Search Results

Search found 19055 results on 763 pages for 'high performance'.

Page 597/763 | < Previous Page | 593 594 595 596 597 598 599 600 601 602 603 604  | Next Page >

  • Sixeyed.Caching available now on NuGet and GitHub!

    - by Elton Stoneman
    Originally posted on: http://geekswithblogs.net/EltonStoneman/archive/2013/10/22/sixeyed.caching-available-now-on-nuget-and-github.aspxThe good guys at Pluralsight have okayed me to publish my caching framework (as seen in Caching in the .NET Stack: Inside-Out) as an open-source library, and it’s out now. You can get it here: Sixeyed.Caching source code on GitHub, and here: Sixeyed.Caching package v1.0.0 on NuGet. If you haven’t seen the course, there’s a preview here on YouTube: In-Process and Out-of-Process Caches, which gives a good flavour. The library is a wrapper around various cache providers, including the .NET MemoryCache, AppFabric cache, and  memcached*. All the wrappers inherit from a base class which gives you a set of common functionality against all the cache implementations: •    inherits OutputCacheProvider, so you can use your chosen cache provider as an ASP.NET output cache; •    serialization and encryption, so you can configure whether you want your cache items serialized (XML, JSON or binary) and encrypted; •    instrumentation, you can optionally use performance counters to monitor cache attempts and hits, at a low level. The framework wraps up different caches into an ICache interface, and it lets you use a provider directly like this: Cache.Memory.Get<RefData>(refDataKey); - or with configuration to use the default cache provider: Cache.Default.Get<RefData>(refDataKey); The library uses Unity’s interception framework to implement AOP caching, which you can use by flagging methods with the [Cache] attribute: [Cache] public RefData GetItem(string refDataKey) - and you can be more specific on the required cache behaviour: [Cache(CacheType=CacheType.Memory, Days=1] public RefData GetItem(string refDataKey) - or really specific: [Cache(CacheType=CacheType.Disk, SerializationFormat=SerializationFormat.Json, Hours=2, Minutes=59)] public RefData GetItem(string refDataKey) Provided you get instances of classes with cacheable methods from the container, the attributed method results will be cached, and repeated calls will be fetched from the cache. You can also set a bunch of cache defaults in application config, like whether to use encryption and instrumentation, and whether the cache system is enabled at all: <sixeyed.caching enabled="true"> <performanceCounters instrumentCacheTotalCounts="true" instrumentCacheTargetCounts="true" categoryNamePrefix ="Sixeyed.Caching.Tests"/> <encryption enabled="true" key="1234567890abcdef1234567890abcdef" iv="1234567890abcdef"/> <!-- key must be 32 characters, IV must be 16 characters--> </sixeyed.caching> For AOP and methods flagged with the cache attribute, you can override the compile-time cache settings at runtime with more config (keyed by the class and method name): <sixeyed.caching enabled="true"> <targets> <target keyPrefix="MethodLevelCachingStub.GetRandomIntCacheConfiguredInternal" enabled="false"/> <target keyPrefix="MethodLevelCachingStub.GetRandomIntCacheExpiresConfiguredInternal" seconds="1"/> </targets> It’s released under the MIT license, so you can use it freely in your own apps and modify as required. I’ll be adding more content to the GitHub wiki, which will be the main source of documentation, but for now there’s an FAQ to get you started. * - in the course the framework library also wraps NCache Express, but there's no public redistributable library that I can find, so it's not in Sixeyed.Caching.

    Read the article

  • Internet Explorer 9 RC Now Available: Here’s the Most Interesting New Stuff

    - by The Geek
    Yesterday Microsoft announced the release candidate of Internet Explorer 9, which is very close to the final product. Here’s a screenshot tour of the most interesting new stuff, as well as answers to your questions. The most important question is should you install this version? And the answer is absolutely yes. Even if you don’t use IE, it’s better to have a newer, more secure version on your PC. What’s New Under the Hood in Release Candidate vs Beta? If you want to see the full list of changes with all the original marketing detail, you can read Microsoft’s Beauty of the Web page, but here’s the highlights that you might be interested in. Improved Performance – they’ve made a lot of changes, and it really feels faster, especially when using more intensive web apps like Gmail. Power Consumption Settings – since the JavaScript engine in any browser uses a lot of CPU power, they’ve now integrated it into the power settings, so if you’re on battery it will use less CPU, and save battery life. This is really a great change. UI Changes – The tab bar can now be moved below the address bar (see below for more), they’ve shaved some pixels off the design to save space, and now you can toggle the Menu bar to be always on. Pinned Sites – now you can pin multiple pages to a single taskbar button. Very useful if you always use a couple web apps together. You can also pin a site in InPrivate mode. FlashBlock and AdBlock are Integrated (sorta) – there’s a new ActiveX filtering that lets you enable plug-ins only for sites you trust. There’s also a tracking protection list that can block certain content (which can obviously be used to block ads). Geolocation – while a lot of privacy conscious people might complain about this, if you use your laptop while traveling, it’s really useful to have geo-located features when using Google Maps, etc. Don’t worry, it won’t leak your privacy by default. WebM Video – Yeah, Google recently removed H.264 from Chrome, but Microsoft has added Google’s WebM video format to Internet Explorer. Keep reading for more about using the new features Latest Features How-To Geek ETC Internet Explorer 9 RC Now Available: Here’s the Most Interesting New Stuff Here’s a Super Simple Trick to Defeating Fake Anti-Virus Malware How to Change the Default Application for Android Tasks Stop Believing TV’s Lies: The Real Truth About "Enhancing" Images The How-To Geek Valentine’s Day Gift Guide Inspire Geek Love with These Hilarious Geek Valentines The 50 Faces of Mario Death [Infographic] Clean Up Google Calendar’s Interface in Chrome and Iron The Rise and Fall of Kramerica? [Seinfeld Video] GNOME Shell 3 Live CDs for OpenSUSE and Fedora Available for Testing Picplz Offers Special FX, Sharing, and Backup of Your Smartphone Pics BUILD! An Epic LEGO Stop Motion Film [VIDEO]

    Read the article

  • Oracle collaborates with leading IT vendors on Cloud Management Standards

    - by Anand Akela
    During the last couple of days, two key specifications for cloud management standards have been announced. Oracle collaborated with leading technology vendors from the IT industry on both of these cloud management specifications. One of the specifications focuses "Infrastructure as a Service" ( IaaS )  cloud service model , while the other specification announced today focuses on "Platform as a Service" ( PaaS ) cloud service model. Please see The NIST Definition of Cloud Computing to learn more about IaaS and PaaS . Earlier today Oracle , CloudBees, Cloudsoft, Huawei, Rackspace, Red Hat, and Software AG   announced the Cloud Application Management for Platforms (CAMP) specification that will be submitted to Organization for the Advancement of Structured Information Standards (OASIS) for development of an industry standard, in an effort to help ensure interoperability for deploying and managing applications across cloud environments.  Typical PaaS architecture - Source : CAMP specification The CAMP specification defines the artifacts and APIs that need to be offered by a PaaS cloud to manage the building, running, administration, monitoring and patching of applications in the cloud. Its purpose is to enable interoperability among self-service interfaces to PaaS clouds by defining artifacts and formats that can be used with any conforming cloud and enable independent vendors to create tools and services that interact with any conforming cloud using the defined interfaces. Cloud vendors can use these interfaces to develop new PaaS offerings that will interact with independently developed tools and components. In a separate cloud standards announcement yesterday, the Distributed Management Task Force ( DMTF ), the organization bringing the IT industry together to collaborate on systems management standards development, validation, promotion and adoption, released the new Cloud Infrastructure Management Interface (CIMI) specification. Oracle collaborated with various technology vendors and industry organizations on this specification. CIMI standardizes interactions between cloud environments to achieve interoperable cloud infrastructure management between service providers and their consumers and developers, enabling users to manage their cloud infrastructure use easily and without complexity. DMTF developed CIMI as a self-service interface for infrastructure clouds ( IaaS focus ) , allowing users to dynamically provision, configure and administer their cloud usage with a high-level interface that greatly simplifies cloud systems management. Mark Carlson, Principal Cloud Strategist at Oracle provides more details about CAMP  and CIMI his blog . Stay Connected: Twitter |  Face book |  You Tube |  Linked in |  Newsletter

    Read the article

  • Silverlight Cream for March 10, 2010 - 2 -- #811

    - by Dave Campbell
    In this Issue: AfricanGeek, Phil Middlemiss, Damon Payne, David Anson, Jesse Liberty, Jeremy Likness, Jobi Joy(-2-), Fredrik Normén, Bobby Diaz, and Mike Taulty(-2-). Shoutouts: Shawn Wildermuth blogged that they posted My "What's New in Silverlight 3" Video from 0reDev Last Fall Shawn Wildermuth also has a post up for his loyal followers: Where to See Me At MIX10 Jonas Follesø has presentation materials up as well: MVVM presentation from NDC2009 on Vimeo Adam Kinney updated his Favorite Tool and Library Downloads for Silverlight From SilverlightCream.com: Styling Silverlight ListBox with Blend 3 In his latest Video Tutorial, AfricanGeek is animating the ListBox control by way of Expression Blend 3. Animating the Silverlight Opacity Mask Phil Middlemiss has written a Behavior that lets you turn a FrameworkElement into an opacity mask for it's parent container... check out his tutorial and grab the code. AddRange for ObservableCollection in Silverlight 3 Damon Payne has a post up discussing the problem with large amounts of data in an ObservableCollection, and how using AddRange is a performance booster. Easily rotate the axis labels of a Silverlight/WPF Toolkit chart David Anson blogged a solution to rotating the axis labels of a Silverlight and WPF chart. Persisting the Configuration (Updated) Jesse Liberty has a good discussion on the continuation of his HyperVideo Platform talking about what all he is needing from the database in the form of configuration information... including the relationships. Animations and View Models: IAnimationDelegate Check out Jeremy Likness' IAnimationDelegate that lets your ViewModel fire and respond to animations without having to know all about them. Button Style - Silverlight Jobi Joy converted a WPF control template into Silverlight... and you'll want to download the XAML he's got for this :) A Simple Accordion banner using ListBox Jobi Joy also has an Image Accordian created in Expression Blend... and it's a 'drop this XAML in your User Control' kinda thing... again, go grab the XAML :) WCF RIA Services Silverlight Business Application – Using ASP.NET SiteMap for Navigation Fredrik Normén has a code-laden post up on RIA Services and the ASP.NET SiteMap. He is using the Silverlight Business app template that comes with WCF RIA Services. A Simple, Selectable Silverlight TextBlock (sort of)... Bobby Diaz shares with us his solution for a Text control that can be copied from in the same manner 'normal' web controls can be. He also includes a link to another post on the same topic. Silverlight 4 Beta Networking. Part 11 - WCF and TCP Mike Taulty has another pair of video tutorials up in his Networking series. This one is on WCF over TCP Silverlight 4 Beta Networking. Part 12 - WCF and Polling HTTP Mike Taulty's 12th networking video tutorial is on WCF with HTTP polling duplex. Stay in the 'Light! Twitter SilverlightNews | Twitter WynApse | WynApse.com | Tagged Posts | SilverlightCream Join me @ SilverlightCream | Phoenix Silverlight User Group Technorati Tags: Silverlight    Silverlight 3    Silverlight 4    MIX10

