Search Results

Search found 1774 results on 71 pages for 'parallel'.

Page 53/71 | < Previous Page | 49 50 51 52 53 54 55 56 57 58 59 60  | Next Page >

• Manic Monday - More OpenWorld Solaris Sessions: Developers, Cloud, Customer Insights, Hardware Optimization

  - by Larry Wake

  We're overflowing with Monday sessions; literally more than one person can take in. Learn more about what's new in Oracle Solaris Studio, hear about the latest x86 and SPARC hardware optimizations, get some insights on cloud deployment strategies, and find out from your peers what they're doing with Oracle Solaris. If you're an OpenWorld attendee, go to to Schedule Builder to guarantee your space in any session or lab. See yesterday's blog post and the "Focus on Oracle Solaris" guide for even more sessions. Monday, October 1st: 10:45 AM - Maximizing Your SPARC T4 Oracle Solaris Application Performance(CON6382,  Marriott Marquis - Golden Gate C3) Hear how customers and commercial software partners have reached peak performance on SPARC T4 servers and engineered systems with Oracle Solaris Studio and its latest tools for analyzing, reporting, and improving runtime performance: Autoparallelizing, high-performance compilers Performance Analyzer (used to find performance hotspots) Thread Analyzer (to expose data races and deadlocks) Code Analyzer (used to discover latent memory corruption issues) 10:45 Cloud Formation: Implementing IaaS in Practice with Oracle Solaris(CON8787, Moscone South 302) Decisions, decisions--at the same time, we've got a session that covers why Oracle Solaris is the ideal OS for public or private clouds, IaaS or PaaS, with built-in features for elastic infrastructure, unrivaled security, superfast installation and deployment, nonstop availability, and crystal-clear observability. This session will include a customer study on how Oracle Solaris is used in the cloud today to implement the Oracle stack. 12:15 PM - Customer Insight: Oracle Solaris on Oracle Exadata, Oracle Exalogic, and SPARC SuperCluster(CON8760, Moscone South 270) Hear from customers what benefits they have realized from using the Oracle stack on Oracle Exadata and Oracle’s SPARC SuperCluster and from using Oracle Solaris on those engineered systems, taking advantage of built-in lightweight OS virtualization (Zones), enterprise reliability and scale, and other key features. 1:45 PM - Case Study: Mobile Tornado Uses Oracle Technology for Better RAS and TCO?(CON4281, Moscone West 2005) Mobile Tornado develops and markets instant communication platforms, replacing traditional radio networks with cellular networks. Its critical concern is uptime. Find out how they've used Oracle Solaris, Netra SPARC T4, and Oracle Solaris Cluster, including Oracle Solaris ZFS and Zones, for their Oracle Database deployments to improve reliability and drive down cost. 3:15 PM - Technical Panel: Developing High Performance Applications on Oracle Solaris(CON7196, Marriott Marquis - Golden Gate C2) Engineers from the Oracle Solaris, Oracle Database, and Oracle Tuxedo development teams, and Oracle ISV Engineering discuss how they develop high-performance enterprise applications that take advantage of Oracle's SPARC and x86 servers, with Oracle Solaris Studio and new Oracle Solaris 11 features. Topics will include developer tools, parallel frameworks, best practices, and methodologies, as well as insights and case studies on parallelizing and optimizing application performance on Oracle Solaris. Bring your best questions! 3:15 PM -  x86 Power Management with Oracle Solaris: Current State, Opportunities, and Future(CON6271, Moscone West 2012) Another option for this time slot: learn about how Intel Xeon and Oracle Solaris work together to reduce server power consumption. This presentation addresses some of the recent power management improvements in Oracle Solaris, opportunities to further improve energy efficiency, and some future directions for Oracle Solaris power management.

  Read the article

• Hadoop, NOSQL, and the Relational Model

  - by Phil Factor

  (Guest Editorial for the IT Pro/SysAdmin Newsletter)Whereas Relational Databases fit the world of commerce like a glove, it is useless to pretend that they are a perfect fit for all human endeavours. Although, with SQL Server, we’ve made great strides with indexing text, in processing spatial data and processing markup, there is still a problem in dealing efficiently with large volumes of ephemeral semi-structured data. Key-value stores such as Cassandra, Project Voldemort, and Riak are of great value for ephemeral data, and seem of equal value as a data-feed that provides aggregations to an RDBMS. However, the Document databases such as MongoDB and CouchDB are ideal for semi-structured data for which no fixed schema exists; analytics and logging are obvious examples. NoSQL products, such as MongoDB, tackle the semi-structured data problem with panache. MongoDB is designed with a simple document-oriented data model that scales horizontally across multiple servers. It doesn’t impose a schema, and relies on the application to enforce the data structure. This is another take on the old ‘EAV’ problem (where you don’t know in advance all the attributes of a particular entity) It uses a clever replica set design that allows automatic failover, and uses journaling for data durability. It allows indexing and ad-hoc querying. However, for SQL Server users, the obvious choice for handling semi-structured data is Apache Hadoop. There will soon be an ODBC Driver for Apache Hive .and an Add-in for Excel. Additionally, there are now two Hadoop-based connectors for SQL Server; the Apache Hadoop connector for SQL Server 2008 R2, and the SQL Server Parallel Data Warehouse (PDW) connector. We can connect to Hadoop process the semi-structured data and then store it in SQL Server. For one steeped in the culture of Relational SQL Databases, I might be expected to throw up my hands in the air in a gesture of contempt for a technology that was, judging by the overblown journalism on the subject, about to make my own profession as archaic as the Saggar makers bottom knocker (a potter’s assistant who helped the saggar maker to make the bottom of the saggar by placing clay in a metal hoop and bashing it). However, on the contrary, I find that I'm delighted with the advances made by the NoSQL databases in the past few years. Having the flow of ideas from the NoSQL providers will knock any trace of complacency out of the providers of Relational Databases and inspire them into back-fitting some features, such as horizontal scaling, with sharding and automatic failover into SQL-based RDBMSs. It will do the breed a power of good to benefit from all this lateral thinking.

