Search Results

Search found 143 results on 6 pages for 'perf'.

Page 1/6 | 1 2 3 4 5 6  | Next Page >

  • perf events documentation

    - by Thanatos
    I've searched for an exhaustive explanation of the meaning of each event monitored by the perf stat command; I've found a tutorial which explains quite well how to use different the features of the perf tool. However, it doesn't explain the meaning of several events that can be observed (and there are a lot!!). Someone know where is a quite simple and complete documentation about the events listed by the perf list command? In particular, I'm interested in finding out the percentage of cpu used by some application I wrote. Can i measure it directly through cpu-clock or task-clock? What's the meaning of these two events? Thanks in advance

    Read the article

  • SPARC T4-4 Delivers World Record Performance on Oracle OLAP Perf Version 2 Benchmark

    - by Brian
    Oracle's SPARC T4-4 server delivered world record performance with subsecond response time on the Oracle OLAP Perf Version 2 benchmark using Oracle Database 11g Release 2 running on Oracle Solaris 11. The SPARC T4-4 server achieved throughput of 430,000 cube-queries/hour with an average response time of 0.85 seconds and the median response time of 0.43 seconds. This was achieved by using only 60% of the available CPU resources leaving plenty of headroom for future growth. The SPARC T4-4 server operated on an Oracle OLAP cube with a 4 billion row fact table of sales data containing 4 dimensions. This represents as many as 90 quintillion aggregate rows (90 followed by 18 zeros). Performance Landscape Oracle OLAP Perf Version 2 Benchmark 4 Billion Fact Table Rows System Queries/hour Users* Response Time (sec) Average Median SPARC T4-4 430,000 7,300 0.85 0.43 * Users - the supported number of users with a given think time of 60 seconds Configuration Summary and Results Hardware Configuration: SPARC T4-4 server with 4 x SPARC T4 processors, 3.0 GHz 1 TB memory Data Storage 1 x Sun Fire X4275 (using COMSTAR) 2 x Sun Storage F5100 Flash Array (each with 80 FMODs) Redo Storage 1 x Sun Fire X4275 (using COMSTAR with 8 HDD) Software Configuration: Oracle Solaris 11 11/11 Oracle Database 11g Release 2 (11.2.0.3) with Oracle OLAP option Benchmark Description The Oracle OLAP Perf Version 2 benchmark is a workload designed to demonstrate and stress the Oracle OLAP product's core features of fast query, fast update, and rich calculations on a multi-dimensional model to support enhanced Data Warehousing. The bulk of the benchmark entails running a number of concurrent users, each issuing typical multidimensional queries against an Oracle OLAP cube consisting of a number of years of sales data with fully pre-computed aggregations. The cube has four dimensions: time, product, customer, and channel. Each query user issues approximately 150 different queries. One query chain may ask for total sales in a particular region (e.g South America) for a particular time period (e.g. Q4 of 2010) followed by additional queries which drill down into sales for individual countries (e.g. Chile, Peru, etc.) with further queries drilling down into individual stores, etc. Another query chain may ask for yearly comparisons of total sales for some product category (e.g. major household appliances) and then issue further queries drilling down into particular products (e.g. refrigerators, stoves. etc.), particular regions, particular customers, etc. Results from version 2 of the benchmark are not comparable with version 1. The primary difference is the type of queries along with the query mix. Key Points and Best Practices Since typical BI users are often likely to issue similar queries, with different constants in the where clauses, setting the init.ora prameter "cursor_sharing" to "force" will provide for additional query throughput and a larger number of potential users. Except for this setting, together with making full use of available memory, out of the box performance for the OLAP Perf workload should provide results similar to what is reported here. For a given number of query users with zero think time, the main measured metrics are the average query response time, the median query response time, and the query throughput. A derived metric is the maximum number of users the system can support achieving the measured response time assuming some non-zero think time. The calculation of the maximum number of users follows from the well-known response-time law N = (rt + tt) * tp where rt is the average response time, tt is the think time and tp is the measured throughput. Setting tt to 60 seconds, rt to 0.85 seconds and tp to 119.44 queries/sec (430,000 queries/hour), the above formula shows that the T4-4 server will support 7,300 concurrent users with a think time of 60 seconds and an average response time of 0.85 seconds. For more information see chapter 3 from the book "Quantitative System Performance" cited below. -- See Also Quantitative System Performance Computer System Analysis Using Queueing Network Models Edward D. Lazowska, John Zahorjan, G. Scott Graham, Kenneth C. Sevcik external local Oracle Database 11g – Oracle OLAP oracle.com OTN SPARC T4-4 Server oracle.com OTN Oracle Solaris oracle.com OTN Oracle Database 11g Release 2 oracle.com OTN Disclosure Statement Copyright 2012, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 11/2/2012.