    Read the article

  • Gnome Shell segfault in libglib-2.0

    - by slohui
    I have been using Ubuntu 11.10 + Gnome Shell with a Nvidia card, but now I've moved it to my new PC which has an ATI card, at first it wasn't booting but I installed the driver from amd.com and then it worked. Anyway my problem is that gnome-shell is crashing, mostly when I try to start a VirtualBox machine (it happened in other times but I don't remember what I was doing). Sometimes gnome-shell respawns and it continue working but sometimes it doesn't so I have to restart lightdm and lose all the windows I was using. Here's some of the syslog when the crash occurs: Apr 9 12:20:08 desktop-1 NetworkManager[1032]: SCPlugin-Ifupdown: devices added (path: /sys/devices/virtual/net/vboxnet0, iface: vboxnet0) Apr 9 12:20:08 desktop-1 NetworkManager[1032]: SCPlugin-Ifupdown: device added (path: /sys/devices/virtual/net/vboxnet0, iface: vboxnet0): no ifupdown configuration found. Apr 9 12:20:08 desktop-1 NetworkManager[1032]: <warn> /sys/devices/virtual/net/vboxnet0: couldn't determine device driver; ignoring... Apr 9 12:20:08 desktop-1 kernel: [ 4498.689561] warning: `VirtualBox' uses 32-bit capabilities (legacy support in use) Apr 9 12:24:29 desktop-1 gnome-session[1617]: WARNING: Application 'gnome-shell.desktop' killed by signal Apr 9 12:24:45 desktop-1 gnome-session[1617]: WARNING: App 'gnome-shell.desktop' respawning too quickly Apr 9 12:24:45 desktop-1 gnome-session[1617]: CRITICAL: We failed, but the fail whale is dead. Sorry.... Apr 9 12:25:20 desktop-1 kernel: [ 4810.769775] show_signal_msg: 30 callbacks suppressed |----- > Apr 9 12:25:20 desktop-1 kernel: [ 4810.769785] gnome-shell[3427]: segfault at b0 ip b6bd09cd sp bfc9b650 error 4 in libglib-2.0.so.0.3000.0[b6b71000+f7000]** Apr 9 12:25:20 desktop-1 gnome-session[1617]: WARNING: Application 'gnome-shell.desktop' killed by signal Apr 9 12:25:23 desktop-1 kernel: [ 4814.055705] EXT4-fs (sda1): Unaligned AIO/DIO on inode 133295 by VirtualBox; performance will be poor. Apr 9 12:26:55 desktop-1 gnome-session[1617]: Gdk-WARNING: gnome-session: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.#012 Apr 9 12:26:55 desktop-1 kernel: [ 4905.373256] [fglrx] IRQ 56 Disabled Apr 9 12:26:59 desktop-1 acpid: client 1124[0:0] has disconnected Apr 9 12:26:59 desktop-1 acpid: client connected from 3864[0:0] Apr 9 12:26:59 desktop-1 acpid: 1 client rule loaded Apr 9 12:26:59 desktop-1 kernel: [ 4909.700095] fglrx_pci 0000:02:00.0: irq 56 for MSI/MSI-X Apr 9 12:26:59 desktop-1 kernel: [ 4909.701466] [fglrx] Firegl kernel thread PID: 3867 Apr 9 12:26:59 desktop-1 kernel: [ 4909.701625] [fglrx] Firegl kernel thread PID: 3868 Apr 9 12:26:59 desktop-1 kernel: [ 4909.701852] [fglrx] Firegl kernel thread PID: 3869 Apr 9 12:26:59 desktop-1 kernel: [ 4909.702021] [fglrx] IRQ 56 Enabled Apr 9 12:26:59 desktop-1 kernel: [ 4909.861815] [fglrx] Gart USWC size:1280 M. Apr 9 12:26:59 desktop-1 kernel: [ 4909.861817] [fglrx] Gart cacheable size:508 M. Apr 9 12:26:59 desktop-1 kernel: [ 4909.861820] [fglrx] Reserved FB block: Shared offset:0, size:1000000 Apr 9 12:26:59 desktop-1 kernel: [ 4909.861821] [fglrx] Reserved FB block: Unshared offset:f8fd000, size:403000 Apr 9 12:26:59 desktop-1 kernel: [ 4909.861823] [fglrx] Reserved FB block: Unshared offset:3fff4000, size:c000 Does anyone could guide me on how to fix this? Or the proper place where to ask for help.

    Read the article

  • SQL SERVER – ColumnStore Index – Batch Mode vs Row Mode

    - by pinaldave
    What do you do when you are in a hurry and hear someone say things which you do not agree or is wrong? Well, let me tell you what I do or what I recently did. I was walking by and heard someone mentioning “Columnstore Index are really great as they are using Batch Mode which makes them seriously fast.” While I was passing by and I heard this statement my first reaction was I thought Columnstore Index can use both – Batch Mode and Row Mode. I stopped by even though I was in a hurry and asked the person if he meant that Columnstore indexes are seriously fast because they use Batch Mode all the time or Batch Mode is one of the reasons for Columnstore Index to be faster. He responded that Columnstore Indexes can run only in Batch Mode. However, I do not like to confront anybody without hearing their complete story. Honestly, I like to do information sharing and avoid confronting as much as possible. There are always ways to communicate the same positively. Well, this is what I did, I quickly pull up my earlier article on Columnstore Index and copied the script to SQL Server Management Studio. I created two versions of the script. 1) Very Large Table 2) Reasonably Small Table. I a query which uses columnstore index on both of the versions. I found very interesting result of the my tests. I saved my tests and sent it to the person who mentioned about that Columnstore Indexes are using Batch Mode only. He immediately acknowledged that indeed he was incorrect in saying that Columnstore Index uses only Batch Mode. What really caught my attention is that he also thanked me for sending him detail email instead of just having argument where he and I both were standing in the corridor and neither have no way to prove any theory. Here is the screenshots of the both the scenarios. 1) Columnstore Index using Batch Mode 2) Columnstore Index using Row Mode Here is the logic behind when Columnstore Index uses Batch Mode and when it uses Row Mode. A batch typically represents about 1000 rows of data. Batch mode processing also uses algorithms that are optimized for the multicore CPUs and increased memory throughput.  Batch mode processing spreads metadata access costs and overhead over all the rows in a batch.  Batch mode processing operates on compressed data when possible leading superior performance. Here is one last point – Columnstore Index can use Batch Mode or Row Mode but Batch Mode processing is only available in Columnstore Index. I hope this statement truly sums up the whole concept. Reference: Pinal Dave (http://blog.sqlauthority.com) Filed under: PostADay, SQL, SQL Authority, SQL Index, SQL Query, SQL Server, SQL Tips and Tricks, T SQL, Technology

    Read the article

  • Large File Upload in SharePoint 2010

    - by Sahil Malik
    Ad:: SharePoint 2007 Training in .NET 3.5 technologies (more information). Okay this is a big BIG B-I-G problem. And with SP2010 it’s going to be more prominent, because atleast at the server side, SharePoint can support large files much much better than SharePoint 2007 ever did. The issue with very large files being uploaded through any browser based API are - Reliably transferring gigabyte or bigger files without breakages over a protocol like HTTP, which is better suited for tiny transfers like images and text. Not killing your browser because it has to load all that in memory Not killing your web server because All that you upload through HTTP post, first gets streamed into IIS Memory, w3wp.exe memory before the ENTIRE FILE finishes uploading .. before it is stored. Which means, You cannot show an accurate and live progress bar of the upload, IIS gives you no such accurate metric of an upload. All the counters it gives you are approximate. Your w3wp.exe eats up all server memory – 4GB of it, for a 4GB upload. A thread is kept busy for the entire duration of the upload, thereby greatly limiting your web server’s capability to serve newer requests. Kills effective load balancing. Not killing your content database because, As you are uploading a very large file, that large file gets written sequentially into the DB, and therefore for a very large file very severely impacts the database performance. I had put together another video showing RBS usage in SharePoint 2010. I talked about many practical ramifications of using RBS in SharePoint in that video. Note that enabling large file support will never ever be a point and click job, simply because there are too many questions one needs to ask, and too many things one needs to plan for. However, one part that will remain common across all large file upload scenarios, in SharePoint or outside of SharePoint is to do it efficiently while not killing the web server. In this video, I describe using the Telerik Silverlight Upload control with SharePoint 2010 to enable efficient large file uploads in SharePoint. Presenting .. The video Comment on the article ....

    Read the article

  • Java Spotlight Episode 139: Mark Heckler and José Pereda on JES based Energy Monitoring @MkHeck @JPeredaDnr

    - by Roger Brinkley
    Interview with Mark Heckler and José Pereda on using JavaSE Embedded with the Java Embedded Suite on a RaspberryPI along with a JavaFX client to monitor an energy production system and their JavaOne Tutorial- Java Embedded EXTREME MASHUPS: Building self-powering sensor nets for the IoT Right-click or Control-click to download this MP3 file. You can also subscribe to the Java Spotlight Podcast Feed to get the latest podcast automatically. If you use iTunes you can open iTunes and subscribe with this link: Java Spotlight Podcast in iTunes. Show Notes News Java Virtual Developer Day Session Videos Available JavaFX Maven Plugin 2.0 Released JavaFX Scene Builder 1.1 build b28 FXForm 2 release 0.2.2 OpenJDK8/Zero cross compile build for Foundation model HSAIL-based GPU offload: the Quest for Java Performance Begins Progress on Moving to Gradle Java EE 7 Launch Keynote Replay Java EE 7 Technical Breakouts Replay Java EE 7 support in NetBeans 7.3.1 Java EE 7 support in Eclipse 4.3 Java Magazine - May/June Events Jul 16-19, Uberconf, Denver, USA Jul 22-24, JavaOne Shanghai, China Jul 29-31, JVM Language Summit, Santa Clara Sep 11-12, JavaZone, Oslo, Norway Sep 19-20, Strange Loop, St. Louis Sep 22-26 JavaOne San Francisco 2013, USA Feature Interview Mark Heckler is an Oracle Corporation Java/Middleware/Core Tech Engineer with development experience in numerous environments. He has worked for and with key players in the manufacturing, emerging markets, retail, medical, telecom, and financial industries to develop and deliver critical capabilities on time and on budget. Currently, he works primarily with large government customers using Java throughout the stack and across the enterprise. He also participates in open-source development at every opportunity, being a JFXtras project committer and developer of DialogFX, MonologFX, and various other projects. When Mark isn't working with Java, he enjoys writing about his experiences at the Java Jungle website (https://blogs.oracle.com/javajungle/) and on Twitter (@MkHeck). José Pereda is a Structural Engineer working in the School of Engineers in the University of Valladolid in Spain for more than 15 years, and his passion is related to applying programming to solve real problems. Being involved with Java since 1999, José shares his time between JavaFX and the Embedded world, developing commercial applications and open source projects (https://github.com/jperedadnr), and blogging (http://jperedadnr.blogspot.com.es/) or tweeting (@JPeredaDnr) of both. What’s Cool AquaFX 0.1 - Mac OS X skin for JavaFX by Claudine Zillmann DromblerFX adds a docking framework Part 2 of Gerrit’s taming the Nashorn for writing JavaFX apps in Javascript Tool from mihosoft called JSelect for quickly switching JDKs Apache Maven Javadoc Plugin 2.9.1 Released Proposal: Java Concurrency Stress tests (jcstress) Slide-free Code-driven session at SV JUG JavaOne approvals/rejects gone out