  Read the article

• Why would more CPU cores on virtual machine slow compile times?

  - by Sid

  [edit#2] If anyone from VMWare can hit me up with a copy of VMWare Fusion, I'd be more than happy to do the same as a VirtualBox vs VMWare comparison. Somehow I suspect the VMWare hypervisor will be better tuned for hyperthreading (see my answer too) I'm seeing something curious. As I increase the number of cores on my Windows 7 x64 virtual machine, the overall compile time increases instead of decreasing. Compiling is usually very well suited for parallel processing as in the middle part (post dependency mapping) you can simply call a compiler instance on each of your .c/.cpp/.cs/whatever file to build partial objects for the linker to take over. So I would have imagined that compiling would actually scale very well with # of cores. But what I'm seeing is: 8 cores: 1.89 sec 4 cores: 1.33 sec 2 cores: 1.24 sec 1 core: 1.15 sec Is this simply a design artifact due to a particular vendor's hypervisor implementation (type2:virtualbox in my case) or something more pervasive across more VMs to make hypervisor implementations more simpler? With so many factors, I seem to be able to make arguments both for and against this behavior - so if someone knows more about this than me, I'd be curious to read your answer. Thanks Sid [edit:addressing comments] @MartinBeckett: Cold compiles were discarded. @MonsterTruck: Couldn't find an opensource project to compile directly. Would be great but can't screwup my dev env right now. @Mr Lister, @philosodad: Have 8 hw threads, using VirtualBox, so should be 1:1 mapping without emulation @Thorbjorn: I have 6.5GB for the VM and a smallish VS2012 project - it's quite unlikely that I'm swapping in/out trashing the page file. @All: If someone can point to an open source VS2010/VS2012 project, that might be a better community reference than my (proprietary) VS2012 project. Orchard and DNN seem to need environment tweaking to compile in VS2012. I really would like to see if someone with VMWare Fusion also sees this (for VMWare vs VirtualBox compartmentalization) Test details: Hardware: Macbook Pro Retina CPU : Core i7 @ 2.3Ghz (quad core, hyper threaded = 8 cores in windows task manager) Memory : 16 GB Disk : 256GB SSD Host OS: Mac OS X 10.8 VM type: VirtualBox 4.1.18 (type 2 hypervisor) Guest OS: Windows 7 x64 SP1 Compiler: VS2012 compiling a solution with 3 C# Azure projects Compile times measure by VS2012 plugin called 'VSCommands' All tests run 5 times, first 2 runs discarded, last 3 averaged

  Read the article

• Implementing a wheeled character controller

  - by Lazlo

  I'm trying to implement Boxycraft's character controller in XNA (with Farseer), as Bryan Dysmas did (minus the jumping part, yet). My current implementation seems to sometimes glitch in between two parallel planes, and fails to climb 45 degree slopes. (YouTube videos in links, plane glitch is subtle). How can I fix it? From the textual description, I seem to be doing it right. Here is my implementation (it seems like a huge wall of text, but it's easy to read. I wish I could simplify and isolate the problem more, but I can't): public Body TorsoBody { get; private set; } public PolygonShape TorsoShape { get; private set; } public Body LegsBody { get; private set; } public Shape LegsShape { get; private set; } public RevoluteJoint Hips { get; private set; } public FixedAngleJoint FixedAngleJoint { get; private set; } public AngleJoint AngleJoint { get; private set; } ... this.TorsoBody = BodyFactory.CreateRectangle(this.World, 1, 1.5f, 1); this.TorsoShape = new PolygonShape(1); this.TorsoShape.SetAsBox(0.5f, 0.75f); this.TorsoBody.CreateFixture(this.TorsoShape); this.TorsoBody.IsStatic = false; this.LegsBody = BodyFactory.CreateCircle(this.World, 0.5f, 1); this.LegsShape = new CircleShape(0.5f, 1); this.LegsBody.CreateFixture(this.LegsShape); this.LegsBody.Position -= 0.75f * Vector2.UnitY; this.LegsBody.IsStatic = false; this.Hips = JointFactory.CreateRevoluteJoint(this.TorsoBody, this.LegsBody, Vector2.Zero); this.Hips.MotorEnabled = true; this.AngleJoint = new AngleJoint(this.TorsoBody, this.LegsBody); this.FixedAngleJoint = new FixedAngleJoint(this.TorsoBody); this.Hips.MaxMotorTorque = float.PositiveInfinity; this.World.AddJoint(this.Hips); this.World.AddJoint(this.AngleJoint); this.World.AddJoint(this.FixedAngleJoint); ... public void Move(float m) // -1, 0, +1 { this.Hips.MotorSpeed = 0.5f * m; }

  Read the article

• Java JRE 1.6.0_37 Certified with Oracle E-Business Suite

  - by Steven Chan (Oracle Development)

