } 15: } 16: } Step 2 : In the worker role run method, add the jobs to the toolkit engine. 1: namespace Acme.WorkerRole 2: { 3: using AzureToolkit.Engine; 4: using Jobs; 5:   6: public class WorkerRole : WorkerRoleEntryPoint 7: { 8: public override void Run() 9: { 10: var engine = new CloudEngine(); 11:   12: // Add Scheduled Jobs (using CronJob syntax - see http://www.adminschoice.com/crontab-quick-reference). 13:   14: // 1. Upload Employee job - 8.00 PM every weekday (Mon-Fri) 15: engine.WithJobScheduler().ScheduleJob<UploadEmployeesJob>(c => { c.CronSchedule = "0 20 * * 1-5"; }); 16: // 2. Purge Data job - 10 AM every Saturday 17: engine.WithJobScheduler().ScheduleJob<PurgeDataJob>(c => { c.CronSchedule = "0 10 * * 6"; }); 18: // 3. Process Exceptions job - Every 5 minutes 19: engine.WithJobScheduler().ScheduleJob<ProcessExceptionsJob>(c => { c.CronSchedule = "*/5 * * * *"; }); 20:   21: engine.Run(); 22: base.Run(); 23: } 24: } 25: } Pros Cons Azure Toolkit option is simple to implement. For the AzureToolkit option, you are limited to a single worker role.  Otherwise, the jobs will be executed multiple times, once for each worker role instance.   Paying for a continuously running worker role, even if it just processes a single job once a week.  If you only have a few scheduled tasks to run calling asynchronous services hosted in different web roles, an extra small worker role likely to be sufficient.  However, for an extra small worker role this still costs $14.40/month (03/09/2012). Option 2: Use Scheduled Task on Azure Web Role calling a console app Setup a Windows Scheduled Task on the Azure Web Role. This calls a console application that calls the WCF service methods that run the task actions. This design is described here: http://www.ronaldwidha.net/2011/02/23/cron-job-on-azure-using-scheduled-task-on-a-web-role-to-replace-azure-worker-role-for-background-job/ http://www.voiceoftech.com/swhitley/index.php/2011/07/windows-azure-task-scheduler/ http://devlicio.us/blogs/vinull/archive/2011/10/23/moving-to-azure-worker-roles-for-nothing-and-tasks-for-free.aspx Pros Cons Fairly easy to implement. Supportability - I RDC’ed onto the Azure server and stopped the scheduled task. I then rebooted the machine and the task was re-started. I also tried deleting the task and rebooting, the same thing occurred. The only way to permanently guarantee that a task is disabled is to do a fresh deployment. I think this is a major supportability concern.   Saleability - multiple instances would trigger multiple tasks. You can only have one instance for the scheduled task web role. The guidance implements setup of the scheduled task as part of a web role instance. But if you have more than one instance in a web role, the task will be triggered multiple times for each scheduled action (once per machine). Workaround: If we wanted to use scheduled tasks for another client with a saleable WCF service, then we could include the console & tasks scripts in a separate web role (e.g. a empty WCF service with no real purpose to it). SaaS Option 3: Azure Marketplace I thought that someone might be offering this type of service via the Azure marketplace. At the point of writing this blog post, I did not find anyone doing so. https://datamarket.azure.com/ Pros Cons   Nobody currently offers this on the Azure Marketplace. Option 4: Online Job Scheduling Service Provider There are plenty of online providers that offer this type of service on a pay-as-you-go approach.  Some of these are free for small usage.   Many of these providers are listed here: http://en.wikipedia.org/wiki/Webcron Pros Cons No bespoke development for scheduler. Reliance on third party. IaaS Option 5: Setup Scheduling Software on Azure IaaS VM’s One of job scheduling software offerings could be installed and configured on Azure VM’s.  A list of software options is listed here: http://en.wikipedia.org/wiki/List_of_job_scheduler_software Pros Cons Enterprise distributed\resilient task scheduling service VM Setup and maintenance   Software Licence Costs Option 6: VM Gallery A the time of writing this blog post, I did not spot a VM in the gallery that included pre-installation of any of the above software options. Pros Cons   No current VM template. Summary For my current project that had a small handful of tasks to schedule with a limited project budget I chose option 1 (a worker role using the Azure Toolkit to schedule tasks).  If I was building an enterprise scale solution for the future, options 4 and 5 are currently worthy of consideration. Hopefully, Microsoft will include tasks scheduling in the future as part of their PaaS offerings.