    Read the article

  • Need to Know

    - by Tony Davis
    Sometimes, I wonder whether writers of documentation, tutorials and articles stop to ask themselves one very important question: Does the reader really need to know this? I recently took on the task of writing a concise series of articles about the transaction log, what is it, how it works and why it's important. It was an enjoyable task; rather like peering inside a giant, complex clock mechanism. Initially, one sees only the basic components, which work to guarantee the integrity of database transactions, and preserve these transactions so that data can be restored to a previous point in time. On closer inspection, one notices all of small, arcane mechanisms that are necessary to make this happen; LSNs, virtual log files, log chains, database checkpoints, and so on. It was engrossing, escapist, stuff; what I'd written looked weighty and steeped in mysterious significance. Suddenly, however, I jolted myself back to reality with the awful thought "does anyone really need to know all this?" The driver of a car needs only to be dimly aware of what goes on under the hood, however exciting the mechanism is to the engineer. Similarly, while everyone who uses SQL Server ought to be aware of the transaction log, its role in guaranteeing the ACID properties, and how to control its growth, the intricate mechanisms ticking away under its clock face are a world away from the daily work of the harassed developer. The DBA needs to know more, such as the correct rituals for ensuring optimal performance and data integrity, setting the appropriate growth characteristics, backup routines, restore procedures, and so on. However, even then, the average DBA only needs to understand enough about the arcane processes to spot problems and react appropriately, or to know how to Google for the best way of dealing with it. The art of technical writing is tied up in intimate knowledge of your audience and what they need to know at any point. It means serving up just enough at each point to help the reader in a practical way, but not to overcook it, or stuff the reader with information that does them no good. When I think of the books and articles that have helped me the most, they have been full of brief, practical, and well-informed guidance, based on experience. This seems far-removed from the 900-page "beginner's guides" that one now sees everywhere. The more I write and edit, the more I become convinced that the real art of technical communication lies in knowing what to leave out. In what areas do the SQL Server technical materials suffer from "information overload"? Where else does it seem that concise, practical advice is drowned out by endless discussion of the "clock mechanisms"? Cheers, Tony.

    Read the article

  • HPC Server Dynamic Job Scheduling: when jobs spawn jobs

    - by JoshReuben
    HPC Job Types HPC has 3 types of jobs http://technet.microsoft.com/en-us/library/cc972750(v=ws.10).aspx · Task Flow – vanilla sequence · Parametric Sweep – concurrently run multiple instances of the same program, each with a different work unit input · MPI – message passing between master & slave tasks But when you try go outside the box – job tasks that spawn jobs, blocking the parent task – you run the risk of resource starvation, deadlocks, and recursive, non-converging or exponential blow-up. The solution to this is to write some performance monitoring and job scheduling code. You can do this in 2 ways: manually control scheduling - allocate/ de-allocate resources, change job priorities, pause & resume tasks , restrict long running tasks to specific compute clusters Semi-automatically - set threshold params for scheduling. How – Control Job Scheduling In order to manage the tasks and resources that are associated with a job, you will need to access the ISchedulerJob interface - http://msdn.microsoft.com/en-us/library/microsoft.hpc.scheduler.ischedulerjob_members(v=vs.85).aspx This really allows you to control how a job is run – you can access & tweak the following features: max / min resource values whether job resources can grow / shrink, and whether jobs can be pre-empted, whether the job is exclusive per node the creator process id & the job pool timestamp of job creation & completion job priority, hold time & run time limit Re-queue count Job progress Max/ min Number of cores, nodes, sockets, RAM Dynamic task list – can add / cancel jobs on the fly Job counters When – poll perf counters Tweaking the job scheduler should be done on the basis of resource utilization according to PerfMon counters – HPC exposes 2 Perf objects: Compute Clusters, Compute Nodes http://technet.microsoft.com/en-us/library/cc720058(v=ws.10).aspx You can monitor running jobs according to dynamic thresholds – use your own discretion: Percentage processor time Number of running jobs Number of running tasks Total number of processors Number of processors in use Number of processors idle Number of serial tasks Number of parallel tasks Design Your algorithms correctly Finally , don’t assume you have unlimited compute resources in your cluster – design your algorithms with the following factors in mind: · Branching factor - http://en.wikipedia.org/wiki/Branching_factor - dynamically optimize the number of children per node · cutoffs to prevent explosions - http://en.wikipedia.org/wiki/Limit_of_a_sequence - not all functions converge after n attempts. You also need a threshold of good enough, diminishing returns · heuristic shortcuts - http://en.wikipedia.org/wiki/Heuristic - sometimes an exhaustive search is impractical and short cuts are suitable · Pruning http://en.wikipedia.org/wiki/Pruning_(algorithm) – remove / de-prioritize unnecessary tree branches · avoid local minima / maxima - http://en.wikipedia.org/wiki/Local_minima - sometimes an algorithm cant converge because it gets stuck in a local saddle – try simulated annealing, hill climbing or genetic algorithms to get out of these ruts   watch out for rounding errors – http://en.wikipedia.org/wiki/Round-off_error - multiple iterations can in parallel can quickly amplify & blow up your algo ! Use an epsilon, avoid floating point errors,  truncations, approximations Happy Coding !

    Read the article

  • JRockit R28/JRockit Mission Control 4.0 is out!

    - by Marcus Hirt
    The next major release of JRockit is finally out! Here are some highlights: Includes the all new JRockit Flight Recorder – supersedes the old JRockit Runtime Analyser. The new flight recorder is inspired by the “black box” in airplanes. It uses a highly efficient recording engine and thread local buffers to capture data about the runtime and the application running in the JVM. It can be configured to always be on, so that whenever anything “interesting” happens, data can be dumped for some time back. Think of it as your own personal profiling time machine. Automatic shortest path calculation in Memleak – no longer any need for running around in circles when trying to find your way back to a thread root from an instance. Memleak can now show class loader related information and split graphs on a per class loader basis. More easily configured JMX agent – default port for both RMI Registry and RMI Server can be configured, and is by default the same, allowing easier configuration of firewalls. Up to 64 GB (was 4GB) compressed references. Per thread allocation profiling in the Management Console. Native Memory Tracking – it is now possible to track native memory allocations with very high resolution. The information can either be accessed using JRCMD, or the dedicated Native Memory Tracking experimental plug-in for the Management Console (alas only available for the upcoming 4.0.1 release). JRockit can now produce heap dumps in HPROF format. Cooperative suspension – JRockit is no longer using system signals for stopping threads, which could lead to hangs if signals were lost or blocked (for example bad NFS shares). Now threads check periodically to see if they are suspended. VPAT/Section 508 compliant JRMC – greatly improved keyboard navigation and screen reader support. See New and Noteworthy for more information. JRockit Mission Control 4.0.0 can be downloaded from here: http://www.oracle.com/technology/software/products/jrockit/index.html <shameless ad> There is even a book to go with JRMC 4.0.0/JRockit R28! http://www.packtpub.com/oracle-jrockit-the-definitive-guide/book/ </shameless ad>

    Read the article

  • Announcing Oracle Enterprise Content Management Suite 11g

    - by [email protected]
    Today Oracle announced Oracle Enterprise Content Management Suite 11g. This is a major release for us, and reinforces our three key themes at Oracle: Complete New in this release - Oracle ECM Suite 11g is built on a single, unified repository. Every piece of content - documents, HTML pages, digital assets, scanned images - is stored and accessbile directly from the repository, whether you are working on websites, creating brand logos, processing accounts payable invoices, or running records and retention functions. It makes complete, end-to-end management of content possible, from the point it enters the organization, through its entire lifecycle. Also new in this release, the installation, access, monitoring and administration of Oracle ECM Suite 11g is centralized. As a complete system, organizations can lower the costs of training and usage by having a centralized source of information that is easily administered. As part of this new unified repository release, Oracle has released a benchmarking white paper that shows the extreme performance and scalability of Oracle ECM Suite. When tested on a two node UCM Server running on Sun Oracle DB Machine Half Rack Hardware with an Exadata storage server, Oracle ECM Suite 11g is able to ingest over 178 million documents per day. Open Oracle ECM Suite 11g is built on a service-oriented architecture. All functions are available through standards-based services calls in Web Services or Java. In this release Oracle unveils Open Web Content Management. Open Web Content Management is a revolutionary approach to web content management that decouples the content management process from the process of creating web applications. One piece of this approach is our one-click web content management. With one click, a web application builder can drag content services into their application, enabling their users to also edit content with just one click. Open Web Content Management is also open because it enables Web developers to add Web content management to new and existing JavaServer Pages (JSP), JavaServer Faces (JSF) and Oracle Application Development Framework (ADF) Faces applications Open content distribution - Oracle ECM Suite 11g offers flexible deployment options with a built-in smart cache so organizations can deliver Web sites or Web applications without requiring Oracle ECM Suite as part of the delivery system Integrated Oracle ECM Suite 11g also offers a series of next generation desktop integrations, providing integrations such as: New MS Office integration with menus to access managed content, insert managed links, and compare managed documents using standard MS Office reviewing tools Automatic identity tagging of documents on download - to help users understand which versions they are viewing and prevent duplicate content items in the content repository. New "smart productivity folders" to show a users workflow inbox, saved searches and checked out content directly from Windows Explorer Drag and drop metadata pop-ups Check in and check out for all file formats with any standard WebDAV server As part of Oracle's Enterprise Application Documents initiative, Oracle Content Management 11g also provides certified application integrations with solution templates You can read the press release here. You can see more assets at the launch center here. You can sign up for the announcement webinar and hear more about the new features here. You can read the benchmarking study here.