  My apologies: this certification announcement got lost in the OpenWorld maelstorm.  Better late than never. The section below entitled, "All JRE 1.6 releases are certified with EBS upon release" should obviate the need for these announcements, but I know that people have gotten used to seeing these certifications referenced explicitly.  The latest Java Runtime Environment 1.6.0_37 (a.k.a. JRE 6u37-b06) is now certified with Oracle E-Business Suite Release 11i and 12 desktop clients.   What's new in Java 1.6.0_37?See the 1.6.0_37 Update Release Notes for details about what has changed in this release.  This release is available for download from the usual Sun channels and through the 'Java Automatic Update' mechanism. 32-bit and 64-bit versions certified This certification includes both the 32-bit and 64-bit JRE versions. 32-bit JREs are certified on: Windows XP Service Pack 3 (SP3) Windows Vista Service Pack 1 (SP1) and Service Pack 2 (SP2) Windows 7 and Windows 7 Service Pack 1 (SP1) 64-bit JREs are certified only on 64-bit versions of Windows 7 and Windows 7 Service Pack 1 (SP1). Worried about the 'mismanaged session cookie' issue? No need to worry -- it's fixed.  To recap: JRE releases 1.6.0_18 through 1.6.0_22 had issues with mismanaging session cookies that affected some users in some circumstances. The fix for those issues was first included in JRE 1.6.0_23. These fixes will carry forward and continue to be fixed in all future JRE releases.  In other words, if you wish to avoid the mismanaged session cookie issue, you should apply any release after JRE 1.6.0_22.All JRE 1.6 releases are certified with EBS upon release Our standard policy is that all E-Business Suite customers can apply all JRE updates to end-user desktops from JRE 1.6.0_03 and later updates on the 1.6 codeline.  We test all new JRE 1.6 releases in parallel with the JRE development process, so all new JRE 1.6 releases are considered certified with the E-Business Suite on the same day that they're released by our Java team.  You do not need to wait for a certification announcement before applying new JRE 1.6 releases to your EBS users' desktops. Important For important guidance about the impact of the JRE Auto Update feature on JRE 1.6 desktops, see: Planning Bulletin for JRE 7: What EBS Customers Can Do Today References Recommended Browsers for Oracle Applications 11i (Metalink Note 285218.1) Upgrading Sun JRE (Native Plug-in) with Oracle Applications 11i for Windows Clients (Metalink Note 290807.1) Recommended Browsers for Oracle Applications 12 (MetaLink Note 389422.1) Upgrading JRE Plugin with Oracle Applications R12 (MetaLink Note 393931.1) Related Articles Mismanaged Session Cookie Issue Fixed for EBS in JRE 1.6.0_23 Roundup: Oracle JInitiator 1.3 Desupported for EBS Customers in July 2009

  Read the article

• How employable am I as a programmer?

  - by dsimcha

  I'm currently a Ph.D. student in Biomedical Engineering with a concentration in computational biology and am starting to think about what I want to do after graduate school. I feel like I've accumulated a lot of programming skills while in grad school, but taken a very non-traditional path to learning all this stuff. I'm wondering whether I would have an easy time getting hired as a programmer and could fall back on that if I can't find a good job directly in my field, and if so whether I would qualify for a more prestigious position than "code monkey". Things I Have Going For Me Approximately 4 years of experience programming as part of my research. I believe I have a solid enough grasp of the fundamentals that I could pick up new languages and technologies pretty fast, and could demonstrate this in an interview. Good math and statistics skills. An extensive portfolio of open source work (and the knowledge that working on these projects implies): I wrote a statistics library in D, mostly from scratch. I wrote a parallelism library (parallel map, reduce, foreach, task parallelism, pipelining, etc.) that is currently in review for adoption by the D standard library. I wrote a 2D plotting library for D against the GTK Cairo backend. I currently use it for most of the figures I make for my research. I've contributed several major performance optimizations to the D garbage collector. (Most of these were low-hanging fruit, but it still shows my knowledge of low-level issues like memory management, pointers and bit twiddling.) I've contributed lots of miscellaneous bug fixes to the D standard library and could show the change logs to prove it. (This demonstrates my ability read other people's code.) Things I Have Going Against Me Most of my programming experience is in D and Python. I have very little to virtually no experience in the more established, "enterprise-y" languages like Java, C# and C++, though I have learned a decent amount about these languages from small, one-off projects and discussions about language design in the D community. In general I have absolutely no knowledge of "enterprise-y" technlogies. I've never used a framework before, possibly because most reusable code for scientific work and for D tends to call itself a "library" instead. I have virtually no formal computer science/software engineering training. Almost all of my knowledge comes from talking to programming geek friends, reading blogs, forums, StackOverflow, etc. I have zero professional experience with the official title of "developer", "software engineer", or something similar.

  Read the article

• "Hello World" in C++ AMP

  - by Daniel Moth

  Some say that the equivalent of "hello world" code in the data parallel world is matrix multiplication :) Below is the before C++ AMP and after C++ AMP code. For more on what it all means, watch the recording of my C++ AMP introduction (the example below is part of the session). void MatrixMultiply(vector<float>& vC, const vector<float>& vA, const vector<float>& vB, int M, int N, int W ) { for (int y = 0; y < M; y++) { for (int x = 0; x < N; x++) { float sum = 0; for(int i = 0; i < W; i++) { sum += vA[y * W + i] * vB[i * N + x]; } vC[y * N + x] = sum; } } } Change the function to use C++ AMP and hence offload the computation to the GPU, and now the calling code (which I am not showing) needs no changes and the overall operation gives you really nice speed up for large datasets…  #include <amp.h> using namespace concurrency; void MatrixMultiply(vector<float>& vC, const vector<float>& vA, const vector<float>& vB, int M, int N, int W ) { array_view<const float,2> a(M, W, vA); array_view<const float,2> b(W, N, vB); array_view<writeonly<float>,2> c(M, N, vC); parallel_for_each( c.grid, [=](index<2> idx) mutable restrict(direct3d) { float sum = 0; for(int i = 0; i < a.x; i++) { sum += a(idx.y, i) * b(i, idx.x); } c[idx] = sum; } ); } Again, you can understand the elements above, by using my C++ AMP presentation slides and recording… Stay tuned for more… Comments about this post welcome at the original blog.