    Read the article

  • Clouds Everywhere But not a Drop of Rain – Part 3

    - by sxkumar
    I was sharing with you how a broad-based transformation such as cloud will increase agility and efficiency of an organization if process re-engineering is part of the plan.  I have also stressed on the key enterprise requirements such as “broad and deep solutions, “running your mission critical applications” and “automated and integrated set of capabilities”. Let me walk you through some key cloud attributes such as “elasticity” and “self-service” and what they mean for an enterprise class cloud. I will also talk about how we at Oracle have taken a very enterprise centric view to developing cloud solutions and how our products have been specifically engineered to address enterprise cloud needs. Cloud Elasticity and Enterprise Applications Requirements Easy and quick scalability for a short-period of time is the signature of cloud based solutions. It is this elasticity that allows you to dynamically redistribute your resources according to business priorities, helps increase your overall resource utilization, and reduces operational costs by allowing you to get the most out of your existing investment. Most public clouds are offering a instant provisioning mechanism of compute power (CPU, RAM, Disk), customer pay for the instance-hours(and bandwidth) they use, adding computing resources at peak times and removing them when they are no longer needed. This type of “just-in-time” serving of compute resources is well known for mid-tiers “state less” servers such as web application servers and web servers that just need another machine to start and run on it but what does it really mean for an enterprise application and its underlying data? Most enterprise applications are not as quite as “state less” and justifiably so. As such, how do you take advantage of cloud elasticity and make it relevant for your enterprise apps? This is where Cloud meets Grid Computing. At Oracle, we have invested enormous amount of time, energy and resources in creating enterprise grid solutions. All our technology products offer built-in elasticity via clustering and dynamic scaling. With products like Real Application Clusters (RAC), Automatic Storage Management, WebLogic Clustering, and Coherence In-Memory Grid, we allow all your enterprise applications to benefit from Cloud elasticity –both vertically and horizontally - without requiring any application changes. A number of technology vendors take a rather simplistic route of starting up additional or removing unneeded VM as the "Cloud Scale-Out" solution. While this may work for stateless mid-tier servers where load balancers can handle the addition and remove of instances transparently but following a similar approach for the database tier - often called as "database sharding" - requires significant application modification and typically does not work with off the shelf packaged applications. Technologies like Oracle Database Real Application Clusters, Automatic Storage Management, etc. on the other hand bring the benefits of incremental scalability and on-demand elasticity to ANY application by providing a simplified abstraction layers where the application does not need deal with data spread over multiple database instances. Rather they just talk to a single database and the database software takes care of aggregating resources across multiple hardware components. It is the technologies like these that truly make a cloud solution relevant for enterprises.  For customers who are looking for a next generation hardware consolidation platform, our engineered systems (e.g. Exadata, Exalogic) not only provide incredible amount of performance and capacity, they also reduce the data center complexity and simplify operations. Assemble, Deploy and Manage Enterprise Applications for Cloud Products like Oracle Virtual assembly builder (OVAB) resolve the complex problem of bringing the cloud speed to complex multi-tier applications. With assemblies, you can not only provision all components of a multi-tier application and wire them together by push of a button, other aspects of application lifecycle, such as real-time application testing, scale-up/scale-down, performance and availability monitoring, etc., are also automated using Oracle Enterprise Manager.  An essential criteria for an enterprise cloud to succeed is the ability to ensure business service levels especially when business users have either full visibility on the usage cost with a “show back” or a “charge back”. With Oracle Enterprise Manager 12c, we have created the most comprehensive cloud management solution in the industry that is capable of managing business service levels “applications-to-disk” in a enterprise private cloud – all from a single console. It is the only cloud management platform in the industry that allows you to deliver infrastructure, platform and application cloud services out of the box. Moreover, it offers integrated and complete lifecycle management of the cloud - including planning and set up, service delivery, operations management, metering and chargeback, etc .  Sounds unbelievable? Well, just watch this space for more details on how Oracle Enterprise Manager 12c is the nerve center of Oracle Cloud! Our cloud solution portfolio is also the broadest and most deep in the industry  - covering public, private, hybrid, Infrastructure, platform and applications clouds. It is no coincidence therefore that the Oracle Cloud today offers the most comprehensive set of public cloud services in the industry.  And to a large part, this has been made possible thanks to our years on investment in creating cloud enabling technologies.  Summary  But the intent of this blog post isn't to dwell on how great our solutions are (these are just some examples to illustrate how we at Oracle have approached this problem space). Rather it is to help you ask the right questions before you embark on your cloud journey.  So to summarize, here are the key takeaways.       It is critical that you are clear on why you are building the cloud. Successful organizations keep business benefits as the first and foremost cloud objective. On the other hand, those who approach this purely as a technology project are more likely to fail. Think about where you want to be in 3-5 years before you get started. Your long terms objectives should determine what your first step ought to be. As obvious as it may seem, more people than not make the first move without knowing where they are headed.  Don’t make the mistake of equating cloud to virtualization and Infrastructure-as-a-Service (IaaS). Spinning a VM on-demand will give some short term relief to your IT staff but is unlikely to solve your larger business problems. As such, even if IaaS is your first step towards a more comprehensive cloud, plan the roadmap around those higher level services before you begin. And ask your vendors on how they are going to be your partners in this journey. Capabilities like self-service access and chargeback/showback are absolutely critical if you really expect your cloud to be transformational. Your business won't see the full benefits of the cloud until it empowers them with same kind of control and transparency that they are used to while using a public cloud service.  Evaluate the benefits of integration, as opposed to blindly following the best-of-breed strategy. Integration is a huge challenge and more so in a cloud environment. There are enormous costs associated with stitching a solution out of disparate components and even more in maintaining it. Hope you found these ideas helpful. Looking forward to hearing your thoughts and experiences.

    Read the article

< Previous Page | 289 290 291 292 293 294 295 296 297 298 299 300  | Next Page >