    Read the article

  • The .NET Rocks! Visual Studio 2010 Road Trip

    - by Laila
    Carl Franklin and Richard Campbell, the two .NET Rocks radio show hosts, have decided to set off to 15 cities in the US, between April 19th and May 7th, in their DotNetMobile (a 30 foot RV). What for you'll ask me? Well, to drive around the US, meet up with .NET developers, and show off the latest and greatest in Visual Studio 2010 and .NET 4.0! Each evening, they stop in a city and host a three hour event in front of a 100 to 300 crowd of developers, where Carl is showing off media features in Silverlight 4 and their road trip tracking application, whilst Richard is demo-ing the web performance testing features of VS2010 using his portable server rig. But before they take to the stage, they have a special guest brought in - a rock star from the Visual Studio world - whom they interview for an hour as a .NET Rock episode. So far, they've had - amongst others - Phil Haack, a Program Manager with the ASP.NET team working on ASP.NET MVC, Dan Fernandez, an Evangelism Manager in the Developer and Platform Evangelism team at Microsoft, and Beth Massi, Senior Program Manager on the Visual Studio Community Team at Microsoft. I love the fact that the audience gets a chance to participate, ask questions and have a great laugh, as you can hear in the first episode! Along the way, the .NET Rocks guys are giving away great prizes (including .NET Reflector Pro, ANTS Memory Profiler licenses, and "40" LCD TVs!). Even more out of the ordinary, at each stop on the road trip, one lucky attendee (who entered in the Ride Along competition) gets to jump in the RV with Carl and Richard and ride along with them to the next stop on the roadtrip. How cool is that! Richard told us: "Our first winner in Mountain View was Eric Ziko. I was looking for him to announce that he had won, when he found us and gave us a bottle of scotch he had brought just to say 'thanks for the great show'. We all had a toast from the bottle the next night when he headed back home." Cheeky! There's still space to a few of these events, so if you want to attend, register now, because it's first come first serve. We're grateful to Richard and Carl for giving us the opportunity to sponsor this major .NET event! A unique .NET adventure worth following for sure. Cheers, Laila

    Read the article

  • Is 4-5 years the “Midlife Crisis” for a programming career?

    - by Jeff
    I’ve been programming C# professionally for a bit over 4 years now. For the past 4 years I’ve worked for a few small/medium companies ranging from “web/ads agencies”, small industry specific software shops to a small startup. I've been mainly doing "business apps" that involves using high-level programming languages (garbage collected) and my overall experience was that all of the works I’ve done could have been more professional. A lot of the things were done incorrectly (in a rush) mainly due to cost factor that people always wanted something “now” and with the smallest amount of spendable money. I kept on thinking maybe if I could work for a bigger companies or a company that’s better suited for programmers, or somewhere that's got the money and time to really build something longer term and more maintainable I may have enjoyed more in my career. I’ve never had a “mentor” that guided me through my 4 years career. I am pretty much blog / google / self taught programmer other than my bachelor IT degree. I’ve also observed another issue that most so called “senior” programmer in “my working environment” are really not that senior skill wise. They are “senior” only because they’ve been a long time programmer, but the code they write or the decisions they make are absolutely rubbish! They don't want to learn, they don't want to be better they just want to get paid and do what they've told to do which make sense and most of us are like that. Maybe that’s why they are where they are now. But I don’t want to become like them I want to be better. I’ve run into a mental state that I no longer intend to be a programmer for my future career. I started to think maybe there are better things out there to work on. The more blogs I read, the more “best practices” I’ve tried the more I feel I am drifting away from “my reality”. But I am not a great programmer otherwise I don't think I am where I am now. I think 4-5 years is a stage that can be a step forward career wise or a step out of where you are. I just wanted to hear what other have to say about what I’ve mentioned above and whether you’ve experienced similar situation in your past programming career and how you dealt with it. Thanks.

    Read the article

  • Survey Probes the Project Management Concerns of Financial Services Executives

    - by Melissa Centurio Lopes
    Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* 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-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} Do you wonder what are the top reasons why large projects in the financial industry fail to meet budgets, schedules, and other key performance criteria? Being able to answer this question can provide important insight and value of good project management practices for your organization. According to 400 senior executives who participated in a new survey conducted by the Economist Intelligence Unit and sponsored by Oracle, unrealistic project goals is the main reason for roadblocks to success Other common stumbling blocks are poor alignment between project and organizational goals, inadequate human resources, lack of strong leadership, and unwillingness among team members to point out problems. This survey sample also had a lot to say about the impact of regulatory compliance on the overall portfolio management process. Thirty-nine percent acknowledged that regulations enabled efficient functioning of their businesses. But a similar number said that regulations often require more financial resources than were originally allocated to bring projects in on time. Regulations were seen by 35 percent of the executives as roadblocks to their ability to invest in the organization’s growth and success. These revelations among others are discussed in depth in a new on-demand Webcast titled “Too Good to Fail: Developing Project Management Expertise in Financial Services” now available from Oracle. The Webcast features Brian Gardner, editor of the Economist Intelligence Unit, who presents these findings from this survey along with Guy Barlow, director of industry strategy for Oracle Primavera. Together, they analyze what the numbers mean for project and program managers and the financial services industry. Register today to watch the on-demand Webcast and get a full rundown and analysis of the survey results. Take the Economist Intelligence Unit benchmarking survey and see how your views compare with those of other financial services industry executives in ensuring project success.  Read more in the October Edition of the quarterly Information InDepth EPPM Newsletter

    Read the article

  • Week 21: FY10 in the Rear View Mirror

    - by sandra.haan
    FY10 is coming to a close and before we dive into FY11 we thought we would take a walk down memory lane and reminisce on some of our favorite Oracle PartnerNetwork activities. June 2009 brought One Red Network to partners offering access to the same virtual kickoff environment used by Oracle employees. It was a new way to deliver valuable content to key stakeholders (and without the 100+ degree temperatures). Speaking of hot, Oracle also announced in June new licensing options for our ISV partners. This model enables an even broader community of ISVs to build, deploy and manage SaaS applications on the same platform. While some people took the summer off, the OPN Program team was working away to deliver a brand new partner program - Oracle PartnerNetwork Specialized - at Oracle OpenWorld in October. Specialized. Recognized. Preferred. If you haven't gotten the message yet, we may need an emergency crew to pull you out from that rock you've been hiding under. But seriously, the announcement at the OPN Forum drew a big crowd and our FY11 event is shaping up to be just as exciting. OPN Specialized was announced in October and opened our doors for enrollment in December 2009. To mark our grand opening we held our first ever social webcast allowing partners from around the world to interact with us live throughout the day. We had a lot of great conversations and really enjoyed the chance to speak with so many of you. After a short holiday break we were back at it - just a small announcement - Oracle's acquisition of Sun. In case you missed it, here is a short field report from Ted Bereswill, SVP North America Alliances & Channels on the partner events to support the announcement: And while we're announcing things - did we mention that both Ted Bereswill and Judson Althoff were named Channel Chiefs by CRN? Not only do we have a couple of Channel Chiefs, but Oracle also won the Partner Program 5 Star Programs Award and took top honors at the CRN Channel Champion Awards for Financial Factors/Financial Performance in the category of Data and Information Management and the and Xchange Solution Provider event in March 2010. We actually caught up with Judson at this event for a quick recap of our participation: But awards aside, let's not forget our main focus in FY10 and that is Specialization. In April we announced that we had over 35 Specializations available for partners and a plan to deliver even more in FY11. We are just days away from the end of FY10 but hope you enjoyed our walk down memory lane. We are already planning lots of activity for our partners in FY11 starting with our Partner Kickoff event on June 29th. Join us to hear the vision and strategy for FY11 and interact with regional A&C leaders. We look forward to talking with you then. The OPN Communications Team

    Read the article

  • Raymond James at Oracle OpenWorld: Showcasing Real Time Data Integration.

    - by Christophe Dupupet
    Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* 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-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} In today’s always-on, always connected world, integrating data in real-time is a necessity for most companies and most industries. The experts at Raymond James Financials, using Oracle GoldenGate and Oracle Data Integrator, have designed a real-time data integration solution for their operational data store and services that support applications throughout the enterprise . They boast an amazing number of daily executions, while dramatically reducing data latency,  increasing data service performance, and speeding time to market. Normal 0 false false false EN-US X-NONE 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-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman","serif";} To know more on how they have achieved such results, come listen to Ryan Fonnett and Tim Garrod: they will explain how they implemented their solution, and also illustrate their explanations with a live demonstration of their work. A presentation not to be missed! Real-Time Data Integrationwith Oracle Data Integratorat Raymond James October 1st 2012 at 4:45pm Moscone West, room 3005

    Read the article

  • So it comes to PASS…

    - by Tony Davis
    How does your company gauge the benefit of attending a technical conference? What's the best change you made as a direct result of attendance? It's time again for the PASS Summit and I, like most people go with a set of general goals for enhancing technical knowledge; to learn more about PowerShell, to drill into SQL Server performance tuning techniques, and so on. Most will write up a brief report on the event for the rest of the team. Ideally, however, it will go a bit further than that; each conference should result in a specific improvement to one of your systems, or in the way you do your job. As co-editor of Simple-talk.com, and responsible for the majority of our SQL books, my “high level” goals don't vary much from conference to conference. I'm always on the lookout for good new authors. I target interesting new technologies and tools and try to learn more. I return with a list of actions, new articles to commission, and potential new authors. Three years ago, however, I started setting myself the goal of implementing “one new thing” after each conference. After one, I adopted Kanban for managing my workload, a technique that places strict limits on “work in progress” and makes the overall workload, and backlog, highly visible. After another I trialled a community book project. At PASS 2010, one of my general goals was to delve deeper into SQL Server transaction log mechanics, but on top of that, I set a specific goal of writing something useful on the topic. I started a Stairway series and, ultimately, it's turned into a book! If you're attending the PASS Summit this year, take some time to consider what specific improvement or change you'll implement as a result. Also, try to drop by the Red Gate booth (#101). During the Vendor event on Wednesday evening, Gail Shaw and I will be there to discuss, and hand out copies of the book. Cheers, Tony.  