  Read the article

• Solaris 11

  - by user9154181

  Oracle has a strict policy about not discussing product features until they appear in shipping product. Now that Solaris 11 is publically available, it is time to catch up. I will be shortly posting articles on a variety of new developments in the Solaris linkers and related bits: 64-bit Archives After 40+ years of Unix, the archive file format has run out of room. The ar and link-editor (ld) commands have been enhanced to allow archives to grow past their previous 32-bit limits. Guidance The link-editor is now willing and able to tell you how to alter your link lines in order to build better objects. Stub Objects This is one of the bigger projects I've undertaken since joining the Solaris group. Stub objects are shared objects, built entirely from mapfiles, that supply the same linking interface as the real object, while containing no code or data. You can link to them, but cannot use them at runtime. It was pretty simple to add this ability to the link-editor, but the changes to the OSnet in order to apply them to building Solaris were massive. I discuss how we came to invent stub objects, how we apply them to build the OSnet in a more parallel and scalable manner, and about the follow on opportunities that have emerged from the new stub proto area we created to hold them. The elffile Utility A new standard Solaris utility, elffile is a variant of the file utility, focused exclusively on linker related files. elffile is of particular value for examining archives, as it allows you to find out what is inside them without having to first extract the archive members into temporary files. This release has been a long time coming. I joined the Solaris group in late 2005, and this will be my first FCS. From a user perspective, Solaris 11 is probably the biggest change to Solaris since Solaris 2.0. Solaris 11 polishes the ground breaking features from Solaris 10 (DTrace, FMA, ZFS, Zones), and uses them to add a powerful new packaging system, numerous other enhacements and features, along with a huge modernization effort. I'm excited to see it go out into the world. I hope you enjoy using it as much as we did creating it. Software is never done. On to the next one...

  Read the article

• How to keep a data structure synchronized over a network?

  - by David Gouveia

  Context In the game I'm working on (a sort of a point and click graphic adventure), pretty much everything that happens in the game world is controlled by an action manager that is structured a bit like: So for instance if the result of examining an object should make the character say hello, walk a bit and then sit down, I simply spawn the following code: var actionGroup = actionManager.CreateGroup(); actionGroup.Add(new TalkAction("Guybrush", "Hello there!"); actionGroup.Add(new WalkAction("Guybrush", new Vector2(300, 300)); actionGroup.Add(new SetAnimationAction("Guybrush", "Sit")); This creates a new action group (an entire line in the image above) and adds it to the manager. All of the groups are executed in parallel, but actions within each group are chained together so that the second one only starts after the first one finishes. When the last action in a group finishes, the group is destroyed. Problem Now I need to replicate this information across a network, so that in a multiplayer session, all players see the same thing. Serializing the individual actions is not the problem. But I'm an absolute beginner when it comes to networking and I have a few questions. I think for the sake of simplicity in this discussion we can abstract the action manager component to being simply: var actionManager = new List<List<string>>(); How should I proceed to keep the contents of the above data structure syncronized between all players? Besides the core question, I'm also having a few other concerns related to it (i.e. all possible implications of the same problem above): If I use a server/client architecture (with one of the players acting as both a server and a client), and one of the clients has spawned a group of actions, should he add them directly to the manager, or only send a request to the server, which in turn will order every client to add that group? What about packet losses and the like? The game is deterministic, but I'm thinking that any discrepancy in the sequence of actions executed in a client could lead to inconsistent states of the world. How do I safeguard against that sort of problem? What if I add too many actions at once, won't that cause problems for the connection? Any way to alleviate that?

  Read the article

• Setting up Ubuntu on my mother's computer

  - by idealmachine

  Intended use My mother had an old Compaq desktop computer running Windows 98, which she used for occasional Web browsing and playing cards. The name of her card game is Hoyle Card Games 3. Although I had to repair it several times over the last 10 years, it worked fine until it finally died at the end of last year. Hardware specifications A relative brought up a newer computer soon afterward: Operating system: Windows XP Asus K8N motherboard (with broken on-board sound; getting a sound card) Athlon 64? processor (don't remember the clock speed) 512 MB RAM Hope the graphics card works... Replacement sound card will be one of: Ensoniq ES1370 AudioPCI Diamond Monster Sound MX300 (Aureal chipset) Sound Blaster Audigy 2 SE Peripherals HP Scanjet 3400c scanner (USB connected) HP LaserJet multi-function printer (parallel port connected, and printing works with a PCL driver) Same serial mouse as old computer Question I had set up an SSH/VNC connection to allow for remotely working out problems. Or so I thought. A month later, the computer would not boot, rendering the SSH connection useless and an OS reinstall necessary. Unfortunately, I have neither the original Windows disc nor the product key. Unless I were to pay $200 for a full Windows 7 Home Premium license for my computer, I would not be able to re-install Windows XP on hers. I consider myself an advanced Linux user, having used Debian for years. So here are my questions. I have only one day to decide whether to use Ubuntu or buy Windows: A quick search leads me to believe all the hardware listed above is supposed to work with Linux, but am I mistaken? Would Ubuntu/Xubuntu suffice (specify which one if it matters), or would I be better off paying the $200 necessary for Windows XP? Is the card game likely to run on Wine? I believe the minimum system requirement is Windows 95. Failing Wine compatibility, will VirtualBox run fast enough on such a computer (Windows 98 as the guest OS)? Are there any free card games just as good? She plays mainly Bridge, Poker, and Solitaire. Is there any "Large Fonts" option for those with poor vision? The lack of it would be a big disadvantage. BONUS: Although I would probably replace the old mouse upon a move to Ubuntu, is it even possible to get a serial mouse working?

  Read the article

• New RUP Patch for iSupplier Portal, Sourcing and Supplier Lifecycle Management (SLM)