    Read the article

  • What’s new in IIS8, Perf, Indexing Service-Week 49

    - by OWScott
    You can find this week’s video here. After some delays in the publishing process week 49 is finally live.  This week I'm taking Q&A from viewers, starting with what's new in IIS8, a question on enable32BitAppOnWin64, performance settings for asp.net, the ARR Helper, and Indexing Services. Starting this week for the remaining four weeks of the 52 week series I'll be taking questions and answers from the viewers. Already a number of questions have come in. This week we look at five topics. Pre-topic: We take a look at the new features in IIS8. Last week Internet Information Services (IIS) 8 Beta was released to the public. This week's video touches on the upcoming features in the next version of IIS. Here’s a link to the blog post which was mentioned in the video Question 1: In a number of places (http://learn.iis.net/page.aspx/201/32-bit-mode-worker-processes/, http://channel9.msdn.com/Events/MIX/MIX08/T06), I've saw that enable32BitAppOnWin64 is recommended for performance reasons. I'm guessing it has to do with memory usage... but I never could find detailed explanation on why this is recommended (even Microsoft books are vague on this topic - they just say - do it, but provide no reason why it should be done). Do you have any insight into this? (Predrag Tomasevic) Question 2: Do you have any recommendations on modifying aspnet.config and machine.config to deliver better performance when it comes to "high number of concurrent connections"? I've implemented recommendations for modifying machine.config from this article (http://www.codeproject.com/KB/aspnet/10ASPNetPerformance.aspx - ASP.NET Process Configuration Optimization section)... but I would gladly listen to more recommendations if you have them. (Predrag Tomasevic) Question 3: Could you share more of your experience with ARR Helper? I'm specifically interested in configuring ARR Helper (for example - how to only accept only X-Forwards-For from certain IPs (proxies you trust)). (Predrag Tomasevic) Question 4: What is the replacement for indexing service to use in coding web search pages on a Windows 2008R2 server? (Susan Williams) Here’s the link that was mentioned: http://technet.microsoft.com/en-us/library/ee692804.aspx This is now week 49 of a 52 week series for the web pro. You can view past and future weeks here: http://dotnetslackers.com/projects/LearnIIS7/ You can find this week’s video here.

    Read the article

  • SQLSat65, Great Perf Counters Poster from Quest

    - by merrillaldrich
    I was fortunate to be able to attend the Vancouver BC SQLSaturday this past weekend, and it was excellent! Great sessions, good facility, well attended. Nice work, and a huge thank you to the volunteers that made that happen. One side perk: I got a copy of this terrific performance counters poster from Quest, which you can download as a PDF for free. Very handy, especially as a teaching tool. I'm using it for my SCOM MP work. Check it out....(read more)

    Read the article

  • perf tuning for ESX vmfs3 on RAID

    - by maruti
    looking for recommendations on ESX4 OS - VMFS version3: RAID-5 : matching the stripe size with VMFS block size? (64K, 128K etc) RAID controller options: "adaptive read ahead, write-back" on PERC 6i 90% VMs on server are Windows (2008, 2003, Vista etc, SQL 2005 etc) i have read that smaller stipes are good for writes and larger for reads. Since this is virtual env, not sure whats good.