    Read the article

  • Guidance and Pricing for MSDN 2010

    - by John Alexander
    Sorry for the rather lengthy post here. I get asked this all the time so I decided to post it…Visual Studio 2010 editions will be available on April 12, 2010. Product Features Professional with MSDN Essentials Professional with MSDN Premium with MSDN Ultimate with MSDN Test Professional with MSDN Debugging and Diagnostics IntelliTrace (Historical Debugger)         Static Code Analysis       Code Metrics       Profiling       Debugger   Testing Tools Unit Testing   Code Coverage       Test Impact Analysis       Coded UI Test       Web Performance Testing         Load Testing1         Microsoft Test Manager 2010       Test Case Management2       Manual Test Execution       Fast-Forward for Manual Testing       Lab Management Configuration3       Integrated Development Environment Multiple Monitor Support   Multi-Targeting   One Click Web Deployment   JavaScript and jQuery Support   Extensible WPF-Based Environment Database Development Database Deployment       Database Change Management2       Database Unit Testing       Database Test Data Generation       Data Access   Development Platform Support Windows Development   Web Development   Office and SharePoint Development   Cloud Development   Customizable Development Experience   Architecture and Modeling Architecture Explorer         UML® 2.0 Compliant Diagrams (Activity, Use Case, Sequence, Class, Component)         Layer Diagram and Dependency Validation         Read-only diagrams (UML, Layer, DGML Graphs)         Lab Management Virtual environment setup & tear down3       Provision environment from template3       Checkpoint environment3       Team Foundation Server Version Control2   Work Item Tracking2   Build Automation2   Team Portal2   Reporting & Business Intelligence2   Agile Planning Workbook2   Microsoft Visual Studio Team Explorer 2010   Test Case Management2       MSDN Subscription – Software and Services for Production Use Windows Azure Platform 20 hrs/mo † 50 hrs/mo † 100 hrs/mo † 250 hrs/mo † n/a Microsoft Visual Studio Team Foundation Server 2010   Microsoft Visual Studio Team Foundation Server 2010 CAL   1 1 1 1 Microsoft Expression Studio 3       Microsoft Office Professional Plus 2010, Project Professional 2010, Visio Premium 2010 (following Office 2010 launch)       MSDN Subscription – Software for Development and Testing 4 Windows 7, Windows Server 2008 R2 and SQL Server 2008 Toolkits, Software Development Kits, Driver Development Kits Previous versions of Windows (client and server operation systems)   Previous versions of Microsoft SQL Server   Microsoft Office       Microsoft Dynamics       All other Servers       Windows Embedded operating systems       Teamprise         MSDN Subscription – Other Benefits Technical support incidents 0 2 4 4 2 Priority support in MSDN Forums Microsoft e-learning collections (typically 10 courses or 20 hours) 0 1 2 2 1 MSDN Flash newsletter MSDN Online Concierge MSDN Magazine   System Requirements View View View View View Buy from (MSRP) $799 $1,199 $5,469 $11,899 $2,169 Renew from (MSRP) $549 (upgrade) $799 $2,299 $3,799 $899 † Availability varies by country and subscription level.  Details available on the MSDN site 1. May require one or more Microsoft Visual Studio Load Test Virtual User Pack 2010 2. Requires Team Foundation Server and a Team Foundation Server CAL 3. Requires Microsoft Visual Studio Lab Management 2010 4. Per-user license allows unlimited installations and use for designing, developing, testing, and demonstrating applications. UML is a registered trademark of Object Management Group, Inc. Windows is either a registered trademark or trademark of Microsoft Corporation in the United States and/or other countries.

    Read the article

  • Up in the Air: Team Oracle Play-by-Play

    - by Aaron Lazenby
    Yesterday, I had the amazing opportunity to fly along with Sean D. Tucker and Team Oracle. Leaving from the San Carols airport, we did a 30 minute flight over the Pacific just south of the coastal town of Half Moon Bay. In that half hour, I rode through a massive 4G loop, survived a crushing hammerhead, and took control of the plane to perform a basic wing over (you can learn what the heck I'm talking about by visiting this website). I have lots of great video, but it's going to take me some time to make sense of it. For now, here's my Twitter-based play-by-play of yesterday's events. Many thanks to Sean D. Tucker and the whole crew (Ben and Ian, especially) for this great opportunity to fly with Team Oracle.Live tweets from @OracleProfitI will be spending the afternoon in a stunt plane, upside down above the San Francisco bay. http://bit.ly/cwkrkIAt the San Carlos airport. More than slightly freaked out. Shaking hands diminish texting ability. Slightly reassuring. http://yfrog.com/1qt61nj There go the doors to the photo plane... #teamoracle http://yfrog.com/58ywljSean D Tucker assures me: "The sky is a great place to be." Helpful, but I'm still nervous. #teamoracle"You get a parachute. He gets a harness." How was this decision made? #teamoracleThe plane with @radu43 has returned. I'm up next...Couldn't help myself...drank a soda before flying. Mistake? We'll see... #teamoracleAdvice of the day "If you pull with two hands, you improve the chances of the chute deploying on the first try." Lovely. #teamoracleI feel so strange. But I flew a high performance airplane. And did an aerobatics move. Wild. #teamoracle"Flying ten feet off he ground, upside-down at 250 miles per hour isn't exciting to me." Sean D. Tucker #teamoracle"What is exciting to me is flying that perfect pattern, just like I imagined it in my head." Sean D. Tucker #teamoracle"You're going to sleep well tonight. You just carried four times your body weight." #teamoracle #gforce Just watched the #teamoracle plane take off for its flight home. I'm waiting for Caltrain. #undignifiedanticlimaxEnough with the #teamoracle. Check http://blogs.oracle.com/profit for the video. Coming soon! 