  - by LuciaC

  Just released - the 12.1.3 Rollup (RUP) Patch 17525552:R12.PRC_PF.B for iSupplier Portal, Sourcing and Supplier Lifecycle Management (SLM). Who should apply this patch? Anyone that is on Release 12.1.3 and is using  iSupplier Portal, Sourcing or Supplier Lifecycle Management (SLM) functionality. The following areas have had major fixes: Prospective Supplier Guided Navigation: The train-navigation is introduced for prospective supplier registration so that prospective suppliers can see all steps needed to successfully register themselves. Supplier Registration Workflow Enhancement: With this release, provided the Approval Management Engine (AME) action required notifications for supplier approval, so that all workflow related features can be enabled. Vacation rules can be set, approvals can be forwarded and more information can be requested through the notification itself.  Additionally AME parallel Approval support for Supplier Registration approvals has been added. Reinstate Supplier Request: Allow buyer to reopen/reinstate the rejected supplier. Supplier is able to access their previously rejected registration again and make changes and resubmit request. Contact Address Association: The prospective supplier is allowed to associate addresses with contacts (including Primary) during the prospective supplier registration process. Primary Contact Enhancement: The prospective supplier can be registered without creating a user account for the primary contact. Mandatory Attributes: In the negotiation requirement creation page, the lookup meaning of 'Internal' has been changed to 'Internal Optional', and a new lookup value with meaning as 'Internal Required' has been added. The values available in the 'Type' dropdown now are Display Only, Internal Optional, Internal Required, Supplier Optional and Supplier Required.  So now during supplier evaluations, internal user response can be set as mandatory by using Internal Required type during requirement creation. Notifications to Supplier:  When the supplier saves and submits their supplier registration request, then a notification with a registration status page link will be sent for further access.  When the buyer approves, rejects or returns the request, the supplier will be notified in an email with the current status. There are also 10 major enhancements included in this RUP. For information about this RUP; including, the fixes and enhancements included, how to access and apply the patch, performing an impact analysis on your system, and testing recommendations, see Doc ID 1591198.1.  Don’t delay apply the patch today!

  Read the article

• Java @Contented annotation to help reduce false sharing

  - by Dave

  See this posting by Aleksey Shipilev for details -- @Contended is something we've wanted for a long time. The JVM provides automatic layout and placement of fields. Usually it'll (a) sort fields by descending size to improve footprint, and (b) pack reference fields so the garbage collector can process a contiguous run of reference fields when tracing. @Contended gives the program a way to provide more explicit guidance with respect to concurrency and false sharing. Using this facility we can sequester hot frequently written shared fields away from other mostly read-only or cold fields. The simple rule is that read-sharing is cheap, and write-sharing is very expensive. We can also pack fields together that tend to be written together by the same thread at about the same time. More generally, we're trying to influence relative field placement to minimize coherency misses. Fields that are accessed closely together in time should be placed proximally in space to promote cache locality. That is, temporal locality should condition spatial locality. Fields accessed together in time should be nearby in space. That having been said, we have to be careful to avoid false sharing and excessive invalidation from coherence traffic. As such, we try to cluster or otherwise sequester fields that tend to written at approximately the same time by the same thread onto the same cache line. Note that there's a tension at play: if we try too hard to minimize single-threaded capacity misses then we can end up with excessive coherency misses running in a parallel environment. Theres no single optimal layout for both single-thread and multithreaded environments. And the ideal layout problem itself is NP-hard. Ideally, a JVM would employ hardware monitoring facilities to detect sharing behavior and change the layout on the fly. That's a bit difficult as we don't yet have the right plumbing to provide efficient and expedient information to the JVM. Hint: we need to disintermediate the OS and hypervisor. Another challenge is that raw field offsets are used in the unsafe facility, so we'd need to address that issue, possibly with an extra level of indirection. Finally, I'd like to be able to pack final fields together as well, as those are known to be read-only.

  Read the article

• How to avoid oscillation by async event based systems?

  - by inf3rno

  Imagine a system where there are data sources which need to be kept in sync. A simple example is model - view data binding by MVC. Now I intend to describe these kind of systems with data sources and hubs. Data sources are publishing and subscribing for events and hubs are relaying events to data sources. By handling an event a data source will change it state described in the event. By publishing an event the data source puts its current state to the event, so other data sources can use that information to change their state accordingly. The only problem with this system, that events can be reflected from the hub or from the other data sources, and that can put the system into an infinite oscillation (by async or infinite loop by sync). For example A -- data source B -- data source H -- hub A -> H -> A -- reflection from the hub A -> H -> B -> H -> A -- reflection from another data source By sync it is relatively easy to solve this issue. You can compare the current state with the event, and if they are equal, you don't change the state and raise the same event again. By async I could not find a solution yet. The state comparison does not work by async event handling because there is eventual consistency, and new events can be published in an inconsistent state causing the same oscillation. For example: A(*->x) -> H -> B(y->x) -- can go parallel with B(*->y) -> H -> A(x->y) -- so first A changes to x state while B changes to y state -- then B changes to x state while A changes to y state -- and so on for eternity... What do you think is there an algorithm to solve this problem? If there is a solution, is it possible to extend it to prevent oscillation caused by multiple hubs, multiple different events, etc... ? update: I don't think I can make this work without a lot of effort. I think this problem is just the same as we have by syncing multiple databases in a distributed system. So I think what I really need is constraints if I want to prevent this problem in an automatic way. What constraints do you suggest?

  Read the article

• Cloud Computing Business Benefits

  - by workflowman

  If you have been living under a rock for the past year, you wouldn't have heard about cloud computing. Cloud computing is a loose term that describes anything that is hosted in data centers and accessed via the internet. It is normally associated with developers who draw clouds in diagrams indicating where services or how systems communicate with each other. Cloud computing also incorporates such well-known trends as Web 2.0 and Software as a Service (SaaS) and more recently Infrastructure as a Service (IaaS) and Platform as a Service (PaaS). Its aim is to change the way we compute, moving from traditional desktop and on-premises servers to services and resources that are hosted in the cloud.  Benefits of Cloud Computing  There are clearly benefits in building applications using cloud computing, some of which are listed here:  Zero up- front investment:  Delivering a large-scale system costs a fortune in both time and money. Often IT departments are split into hardware/network and software services. The hardware team provisions servers and so forth under the requirements of the software team. Often the hardware team has a different budget that requires approval. Although hardware and software management are two separate disciplines, sometimes what happens is developers are given the task to estimate CPU cycles, disk space, and so forth, which ends up in underutilized servers.  Usage-based costing:  You pay for what you use, no more, no less, because you never actually own the server. This is similar to car leasing, where in the long run you get a new car every three years and maintenance is never a worry.  Potential for shrinking the processing time:  If processes are split over multiple machines, parallel processing is performed, which decreases processing time.  More office space:  Walk into most offices, and guaranteed you will find a medium- sized room dedicated to servers.  Efficient resource utilization:  The resource utilization is handed by a centralized cloud administrator who is in charge of deciding exactly the right amount of resources for a system. This takes the task away from local administrators, who have to regularly monitor these servers.  Just-in-time infrastructure:  If your system is a success and needs to scale to meet demand, this can cause further time delays or a slow- performing service. Cloud computing solves this because you can add more resources at any time.  Lower environmental impact:  If servers are centralized, potentially an environment initiative is more likely to succeed. As an example, if servers are placed in sunny or windy parts of the world, then why not use these resources to power those servers?  Lower costs:  Unfortunately, this is one point that administrators will not like. If you have people administrating your e-mail server and network along with support staff doing other cloud-based tasks, this workforce can be reduced. This saves costs, though it also reduces jobs.