    Read the article

  • perf tuning for vmfs3 on RAID

    - by maruti
    recommendations for ESX4 OS - VMFS version3: matching: RAID-5 stripe size with VMFS block size? (64K, 128K etc) enabled "adaptive read ahead, write-back" on PERC 6i 90% VMs on server are Windows (2008, 2003, Vista etc, SQL 2005 etc) i have read that smaller stipes are good for writes and larger for reads. Since this is virtual env, not sure whats good.

    Read the article

  • Tracking down origin of I/O in multi-process server

    - by Craig Ringer
    I'm currently trying to track down some phantom I/O in a PostgreSQL build I'm testing. It's a multi-process server and it isn't simple to associate disk I/O back to a particular back-end and query. I thought Linux's perf tool would be ideal for this, but I'm struggling to capture block I/O performance counter metrics and associate them with user-space activity. It's easy to record block I/O requests and completions with, eg: sudo perf record -g -T -u postgres -e 'block:block_rq_*' and the user-space pid is recorded, but there's no kernel or user-space stack captured, or ability to snapshot bits of the user-space process's heap (say, query text) etc. So while you have the pid, you don't know what the process was doing at that point. Just perf script output like: postgres 7462 [002] 301125.113632: block:block_rq_issue: 8,0 W 0 () 208078848 + 1024 [postgres] If I add the -g flag to perf record it'll take snapshots of the kernel stack, but doesn't capture user-space state for perf events captured in the kernel. The user-space stack only goes up to the entry-point from userspace, like LWLockRelease, LWLockAcquire, memcpy (mmap'd IO), __GI___libc_write, etc. So. Any tips? Being able to capture a snapshot of the user-space stack in response to kernel events would be ideal.

    Read the article

  • Perf: Viewing thousands of images in Silverlight 3 on a 3D Wall

    - by Bob Holland
    I currently work on a very cool Silverlight app that displays photos in a 3D wall space like the Wall3D demo that is thrown in with Blend 3. The problem I am currently facing is performance. The app works like this: As you scroll right or left the 3d photo wall rotates As each movement is made, the next column of photos are downloaded, decoded into a BitmapImage and thrown into a 3D Wall Node. As you can imagine users (if you let them) will want to flip through the photos really quickly, but the problem I have is I cannot display the photos quick enough. In most cases it's a beautiful app that works really well, but when an album contains over 300 photos, you can imagine the sort of memory taken up by all the BitmapImage classes and how moving the slider can jump from photo 20 to photo 120 in a second. Of course we have algorithms in place to not download every photo in between, but I still can't work out a fast way to get the photos displayed. It may be a case that we need to throw away the 'great for show' 3D wall and go to a flat DeepZoom like wall like the Playboy archive one that Vertigo did. Still not sure, let me know your thoughts. P.S. We are using Kit3D for all the 3D work, it's using PerspectiveCamera, Model3DGroup, ModelVisual3D, RotateTransform3D & TranslateTransform3D. Cheers, Bob.

    Read the article

  • Performance Overhead of Perf Event Subsystem in Linux Kernel

    - by Bo Xiao
    Performance counters for Linux are a new kernel-based subsystem that provide a framework for all things performance analysis. It covers hardware level (CPU/PMU, Performance Monitoring Unit) features and software features (software counters, tracepoints) as well. Since 2.6.33, the kernel provide 'perf_event_create_kernel_counter' kernel api for developers to create kernel counter to collect system runtime information. What I concern most is the performance impact on overall system when tracepoint/ftrace is enabled. There are no docs I can find about them. I was once told that ftrace was implemented by dynamically patching code, will it slow the system dramatically?

    Read the article

  • Is there a perf hit using mule as a container vs. standard JEE container like Weblogic?