    Read the article

  • Global Cache CR Requested But Current Block Received

    - by Liu Maclean(???)
    ????????«MINSCN?Cache Fusion Read Consistent» ????,???????????? ??????????????????: SQL> select * from V$version; BANNER -------------------------------------------------------------------------------- Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production PL/SQL Release 11.2.0.3.0 - Production CORE 11.2.0.3.0 Production TNS for Linux: Version 11.2.0.3.0 - Production NLSRTL Version 11.2.0.3.0 - Production SQL> select count(*) from gv$instance; COUNT(*) ---------- 2 SQL> select * from global_name; GLOBAL_NAME -------------------------------------------------------------------------------- www.oracledatabase12g.com ?11gR2 2??RAC??????????status???XG,????Xcurrent block???INSTANCE 2?hold?,?????INSTANCE 1?????????,?????: SQL> select * from test; ID ---------- 1 2 SQL> select dbms_rowid.rowid_block_number(rowid),dbms_rowid.rowid_relative_fno(rowid) from test; DBMS_ROWID.ROWID_BLOCK_NUMBER(ROWID) DBMS_ROWID.ROWID_RELATIVE_FNO(ROWID) ------------------------------------ ------------------------------------ 89233 1 89233 1 SQL> alter system flush buffer_cache; System altered. INSTANCE 1 Session A: SQL> update test set id=id+1 where id=1; 1 row updated. INSTANCE 1 Session B: SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 1 0 3 1755287 SQL> oradebug setmypid; Statement processed. SQL> oradebug dump gc_elements 255; Statement processed. SQL> oradebug tracefile_name; /s01/orabase/diag/rdbms/vprod/VPROD1/trace/VPROD1_ora_19111.trc GLOBAL CACHE ELEMENT DUMP (address: 0xa4ff3080): id1: 0x15c91 id2: 0x1 pkey: OBJ#76896 block: (1/89233) lock: X rls: 0x0 acq: 0x0 latch: 3 flags: 0x20 fair: 0 recovery: 0 fpin: 'kdswh11: kdst_fetch' bscn: 0x0.146e20 bctx: (nil) write: 0 scan: 0x0 lcp: (nil) lnk: [NULL] lch: [0xa9f6a6f8,0xa9f6a6f8] seq: 32 hist: 58 145:0 118 66 144:0 192 352 197 48 121 113 424 180 58 LIST OF BUFFERS LINKED TO THIS GLOBAL CACHE ELEMENT: flg: 0x02000001 lflg: 0x1 state: XCURRENT tsn: 0 tsh: 2 addr: 0xa9f6a5c8 obj: 76896 cls: DATA bscn: 0x0.1ac898 BH (0xa9f6a5c8) file#: 1 rdba: 0x00415c91 (1/89233) class: 1 ba: 0xa9e56000 set: 5 pool: 3 bsz: 8192 bsi: 0 sflg: 3 pwc: 0,15 dbwrid: 0 obj: 76896 objn: 76896 tsn: 0 afn: 1 hint: f hash: [0x91f4e970,0xbae9d5b8] lru: [0x91f58848,0xa9f6a828] lru-flags: debug_dump obj-flags: object_ckpt_list ckptq: [0x9df6d1d8,0xa9f6a740] fileq: [0xa2ece670,0xbdf4ed68] objq: [0xb4964e00,0xb4964e00] objaq: [0xb4964de0,0xb4964de0] st: XCURRENT md: NULL fpin: 'kdswh11: kdst_fetch' tch: 2 le: 0xa4ff3080 flags: buffer_dirty redo_since_read LRBA: [0x19.5671.0] LSCN: [0x0.1ac898] HSCN: [0x0.1ac898] HSUB: [1] buffer tsn: 0 rdba: 0x00415c91 (1/89233) scn: 0x0000.001ac898 seq: 0x01 flg: 0x00 tail: 0xc8980601 frmt: 0x02 chkval: 0x0000 type: 0x06=trans data ??????block: (1/89233)?GLOBAL CACHE ELEMENT DUMP?LOCK????X ??XG , ??????Current Block????Instance??modify???,????????????? ????Instance 2 ????: Instance 2 Session C: SQL> update test set id=id+1 where id=2; 1 row updated. Instance 2 Session D: SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 1 0 3 1756658 SQL> oradebug setmypid; Statement processed. SQL> oradebug dump gc_elements 255; Statement processed. SQL> oradebug tracefile_name; /s01/orabase/diag/rdbms/vprod/VPROD2/trace/VPROD2_ora_13038.trc GLOBAL CACHE ELEMENT DUMP (address: 0x89fb25a0): id1: 0x15c91 id2: 0x1 pkey: OBJ#76896 block: (1/89233) lock: XG rls: 0x0 acq: 0x0 latch: 3 flags: 0x20 fair: 0 recovery: 0 fpin: 'kduwh01: kdusru' bscn: 0x0.1acdf3 bctx: (nil) write: 0 scan: 0x0 lcp: (nil) lnk: [NULL] lch: [0x96f4cf80,0x96f4cf80] seq: 61 hist: 324 21 143:0 19 16 352 329 144:6 14 7 352 197 LIST OF BUFFERS LINKED TO THIS GLOBAL CACHE ELEMENT: flg: 0x0a000001 state: XCURRENT tsn: 0 tsh: 1 addr: 0x96f4ce50 obj: 76896 cls: DATA bscn: 0x0.1acdf6 BH (0x96f4ce50) file#: 1 rdba: 0x00415c91 (1/89233) class: 1 ba: 0x96bd4000 set: 5 pool: 3 bsz: 8192 bsi: 0 sflg: 2 pwc: 0,15 dbwrid: 0 obj: 76896 objn: 76896 tsn: 0 afn: 1 hint: f hash: [0x96ee1fe8,0xbae9d5b8] lru: [0x96f4d0b0,0x96f4cdc0] obj-flags: object_ckpt_list ckptq: [0xbdf519b8,0x96f4d5a8] fileq: [0xbdf519d8,0xbdf519d8] objq: [0xb4a47b90,0xb4a47b90] objaq: [0x96f4d0e8,0xb4a47b70] st: XCURRENT md: NULL fpin: 'kduwh01: kdusru' tch: 1 le: 0x89fb25a0 flags: buffer_dirty redo_since_read remote_transfered LRBA: [0x11.9e18.0] LSCN: [0x0.1acdf6] HSCN: [0x0.1acdf6] HSUB: [1] buffer tsn: 0 rdba: 0x00415c91 (1/89233) scn: 0x0000.001acdf6 seq: 0x01 flg: 0x00 tail: 0xcdf60601 frmt: 0x02 chkval: 0x0000 type: 0x06=trans data GCS CLIENT 0x89fb2618,6 resp[(nil),0x15c91.1] pkey 76896.0 grant 2 cvt 0 mdrole 0x42 st 0x100 lst 0x20 GRANTQ rl G0 master 1 owner 2 sid 0 remote[(nil),0] hist 0x94121c601163423c history 0x3c.0x4.0xd.0xb.0x1.0xc.0x7.0x9.0x14.0x1. cflag 0x0 sender 1 flags 0x0 replay# 0 abast (nil).x0.1 dbmap (nil) disk: 0x0000.00000000 write request: 0x0000.00000000 pi scn: 0x0000.00000000 sq[(nil),(nil)] msgseq 0x1 updseq 0x0 reqids[6,0,0] infop (nil) lockseq x2b8 pkey 76896.0 hv 93 [stat 0x0, 1->1, wm 32768, RMno 0, reminc 18, dom 0] kjga st 0x4, step 0.0.0, cinc 20, rmno 6, flags 0x0 lb 0, hb 0, myb 15250, drmb 15250, apifrz 0 ?Instance 2??????block: (1/89233)? GLOBAL CACHE ELEMENT Lock Convert?lock: XG ????GC_ELEMENTS DUMP???XCUR Cache Fusion?,???????X$ VIEW,??? X$LE X$KJBR X$KJBL, ???X$ VIEW???????????????????: INSTANCE 2 Session D: SELECT * FROM x$le WHERE le_addr IN (SELECT le_addr FROM x$bh WHERE obj IN (SELECT data_object_id FROM dba_objects WHERE owner = 'SYS' AND object_name = 'TEST') AND class = 1 AND state != 3); ADDR INDX INST_ID LE_ADDR LE_ID1 LE_ID2 ---------------- ---------- ---------- ---------------- ---------- ---------- LE_RLS LE_ACQ LE_FLAGS LE_MODE LE_WRITE LE_LOCAL LE_RECOVERY ---------- ---------- ---------- ---------- ---------- ---------- ----------- LE_BLKS LE_TIME LE_KJBL ---------- ---------- ---------------- 00007F94CA14CF60 7003 2 0000000089FB25A0 89233 1 0 0 32 2 0 1 0 1 0 0000000089FB2618 PCM Resource NAME?[ID1][ID2],[BL]???, ID1?ID2 ??blockno? fileno????, ??????????GC_elements dump?? id1: 0x15c91 id2: 0×1 pkey: OBJ#76896 block: (1/89233)?? ,?  kjblname ? kjbrname ??”[0x15c91][0x1],[BL]” ??: INSTANCE 2 Session D: SQL> set linesize 80 pagesize 1400 SQL> SELECT * 2 FROM x$kjbl l 3 WHERE l.kjblname LIKE '%[0x15c91][0x1],[BL]%'; ADDR INDX INST_ID KJBLLOCKP KJBLGRANT KJBLREQUE ---------------- ---------- ---------- ---------------- --------- --------- KJBLROLE KJBLRESP KJBLNAME ---------- ---------------- ------------------------------ KJBLNAME2 KJBLQUEUE ------------------------------ ---------- KJBLLOCKST KJBLWRITING ---------------------------------------------------------------- ----------- KJBLREQWRITE KJBLOWNER KJBLMASTER KJBLBLOCKED KJBLBLOCKER KJBLSID KJBLRDOMID ------------ ---------- ---------- ----------- ----------- ---------- ---------- KJBLPKEY ---------- 00007F94CA22A288 451 2 0000000089FB2618 KJUSEREX KJUSERNL 0 00 [0x15c91][0x1],[BL][ext 0x0,0x 89233,1,BL 0 GRANTED 0 0 1 0 0 0 0 0 76896 SQL> SELECT r.* FROM x$kjbr r WHERE r.kjbrname LIKE '%[0x15c91][0x1],[BL]%'; no rows selected Instance 1 session B: SQL> SELECT r.* FROM x$kjbr r WHERE r.kjbrname LIKE '%[0x15c91][0x1],[BL]%'; ADDR INDX INST_ID KJBRRESP KJBRGRANT KJBRNCVL ---------------- ---------- ---------- ---------------- --------- --------- KJBRROLE KJBRNAME KJBRMASTER KJBRGRANTQ ---------- ------------------------------ ---------- ---------------- KJBRCVTQ KJBRWRITER KJBRSID KJBRRDOMID KJBRPKEY ---------------- ---------------- ---------- ---------- ---------- 00007F801ACA68F8 1355 1 00000000B5A62AE0 KJUSEREX KJUSERNL 0 [0x15c91][0x1],[BL][ext 0x0,0x 0 00000000B48BB330 00 00 0 0 76896 ??????Instance 1???block: (1/89233),??????Instance 2 build cr block ????Instance 1, ?????????? ????? Instance 1? Foreground Process ? Instance 2?LMS??????RAC  TRACE: Instance 2: [oracle@vrh2 ~]$ ps -ef|grep ora_lms|grep -v grep oracle 23364 1 0 Apr29 ? 00:33:15 ora_lms0_VPROD2 SQL> oradebug setospid 23364 Oracle pid: 13, Unix process pid: 23364, image: [email protected] (LMS0) SQL> oradebug event 10046 trace name context forever,level 8:10708 trace name context forever,level 103: trace[rac.*] disk high; Statement processed. SQL> oradebug tracefile_name /s01/orabase/diag/rdbms/vprod/VPROD2/trace/VPROD2_lms0_23364.trc Instance 1 session B : SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 3 1756658 3 1756661 3 1755287 Instance 1 session A : SQL> alter session set events '10046 trace name context forever,level 8:10708 trace name context forever,level 103: trace[rac.*] disk high'; Session altered. SQL> select * from test; ID ---------- 2 2 SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 3 1761520 ?x$BH?????,???????Instance 1???build??CR block,????? TRACE ??: Instance 1 foreground Process: PARSING IN CURSOR #140336527348792 len=18 dep=0 uid=0 oct=3 lid=0 tim=1335939136125254 hv=1689401402 ad='b1a4c828' sqlid='c99yw1xkb4f1u' select * from test END OF STMT PARSE #140336527348792:c=2999,e=2860,p=0,cr=0,cu=0,mis=1,r=0,dep=0,og=1,plh=1357081020,tim=1335939136125253 EXEC #140336527348792:c=0,e=40,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=1357081020,tim=1335939136125373 WAIT #140336527348792: nam='SQL*Net message to client' ela= 6 driver id=1650815232 #bytes=1 p3=0 obj#=0 tim=1335939136125420 *** 2012-05-02 02:12:16.125 kclscrs: req=0 block=1/89233 2012-05-02 02:12:16.125574 : kjbcro[0x15c91.1 76896.0][4] *** 2012-05-02 02:12:16.125 kclscrs: req=0 typ=nowait-abort *** 2012-05-02 02:12:16.125 kclscrs: bid=1:3:1:0:f:1e:0:0:10:0:0:0:1:2:4:1:20:0:0:0:c3:49:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:4:3:2:1:2:0:1c:0:4d:26:a3:52:0:0:0:0:c7:c:ca:62:c3:49:0:0:0:0:1:0:14:8e:47:76:1:2:dc:5:a9:fe:17:75:0:0:0:0:0:0:0:0:0:0:0:0:99:ed:0:0:0:0:0:0:10:0:0:0 2012-05-02 02:12:16.125718 : kjbcro[0x15c91.1 76896.0][4] 2012-05-02 02:12:16.125751 : GSIPC:GMBQ: buff 0xba0ee018, queue 0xbb79a7b8, pool 0x60013fa0, freeq 0, nxt 0xbb79a7b8, prv 0xbb79a7b8 2012-05-02 02:12:16.125780 : kjbsentscn[0x0.1ae0f0][to 2] 2012-05-02 02:12:16.125806 : GSIPC:SENDM: send msg 0xba0ee088 dest x20001 seq 177740 type 36 tkts xff0000 mlen x1680198 2012-05-02 02:12:16.125918 : kjbmscr(0x15c91.1)reqid=0x8(req 0xa4ff30f8)(rinst 1)hldr 2(infosz 200)(lseq x2b8) 2012-05-02 02:12:16.126959 : GSIPC:KSXPCB: msg 0xba0ee088 status 30, type 36, dest 2, rcvr 1 *** 2012-05-02 02:12:16.127 kclwcrs: wait=0 tm=1233 *** 2012-05-02 02:12:16.127 kclwcrs: got 1 blocks from ksxprcv WAIT #140336527348792: nam='gc cr block 2-way' ela= 1233 p1=1 p2=89233 p3=1 obj#=76896 tim=1335939136127199 2012-05-02 02:12:16.127272 : kjbcrcomplete[0x15c91.1 76896.0][0] 2012-05-02 02:12:16.127309 : kjbrcvdscn[0x0.1ae0f0][from 2][idx 2012-05-02 02:12:16.127329 : kjbrcvdscn[no bscn <= rscn 0x0.1ae0f0][from 2] ???? kjbcro[0x15c91.1 76896.0][4] kjbsentscn[0x0.1ae0f0][to 2] ?Instance 2??SCN=1ae0f0=1761520? block: (1/89233),???’gc cr block 2-way’ ??,?????????CR block? Instance 2 LMS TRACE 2012-05-02 02:12:15.634057 : GSIPC:RCVD: ksxp msg 0x7f16e1598588 sndr 1 seq 0.177740 type 36 tkts 0 2012-05-02 02:12:15.634094 : GSIPC:RCVD: watq msg 0x7f16e1598588 sndr 1, seq 177740, type 36, tkts 0 2012-05-02 02:12:15.634108 : GSIPC:TKT: collect msg 0x7f16e1598588 from 1 for rcvr -1, tickets 0 2012-05-02 02:12:15.634162 : kjbrcvdscn[0x0.1ae0f0][from 1][idx 2012-05-02 02:12:15.634186 : kjbrcvdscn[no bscn1, wm 32768, RMno 0, reminc 18, dom 0] kjga st 0x4, step 0.0.0, cinc 20, rmno 6, flags 0x0 lb 0, hb 0, myb 15250, drmb 15250, apifrz 0 GCS CLIENT END 2012-05-02 02:12:15.635211 : kjbdowncvt[0x15c91.1 76896.0][1][options x0] 2012-05-02 02:12:15.635230 : GSIPC:AMBUF: rcv buff 0x7f16e1c56420, pool rcvbuf, rqlen 1103 2012-05-02 02:12:15.635308 : GSIPC:GPBMSG: new bmsg 0x7f16e1c56490 mb 0x7f16e1c56420 msg 0x7f16e1c564b0 mlen 152 dest x101 flushsz -1 2012-05-02 02:12:15.635334 : kjbmslset(0x15c91.1)) seq 0x4 reqid=0x6 (shadow 0xb48bb330.xb)(rsn 2)(mas@1) 2012-05-02 02:12:15.635355 : GSIPC:SPBMSG: send bmsg 0x7f16e1c56490 blen 184 msg 0x7f16e1c564b0 mtype 33 attr|dest x30101 bsz|fsz x1ffff 2012-05-02 02:12:15.635377 : GSIPC:SNDQ: enq msg 0x7f16e1c56490, type 65521 seq 118669, inst 1, receiver 1, queued 1 *** 2012-05-02 02:12:15.635 kclccctx: cleanup copy 0x7f16e1d94798 2012-05-02 02:12:15.635479 : [kjmpmsgi:compl][type 36][msg 0x7f16e1598588][seq 177740.0][qtime 0][ptime 1257] 2012-05-02 02:12:15.635511 : GSIPC:BSEND: flushing sndq 0xb491dd28, id 1, dcx 0xbc516778, inst 1, rcvr 1 qlen 0 1 2012-05-02 02:12:15.635536 : GSIPC:BSEND: no batch1 msg 0x7f16e1c56490 type 65521 len 184 dest (1:1) 2012-05-02 02:12:15.635557 : kjbsentscn[0x0.1ae0f1][to 1] 2012-05-02 02:12:15.635578 : GSIPC:SENDM: send msg 0x7f16e1c56490 dest x10001 seq 118669 type 65521 tkts x10002 mlen xb800e8 WAIT #0: nam='gcs remote message' ela= 180 waittime=1 poll=0 event=0 obj#=0 tim=1335939135635819 2012-05-02 02:12:15.635853 : GSIPC:RCVD: ksxp msg 0x7f16e167e0b0 sndr 1 seq 0.177741 type 32 tkts 0 2012-05-02 02:12:15.635875 : GSIPC:RCVD: watq msg 0x7f16e167e0b0 sndr 1, seq 177741, type 32, tkts 0 2012-05-02 02:12:15.636012 : GSIPC:TKT: collect msg 0x7f16e167e0b0 from 1 for rcvr -1, tickets 0 2012-05-02 02:12:15.636040 : kjbrcvdscn[0x0.1ae0f1][from 1][idx 2012-05-02 02:12:15.636060 : kjbrcvdscn[no bscn <= rscn 0x0.1ae0f1][from 1] 2012-05-02 02:12:15.636082 : GSIPC:TKT: dest (1:1) rtkt not acked 1  unassigned bufs 0  tkts 0  newbufs 0 2012-05-02 02:12:15.636102 : GSIPC:TKT: remove ctx dest (1:1) 2012-05-02 02:12:15.636125 : [kjmxmpm][type 32][seq 0.177741][msg 0x7f16e167e0b0][from 1] 2012-05-02 02:12:15.636146 : kjbmpocr(0xb0.6)seq 0x1,reqid=0x23a,(client 0x9fff7b58,0x1)(from 1)(lseq xdf0) 2????LMS????????? ??gcs remote message GSIPC ????SCN=[0x0.1ae0f0] block=1/89233???,??BAST kjbmpbast(0x15c91.1),?? block=1/89233??????? ??fairness??(?11.2.0.3???_fairness_threshold=2),?current block?KCL: F156: fairness downconvert,?Xcurrent DownConvert? Scurrent: Instance 2: SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 2 0 3 1756658 ??Instance 2 LMS ?cr block??? kjbmslset(0x15c91.1)) ????SEND QUEUE GSIPC:SNDQ: enq msg 0x7f16e1c56490? ???????Instance 1???? block: (1/89233)??? ??????: Instance 2: SQL> select CURRENT_RESULTS,LIGHT_WORKS from v$cr_block_server; CURRENT_RESULTS LIGHT_WORKS --------------- ----------- 29273 437 Instance 1 session A: SQL> SQL> select * from test; ID ---------- 2 2 SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 3 1761942 3 1761932 1 0 3 1761520 Instance 2: SQL> select CURRENT_RESULTS,LIGHT_WORKS from v$cr_block_server; CURRENT_RESULTS LIGHT_WORKS --------------- ----------- 29274 437 select * from test END OF STMT PARSE #140336529675592:c=0,e=337,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=1357081020,tim=1335939668940051 EXEC #140336529675592:c=0,e=96,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=1357081020,tim=1335939668940204 WAIT #140336529675592: nam='SQL*Net message to client' ela= 5 driver id=1650815232 #bytes=1 p3=0 obj#=0 tim=1335939668940348 *** 2012-05-02 02:21:08.940 kclscrs: req=0 block=1/89233 2012-05-02 02:21:08.940676 : kjbcro[0x15c91.1 76896.0][5] *** 2012-05-02 02:21:08.940 kclscrs: req=0 typ=nowait-abort *** 2012-05-02 02:21:08.940 kclscrs: bid=1:3:1:0:f:21:0:0:10:0:0:0:1:2:4:1:20:0:0:0:c3:49:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:4:3:2:1:2:0:1f:0:4d:26:a3:52:0:0:0:0:c7:c:ca:62:c3:49:0:0:0:0:1:0:17:8e:47:76:1:2:dc:5:a9:fe:17:75:0:0:0:0:0:0:0:0:0:0:0:0:99:ed:0:0:0:0:0:0:10:0:0:0 2012-05-02 02:21:08.940799 : kjbcro[0x15c91.1 76896.0][5] 2012-05-02 02:21:08.940833 : GSIPC:GMBQ: buff 0xba0ee018, queue 0xbb79a7b8, pool 0x60013fa0, freeq 0, nxt 0xbb79a7b8, prv 0xbb79a7b8 2012-05-02 02:21:08.940859 : kjbsentscn[0x0.1ae28c][to 2] 2012-05-02 02:21:08.940870 : GSIPC:SENDM: send msg 0xba0ee088 dest x20001 seq 177810 type 36 tkts xff0000 mlen x1680198 2012-05-02 02:21:08.940976 : kjbmscr(0x15c91.1)reqid=0xa(req 0xa4ff30f8)(rinst 1)hldr 2(infosz 200)(lseq x2b8) 2012-05-02 02:21:08.941314 : GSIPC:KSXPCB: msg 0xba0ee088 status 30, type 36, dest 2, rcvr 1 *** 2012-05-02 02:21:08.941 kclwcrs: wait=0 tm=707 *** 2012-05-02 02:21:08.941 kclwcrs: got 1 blocks from ksxprcv 2012-05-02 02:21:08.941818 : kjbassume[0x15c91.1][sender 2][mymode x1][myrole x0][srole x0][flgs x0][spiscn 0x0.0][swscn 0x0.0] 2012-05-02 02:21:08.941852 : kjbrcvdscn[0x0.1ae28d][from 2][idx 2012-05-02 02:21:08.941871 : kjbrcvdscn[no bscn ??????????????SCN=[0x0.1ae28c]=1761932 Version?CR block, ????receive????Xcurrent Block??SCN=1ae28d=1761933,Instance 1???Xcurrent Block???build????????SCN=1761932?CR BLOCK, ????????Current block,?????????'gc current block 2-way'? ?????????????request current block,?????kjbcro;?????Instance 2?LMS???????Current Block: Instance 2 LMS trace: 2012-05-02 02:21:08.448743 : GSIPC:RCVD: ksxp msg 0x7f16e14a4398 sndr 1 seq 0.177810 type 36 tkts 0 2012-05-02 02:21:08.448778 : GSIPC:RCVD: watq msg 0x7f16e14a4398 sndr 1, seq 177810, type 36, tkts 0 2012-05-02 02:21:08.448798 : GSIPC:TKT: collect msg 0x7f16e14a4398 from 1 for rcvr -1, tickets 0 2012-05-02 02:21:08.448816 : kjbrcvdscn[0x0.1ae28c][from 1][idx 2012-05-02 02:21:08.448834 : kjbrcvdscn[no bscn <= rscn 0x0.1ae28c][from 1] 2012-05-02 02:21:08.448857 : GSIPC:TKT: dest (1:1) rtkt not acked 2  unassigned bufs 0  tkts 0  newbufs 0 2012-05-02 02:21:08.448875 : GSIPC:TKT: remove ctx dest (1:1) 2012-05-02 02:21:08.448970 : [kjmxmpm][type 36][seq 0.177810][msg 0x7f16e14a4398][from 1] 2012-05-02 02:21:08.448993 : kjbmpbast(0x15c91.1) reqid=0x6 (req 0xa4ff30f8)(reqinst 1)(reqid 10)(flags x0) *** 2012-05-02 02:21:08.449 kclcrrf: req=48054 block=1/89233 *** 2012-05-02 02:21:08.449 kcl_compress_block: compressed: 6 free space: 7680 2012-05-02 02:21:08.449085 : kjbsentscn[0x0.1ae28d][to 1] 2012-05-02 02:21:08.449142 : kjbdeliver[to 1][0xa4ff30f8][10][current 1] 2012-05-02 02:21:08.449164 : kjbmssch(reqlock 0xa4ff30f8,10)(to 1)(bsz 344) 2012-05-02 02:21:08.449183 : GSIPC:AMBUF: rcv buff 0x7f16e18bcec8, pool rcvbuf, rqlen 1102 *** 2012-05-02 02:21:08.449 kclccctx: cleanup copy 0x7f16e1d94838 *** 2012-05-02 02:21:08.449 kcltouched: touch seconds 3271 *** 2012-05-02 02:21:08.449 kclgrantlk: req=48054 2012-05-02 02:21:08.449347 : [kjmpmsgi:compl][type 36][msg 0x7f16e14a4398][seq 177810.0][qtime 0][ptime 1119] WAIT #0: nam='gcs remote message' ela= 568 waittime=1 poll=0 event=0 obj#=0 tim=1335939668449962 2012-05-02 02:21:08.450001 : GSIPC:RCVD: ksxp msg 0x7f16e1bb22a0 sndr 1 seq 0.177811 type 32 tkts 0 2012-05-02 02:21:08.450024 : GSIPC:RCVD: watq msg 0x7f16e1bb22a0 sndr 1, seq 177811, type 32, tkts 0 2012-05-02 02:21:08.450043 : GSIPC:TKT: collect msg 0x7f16e1bb22a0 from 1 for rcvr -1, tickets 0 2012-05-02 02:21:08.450060 : kjbrcvdscn[0x0.1ae28e][from 1][idx 2012-05-02 02:21:08.450078 : kjbrcvdscn[no bscn <= rscn 0x0.1ae28e][from 1] 2012-05-02 02:21:08.450097 : GSIPC:TKT: dest (1:1) rtkt not acked 3  unassigned bufs 0  tkts 0  newbufs 0 2012-05-02 02:21:08.450116 : GSIPC:TKT: remove ctx dest (1:1) 2012-05-02 02:21:08.450136 : [kjmxmpm][type 32][seq 0.177811][msg 0x7f16e1bb22a0][from 1] 2012-05-02 02:21:08.450155 : kjbmpocr(0xb0.6)seq 0x1,reqid=0x23e,(client 0x9fff7b58,0x1)(from 1)(lseq xdf4) ???Instance 2??LMS???,???build cr block,??????Instance 1?????Current Block??????Instance 2??v$cr_block_server??????LIGHT_WORKS?????current block transfer??????,??????? CR server? Light Work Rule(Light Work Rule?8i Cr Server?????????,?Remote LMS?? build CR????????,resource holder?LMS???????block,????CR build If creating the consistent read version block involves too much work (such as reading blocks from disk), then the holder sends the block to the requestor, and the requestor completes the CR fabrication. The holder maintains a fairness counter of CR requests. After the fairness threshold is reached, the holder downgrades it to lock mode.)? ??????? CR Request ????Current Block?? ???:??????class?block,CR server??????? ??undo block?? undo header block?CR quest, LMS????Current Block, ????? ???? ??????? block cleanout? CR  Version??????? ???????? data blocks, ??????? CR quest  & CR received?(???????Light Work Rule,LMS"??"), ??Current Block??DownConvert???S lock,??LMS???????ship??current version?block? ??????? , ?????? ,???????DownConvert?????”_fairness_threshold“???200,????Xcurrent Block?????Scurrent, ????LMS?????Current Version?Data Block: SQL> show parameter fair NAME TYPE VALUE ------------------------------------ ----------- ------------------------------ _fairness_threshold integer 200 Instance 1: SQL> update test set id=id+1 where id=4; 1 row updated. Instance 2: SQL> update test set id=id+1 where id=2; 1 row updated. SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 1 0 3 1838166 ?Instance 1? ????,? ??instance 2? v$cr_block_server?? instance 1 SQL> select * from test; ID ---------- 10 3 instance 2: SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 1 0 3 1883707 8 0 SQL> select * from test; ID ---------- 10 3 SQL> select state,cr_scn_bas from x$bh where file#=1 and dbablk=89233 and state!=0; STATE CR_SCN_BAS ---------- ---------- 1 0 3 1883707 8 0 ................... SQL> / STATE CR_SCN_BAS ---------- ---------- 2 0 3 1883707 3 1883695 repeat cr request on Instance 1 SQL> / STATE CR_SCN_BAS ---------- ---------- 8 0 3 1883707 3 1883695 ??????_fairness_threshold????????,?????200 ????????CR serve??Downgrade?lock, ????data block? CR Request????Receive? Current Block?