  Read the article

• IndexOutOfRangeException on World.Step after enabling/disabling a Farseer physics body?

  - by WilHall

  Earlier, I posted a question asking how to swap fixtures on the fly in a 2D side-scroller using Farseer Physics Engine. The ultimate goal being that the player's physical body changes when the player is in different states (I.e. standing, walking, jumping, etc). After reading this answer, I changed my approach to the following: Create a physical body for each state when the player is loaded Save those bodies and their corresponding states in parallel lists Swap those physical bodies out when the player state changes (which causes an exception, see below) The following is my function to change states and swap physical bodies: new protected void SetState(object nState) { //If mBody == null, the player is being loaded for the first time if (mBody == null) { mBody = mBodies[mStates.IndexOf(nState)]; mBody.Enabled = true; } else { //Get the body for the given state Body nBody = mBodies[mStates.IndexOf(nState)]; //Enable the new body nBody.Enabled = true; //Disable the current body mBody.Enabled = false; //Copy the current body's attributes to the new one nBody.SetTransform(mBody.Position, mBody.Rotation); nBody.LinearVelocity = mBody.LinearVelocity; nBody.AngularVelocity = mBody.AngularVelocity; mBody = nBody; } base.SetState(nState); } Using the above method causes an IndexOutOfRangeException when calling World.Step: mWorld.Step(Math.Min((float)nGameTime.ElapsedGameTime.TotalSeconds, (1f / 30f))); I found that the problem is related to changing the .Enabled setting on a body. I tried the above function without setting .Enabled, and there was no error thrown. Turning on the debug views, I saw that the bodies were updating positions/rotations/etc properly when the state was changes, but since they were all enabled, they were just colliding wildly with each other. Does Enabling/Disabling a body remove it from the world's body list, which then causes the error because the list is shorter than expected? Update: For such a straightforward issue, I feel this question has not received enough attention. Has anyone else experienced this? Would anyone try a quick test case? I know this issue can be sidestepped - I.e. by not disabling a body during the simulation - but it seems strange that this issue would exist in the first place, especially when I see no mention of it in the documentation for farseer or box2d. I can't find any cases of the issue online where things are more or less kosher, like in my case. Any leads on this would be helpful.

  Read the article

• blu-ray archiving in vmware ESXi 4

  - by spacecadet77

  Hi, I need some advice about using blu-ray writer for archiving data on vmware ESXi 4. At office we have IBM System x3400 Tower server with ESXi 4 hipervisor and OpenSuse and CentOS GNU/Linux system as guests. Will blu-ray writer work in this setup, and if it will is there any particular model you can suggest. Best regards IBM System x3400 Tower server specification: 1x Intel Quad-Core Xeon E5410 2.33GHz/ 12MB/ 1333MHz (2x CPU max) Intel 5000P chipset, 2x 1GB PC2-5300 DDR2 667MHz SDRAM ECC Chipkill (32GB max) 2x4GB (2x2GB) PC2-5300 CL5 ECC DDR2 FBDIMM (x3400, x3550, x3650) SAS/SATA Hot-Swap Open Bay (0xHDD std, 4xHDD max, 8xHDD optional) ServeRAID 8K dual channel SAS/SATA controller (RAID 0,1,1E,10,5,6, 256MB, Battery Backup) Graphics ATI® RN50(ES1000) 16MB DDR, CD-RW/DVD Combo no FDD GigaEthernet, Tower with Power Supply 835W (opt Redudant) Slot 1: half-length, PCI-Express x8(x4 electrical) Slot 2: full, PCI-Express x8 Slot 3: full, PCI-Express x8 Slot 4: full, 64-bit 133MHz 3.3v PCI-X Slot 5: full, 64-bit 133MHz 3.3v PCI-X , Slot 6: half-length, 32-bit 33MHz 5.0v PCI ports: 4x USB (Vers 2.0), 2x PS/2, parallel, 2x serial (9-pin), VGA, RJ-45 (ethernet ), RJ-45 (sys mgm) HDD 4 x TB 7200rpm / Serial ATA II 3.0Gb/s / 16MB, RoHS

  Read the article

• NetBSD Networking

  - by Mike

  Hello, everyone! I have an old Toshiba Satellite 4015CDT, with Pentium II MMX, 32MB RAM, 4GB HDD. It also has one USB 1.0 port, parallel and serial ports, a 3.5" floppy drive and a CD-ROM drive (probably almost dead). I've installed NetBSD on this machine (full install) and now I want to connect it to the Internet. Although it has one PCMCIA modem card in it, it is obviously not an option. I've narrowed down my available options to the following: Connect a USB wireless adapter. I have a Realtek RTL8192 adapter, but although the system gives me the device's name (through dmesg) it fails to recognize it as a network adapter. Connect the laptop to another machine through a serial connection, so that the other machine will serve as a bridge to the Internet. Here is the page from the NetBSD documentation on serial connections. Unfortunately, I was not able to find anything on bridging there. I would very much appreciate your suggestions on the topic. Thanks in advance.

  Read the article

• How do I improve my incremental-backup performance?

  - by Alistair Bell

  I'm currently using the traditional rsync+cp -al method to create incremental/snapshot backups of our server tree. The backups are going onto a pair of eight-disk towers connected to the backup machine (a Sandy Bridge machine with 16 GB of RAM, running CentOS 5.5) via four eSATA connections (four disks per connection). Each disk is a regular 2 TB disk, so we have 32 TB of disk space connected to the backup machine. We're backing up about 20 TB of data on the servers with this. The problem is that each daily backup is taking more than 24 hours, and the real time-killer isn't the actual rsync, but the time it takes to perform a cp -al of the tree locally on the backup machine. It's taking more than 12 hours just to make the shadow copy of the tree, and as far as I can tell the performance backlog is at the disk (top shows the cp using a lot of RAM but not a lot of CPU and mostly in uninterruptible-sleep state) We have the server data split into four major volumes (and a few minor ones), and each of these backups runs in parallel (with some offsets in the cron to try to get some disks' cp done first). There are two volumes on the backup drive, both striped LVM volumes of 16 TB each. So obviously I need to improve the performance because it's unusable as it stands. The first question is: when CentOS 6 comes out, with support for btrfs, will making snapshots of subvolumes with btrfs substantially increase this performance? The second is: is there a way, with ext3 or something else supported in CentOS 5 or 6, to 'encourage' it to put the directories/inodes in one part of a volume (which could happen to be the part that's on an SSD, via LVM) and the files in another? That would presumably solve the problem, but I don't know of ways to hint ext3 like that.

  Read the article

• Computer speakers receive radio station signal

  - by squircle

  I have a set of Logitech 5.1 speakers where each speaker and the source plug into the subwoofer. I'm using a Griffin Firewave with output from my MacBook Pro, and output from my custom-built desktop with a switch in the middle (built it myself out of an old Belkin A/B parallel switch). Recently, I've noticed that I can hear a local Punjabi radio station being picked up by my speakers, and the volume of this interference increases as I increase the volume of the speakers. I'm fairly sure that this radio station is at the low-end of the FM spectrum, below 90MHz (or it may be at the high end, above 105MHz, my memory isn't infallible). It gets quite annoying as I can't put my audio very loud without the interference. I've tried to put a ferrite core on the input cable just before the 3.5mm jacks plug into the subwoofer. I don't know if putting the same core around all three of the cables (green, black, orange) would negate the effects, but I'm assuming not. There has been no change. Is there any reason why this would be happening? I'm assuming the interference is coming somewhere between the FireWave and the subwoofer, because the noise gets amplified with volume increases. If anybody has any suggestions, I'd be grateful!

  Read the article

• ReadyBoost in Windows 7

  - by Robert Koritnik

  I've bought an SD card today for my phot frame, but when I inserted it into my notebook I saw I could use it for ReadyBoost. Some background I'm a .net developer, using VMs and developing web applications (and Sharepoint). I use an HP notebook machine with Core 2 Duo 2GHz + 4GB RAM + 320 7200 HD. I simultaneously run Visual Studio 2010 with some plugins SQL Server Firefox with at least 10 tabs Chrome with about 5 tabs IIS VM with Server 2008 machine Sharepoint and occasionally also Photoshop and some InDesign as well. So I don't let my machine have a break. :D Question If I buy myself some really fast SDHC card (like SanDisk 16GB Extreme 30MB/s - is there anything faster) and use it with my Windows 7 ReadyBoost, will I see any performance gain? Is it going to work something similar to Seagate's HybridDrive Momentus with 4GB of solid state drive? What could I actually expect if I do put this card into my machine? And what would be recommended size? Observations I guess redirecting page file to it would speed up the system. Some VM machines on it would probably run faster as well because they could run parallel to HD host system I guess. Am I right or wrong?

  Read the article

• Firefox and Chrome keeps forcing HTTPS on Rails app using nginx/Passenger

  - by Steve

  I've got a really weird problem here where every time I try to browse my Rails app in non-SSL mode Chrome (v16) and Firefox (v7) keeps forcing my website to be served in HTTPS. My Rails application is deployed on a Ubuntu VPS using Capistrano, nginx, Passenger and a wildcard SSL certificate. I have set these parameters for port 80 in the nginx.conf: passenger_set_cgi_param HTTP_X_FORWARDED_PROTO http; passenger_set_cgi_param HTTPS off; The long version of my nginx.conf can be found here: https://gist.github.com/2eab42666c609b015bff The ssl-redirect.include file contains: rewrite ^/sign_up https://$host$request_uri? permanent ; rewrite ^/login https://$host$request_uri? permanent ; rewrite ^/settings/password https://$host$request_uri? permanent ; It is to make sure those three pages use HTTPS when coming from non-SSL request. My production.rb file contains this line: # Enable HTTP and HTTPS in parallel config.middleware.insert_before Rack::Lock, Rack::SSL, :exclude => proc { |env| env['HTTPS'] != 'on' } I have tried redirecting to HTTP via nginx rewrites, Ruby on Rails redirects and also used Rails view url using HTTP protocol. My application.rb file contains this methods used in a before_filter hook: def force_http if Rails.env.production? if request.ssl? redirect_to :protocol => 'http', :status => :moved_permanently end end end Every time I try to redirect to HTTP non-SSL the browser attempts to redirect it back to HTTPS causing an infinite redirect loop. Safari, however, works just fine. Even when I've disabled serving SSL in nginx the browsers still try to connect to the site using HTTPS. I should also mention that when I pushed my app on to Heroku, the Rails redirect work just fine for all browsers. The reason why I want to use non-SSL is that my homepage contains non-secure dynamic embedded objects and a non-secure CDN and I want to prevent security warnings. I don't know what is causing the browser to keep forcing HTTPS requests.

  Read the article

• Postfix aliases and duplicate e-mails, how to fix?