    Read the article

  • Multi-Threaded Application vs. Single Threaded Application

    Why would we use a multi threaded application vs. a single threaded application? First we must define multithreading. Multithreading is a feature of an operating system that allows programs to run subcomponents or threads in parallel. Typically most applications only need to use one thread because they do not perform time consuming tasks. The use of multiple threads allows an application to distribute long running tasks so that they can be executed in parallel. This gives the user the perceived appearance that the application is working faster due to the fact that while one thread is waiting on an IO process the remaining tasks can make use of the available CPU. The allows working threads to execute in tandem so that they can be competed sooner. Multithreading Benefits Improved responsiveness — Users usually report improved responsiveness compared to single thread applications. Faster applications — Multiple threads can lead to improved application performance. Prioritization — Threads can be assigned a priority which would allow higher priority tasks to take precedence over lower priority tasks. Single Threading Benefits Programming and debugging —These activities are easier compared to multithreaded applications due to the reduced complexity Less Overhead — Threads add overhead to an application When developing multi-threaded applications, the following must be considered. Deadlocks occur when two threads hold a monitor that the other one requires. In essence each task is blocking the other and both tasks are waiting for the other monitor to be released. This forces an application to hang or deadlock. Resource allocation is used to prevent deadlocks because the system determines if approving the resource request will render the system in an unsafe state. An unsafe state could result in a deadlock. The system only approves requests that will lead to safe states. Thread Synchronization is used when multiple threads use the same instance of an object. The threads accessing the object can then be locked and then synchronized so that each task can interact with the static object on at a time.

    Read the article

  • SQL SERVER – PREEMPTIVE and Non-PREEMPTIVE – Wait Type – Day 19 of 28

    - by pinaldave
    In this blog post, we are going to talk about a very interesting subject. I often get questions related to SQL Server 2008 Book-Online about various Preemptive wait types. I got a few questions asking what these wait types are and how they could be interpreted. To get current wait types of the system, you can read this article and run the script: SQL SERVER – DMV – sys.dm_os_waiting_tasks and sys.dm_exec_requests – Wait Type – Day 4 of 28. Before we continue understanding them, let us study first what PREEMPTIVE and Non-PREEMPTIVE waits in SQL Server mean. PREEMPTIVE: Simply put, this wait means non-cooperative. While SQL Server is executing a task, the Operating System (OS) interrupts it. This leads to SQL Server to involuntarily give up the execution for other higher priority tasks. This is not good for SQL Server as it is a particular external process which makes SQL Server to yield. This kind of wait can reduce the performance drastically and needs to be investigated properly. Non-PREEMPTIVE: In simple terms, this wait means cooperative. SQL Server manages the scheduling of the threads. When SQL Server manages the scheduling instead of the OS, it makes sure its own priority. In this case, SQL Server decides the priority and one thread yields to another thread voluntarily. In the earlier version of SQL Server, there was no preemptive wait types mentioned and the associated task status with them was marked as suspended. In SQL Server 2005, preemptive wait types were not listed as well, but their associated task status was marked as running. In SQL Server 2008, preemptive wait types are properly listed and their associated task status is also marked as running. Now, SQL Server is in Non-Preemptive mode by default and it works fine. When CLR, extended Stored Procedures and other external components run, they run in Preemptive mode, leading to the creation of these wait types. There are a wide variety of preemptive wait types. If you see consistent high value in the Preemptive wait types, I strongly suggest that you look into the wait type and try to know the root cause. If you are still not sure, you can send me an email or leave a comment about it and I will do my best to help you reduce this wait type. Read all the post in the Wait Types and Queue series. Reference: Pinal Dave (http://blog.sqlauthority.com) Filed under: Pinal Dave, PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, SQL Wait Stats, SQL Wait Types, T SQL, Technology

    Read the article

  • ING Selecciona Oracle Fusion HCM para la Gestión del Capital Humano

    - by Noelia Gomez
    Normal 0 21 false false false ES X-NONE X-NONE MicrosoftInternetExplorer4 /* 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-qformat:yes; 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-family:"Calibri","sans-serif";} La decisión de ING sobre la elección de Oracle Fusion HCM fue impulsada por su deseo continuo de fortalecer la función de RRHH para impulsar su negocio. Tras mucho tiempo como cliente de Oracle PeopleSoft HCM, evaluó el segmento de mercado de HCM antes de seleccionar Oracle Fusion HCM, con el objetivo de seguir utilizando las herramientas principales de recursos humanos y mantener su posición como uno de los principales empleadores en los Países Bajos. ING implementará Oracle Fusion HCM, en estrecha colaboración con NorthgateArinso, miembro de Oracle PartnerNetwork (Gold), que proporcionará los servicios de alojamiento y de Oracle Consulting. La firma de servicios financieros utilizará toda la suite de Oracle Fusion Global Human Resources y Oracle Talent Management incluyendo Oracle Fusion Goal, Oracle Fusion OTBI, Oracle Fusion Workforce Compensation, Oracle Fusion Benefits y Oracle Fusion Performance Management. Además, ING implementará Oracle Fusion Financials, aprovechando la plataforma unificada de Oracle Fusion Applications. “Como institución financiera líder y mejor empleador, ING considera a sus colaboradores el activo más estratégico. Continuar siendo “best-in-class” en recursos humanos, es un diferenciador competitivo para nosotros, y necesitamos una aplicación líder de HCM para complementar nuestros esfuerzos ", dijo Marijke Brunklaus, Director General de Recursos Humanos de ING Bank Países Bajos. "Siendo un antiguo usuario de Oracle PeopleSoft HCM, cuidadosamente hemos evaluado las posibles soluciones disponibles y elegimos Oracle Fusion HCM como nuestra futura plataforma. Las profundas capacidades globales y las herramientas principales de talento de Oracle Fusion HCM son una buena opción para ayudarnos a evolucionar continuamente en nuestro negocio". Puede conocer más sobre HCM Fusion : · Oracle Fusion Applications · Oracle Fusion Human Capital Management · Oracle PartnerNetwork · Oracle Consulting Services · Oracle Human Capital Management Blog · Oracle HCM on Twitter · Oracle HCM on Facebook

    Read the article

  • Going for Gold

    - by Simple-Talk Editorial Team
    There was a spring in the step of some members of our development teams here at Red Gate, on hearing that on five gold awards at 2012′s SQL Mag Community and Editors Choice Awards. And why not? It’s a nice recognition that their efforts were appreciated by many in the SQL Server community. The team at Simple-Talk don’t tend to spring, but even we felt a twinge of pride in the fact that SQL Scripts Manager received Gold for Editor’s Choice in the Best Free Tools category. The tool began life as a “Down Tools” project and is one that we’ve supported and championed in various articles on Simple-talk.com. Over a Cambridge Bitter in the Waggon and Horses, we’ve often reflected on how nice it would be to nominate our own awards. Of course, we’d have to avoid nominating Red Gate tools in each category, even the free ones, for fear of seeming biased,  but we could still award other people’s free tools, couldn’t we? So allow us to set the stage for the annual Simple-Talk Community Tool awards… Onto the platform we shuffle, to applause from the audience; Chris in immaculate tuxedo, Alice in stunning evening gown, Dave and Tony looking vaguely uncomfortable, Andrew somehow distracted, as if his mind is elsewhere. Tony strides up to the lectern, and coughs lightly…”In the free-tool category we have the three nominations, and they are…” (rustle of the envelope opening) Ola Hallengren’s SQL Server Maintenance Solution (applause) Adam Machanic’s WhoIsActive (cheers, more applause) Brent Ozar’s sp_Blitz (much clapping) “Before we declare the winner, I’d like to say a few words in recognition of a grand tradition in a SQL Server community that continues to offer its members a steady supply of excellent, free tools. It hammers home the fundamental principle that a tool should solve a single, pressing and frustrating problem, but you should only ever build your own solution to that problem if you are certain that you cannot buy it, or that someone has not already provided it free. We have only three finalists tonight, but I feel compelled to mention a few other tools that we also use and appreciate, such as Microsoft’s Logparser, Open source Curl, Microsoft’s TableDiff.exe, Performance Analysis of Logs (PAL) Tool, SQL Server Cache Manager and SQLPSX.” “And now I’ll hand over to Alice to announce the winner.” Alice strides over to the microphone, tearing open the envelope. “The winner,” she pauses for dramatic effect “… is …Ola Hallengren’s SQL Server Maintenance Solution!” Queue much applause and consumption of champagne. Did we get it wrong? What free tool would you nominate? Let us know! Cheers, Simple-Talk Editorial Team (Andrew, Alice, Chris, Dave, Tony)

    Read the article

< Previous Page | 593 594 595 596 597 598 599 600 601 602 603 604  | Next Page >