  - by macke

  I have aliases set up in postfix, such as the following: [email protected]: [email protected], [email protected] ... When an email is sent to [email protected], and any of the recipients in that alias is cc:ed which is quite common (ie: "Reply all"), the e-mail is delivered in duplicates. For instance, if an e-mail is sent to [email protected] and [email protected] is cc:ed, it'll get delivered twice. According to the Postfix FAQ, this is by design as Postfix sends e-mail in parallel without expanding the groups, which makes it faster than sendmail. Now that's all fine and dandy, but is it possible to configure Postfix to actually remove duplicate recipients before sending the e-mail? I've found a lot of posts from people all over the net that has the same problem, but I have yet to find an answer. If this is not possible to do in Postfix, is it possible to do it somewhere on the way? I've tried educating my users, but it's rather futile I'm afraid... I'm running postfix on Mac OS X Server 10.6, amavis is set as content_filter and dovecot is set as mailbox_command. I've tried setting up procmail as a content_filter for smtp delivery (as per the suggestion below), but I can't seem to get it right. For various reasons, I can't replace the standard OS X configuration, meaning postfix, amavis and dovecot stay put. I can however add to it if I wish.

  Read the article

• Unable to activate Windows XP

  - by Josh Kelley

  The latest round of Patch Tuesday updates left my Windows XP computer unbootable. ("Fatal System Error: The Windows Logon Process system process terminated unexpectedly.") After much messing around with the recovery console, an XP CD's repair mode, and manually copying registry files around, I have a system that can boot again. However, I overwrote my OEM XP installation's activation information while trying to run a retail XP CD's setup, so it needs reactivation. Here's my problem: I cannot activate it at all. I log in, Windows tells me I have to activate to continue, I click Yes, and absolutely nothing happens: no windows, no response to keyboard or mouse, no response to Ctrl-Alt-Del, nothing. Safe mode works, but I can't activate in safe mode (EDIT: not even safe mode with networking). I read a trick online of pressing [Windows Key]+U to bring up the Microsoft Narrator, and that works, but clicking its Microsoft Web Site link does nothing. My last attempt to resolve this was to reinstall Windows off of the OEM CD. Now I have two parallel Windows installations, both on the same hard drive, one with all of my stuff and no way to activate it, one fully activated with no usable programs. Any ideas? Any way to activate in safe mode? Any way to copy activation information from my activated installation to my unactivated installation (since they're both on the same hard drive)?

  Read the article

• Ubuntu 10.04 Keyboard and Mouse Freezing Problem

  - by nitbuntu

  I had a partition setup with Windows XP and Ubuntu 8.04 dual booting. I recently upgraded to Ubuntu 10.04 by installing fresh from CD but leaving the previous /home folder as is. Things seemed to be working fine, but started finding that my mouse and keyboard were freezing. After a quick search on the internet, I found the following suggestions as shown here:- Ubuntu Forums Here the suggestion was to:- Edit /etc/default/grub, go to the line that begins like: GRUB_CMDLINE_LINUX_DEFAULT= Change it to: GRUB_CMDLINE_LINUX_DEFAULT="quiet splash acpi=off" After that, run this command: sudo update-grub and Reboot This seemed to have resolved the issue but after a couple of days I again find my mouse and keyboard freezing. I also find that my parallel port printer had also stopped working. I have saved the output of dmesg and my syslog. The first can be viewed here but the syslog had too many characters, so if someone can suggest an alternative to freetexthost, I can post it there. Moreover, if there is any other information that should be provided, do let me know. I do hope we can get to the bottom of this issue. Thank you in advance for any help that could be provided.

  Read the article

• Tomcat web application intermittent freeze

  - by tinny

  I have a Grails web application (just a standard war file) deployed on a Ubuntu 10.10 server running on tomcat 6. My database is postgresql. The problem is that every so often (once or twice a day after inactivity) when I try to log into this web application it just freezes. I can navigate to the login page but when I try and login (first time the DB is hit, might be a clue..?) the application just freezes indefinitely, no 500 response code... the browser just waits and waits. I followed the instructions detailed here because the problem described sounded the same as mine. My GC logging showed no long running GC, all sub sec. When the application freezes a jmap heap output is... using parallel threads in the new generation. using thread-local object allocation. Concurrent Mark-Sweep GC Heap Configuration: MinHeapFreeRatio = 40 MaxHeapFreeRatio = 70 MaxHeapSize = 536870912 (512.0MB) NewSize = 21757952 (20.75MB) MaxNewSize = 87228416 (83.1875MB) OldSize = 65404928 (62.375MB) NewRatio = 7 SurvivorRatio = 8 PermSize = 21757952 (20.75MB) MaxPermSize = 85983232 (82.0MB) Heap Usage: New Generation (Eden + 1 Survivor Space): capacity = 19595264 (18.6875MB) used = 11411976 (10.883308410644531MB) free = 8183288 (7.804191589355469MB) 58.23843965562291% used Eden Space: capacity = 17432576 (16.625MB) used = 9249296 (8.820816040039062MB) free = 8183280 (7.8041839599609375MB) 53.05754009046053% used From Space: capacity = 2162688 (2.0625MB) used = 2162680 (2.0624923706054688MB) free = 8 (7.62939453125E-6MB) 99.99963008996212% used To Space: capacity = 2162688 (2.0625MB) used = 0 (0.0MB) free = 2162688 (2.0625MB) 0.0% used concurrent mark-sweep generation: capacity = 101556224 (96.8515625MB) used = 83906080 (80.01907348632812MB) free = 17650144 (16.832489013671875MB) 82.62032270912317% used Perm Generation: capacity = 85983232 (82.0MB) used = 62866832 (59.95448303222656MB) free = 23116400 (22.045516967773438MB) 73.1152232100324% used Anyone know what "From Space:" is? Any ideas on further fault finding ideas? I dont have much experience with this type of fault finding.

  Read the article

< Previous Page | 49 50 51 52 53 54 55 56 57 58 59 60  | Next Page >