    - by Victor Grazi
    Our team is considering using Mule in a large scale medium volume internal facing transactional banking application. At first Mule would just be used as an application server although it is possible some of its esb/orchestration features would be used in the future. I have no experience with mule, being new on the team. But my gut says Mule would not be as performant as Weblogic or Glassfish as a deployment container. Does anyone have any comparison stories to share that might shed light?

    Read the article

  • How to collect the performance data of a server during an unreachable/down period using Nagios?

    - by gsc-frank
    Some time services and host stop responding due to a poor server performance. I mean, if for some reason (could be lot of concurrency services access, a expensive backup execution on the server or whatever that consume tons of server resources) a server performance is very degraded, that could lead that the server isn't capable to establish any "normal network communication" (without trigger whatever standards timeouts defined for such communication). Knowing host's performance data (cpu, memory, ...) in case of available during that period (host is not down and despite of its performance degradation still allow plugins collect performance data) could be very useful for sysadmin to try to determine what cause the problem, or at least, if the host performance was good and don't interfered at all in the host/service down. This problem could be solved using remote active (NRPE) or remote passive (NSCA) if such remote solutions could store (buffered) perf data to be send to central Nagios server when host performance or network outage allow it. I read the doc of both solutions and can't find any reference to such buffer mechanism neither what happened in case that NSCA can't reach Nagios server. Any idea of how solve this lack of info? so useful for forensic analysis. EDIT: My questions isn about which tools I can use to debug perf problems or gather perf data to analysis, but is about how collect (using Nagios) host perf data even during a network outage for its posterior analysis (kind of forensic analysis). The idea is integrate such data to Nagios graphers like pnp4nagios and NagiosGrapther. I know that I could install tools like Cacti in each of my host, and have a kind of performance data collection redundancy, but I really want avoid that and try to solve all perf analysis requirements with one tools: Nagios

    Read the article

  • Problems creating a functioning table

    - by Hoser
    This is a pretty simple SQL query I would assume, but I'm having problems getting it to work. if (object_id('#InfoTable')is not null) Begin Drop Table #InfoTable End create table #InfoTable (NameOfObject varchar(50), NameOfCounter varchar(50), SampledValue float(30), DayStamp datetime) insert into #InfoTable(NameOfObject, NameOfCounter, SampledValue, DayStamp) select vPerformanceRule.ObjectName AS NameOfObject, vPerformanceRule.CounterName AS NameOfCounter, Perf.vPerfRaw.SampleValue AS SampledValue, Perf.vPerfHourly.DateTime AS DayStamp from vPerformanceRule, vPerformanceRuleInstance, Perf.vPerfHourly, Perf.vPerfRaw where (ObjectName like 'Logical Disk' and CounterName like '% Free Space' AND SampleValue > 95 AND SampleValue < 100) order by DayStamp desc select NameOfObject, NameOfCounter, SampledValue, DayStamp from #InfoTable Drop Table #InfoTable I've tried various other forms of syntax, but no matter what I do, I get these error messages. Msg 207, Level 16, State 1, Line 10 Invalid column name 'NameOfObject'. Msg 207, Level 16, State 1, Line 10 Invalid column name 'NameOfCounter'. Msg 207, Level 16, State 1, Line 10 Invalid column name 'SampledValue'. Msg 207, Level 16, State 1, Line 10 Invalid column name 'DayStamp'. Msg 207, Level 16, State 1, Line 22 Invalid column name 'NameOfObject'. Msg 207, Level 16, State 1, Line 22 Invalid column name 'NameOfCounter'. Msg 207, Level 16, State 1, Line 22 Invalid column name 'SampledValue'. Msg 207, Level 16, State 1, Line 22 Invalid column name 'DayStamp'. Line 10 is the first 'insert into' line, and line 22 is the second select line. Any ideas?

    Read the article

  • My cpus are powered down periodically

    - by mgiammarco
    I post here because I am using Ubuntu but this is probably an hardware problem. Since I bought my new setup with AMD Athlon(tm) II X4 635 Processor and asus m4a89td pro/usb3 motherboard with ecc ram I have stuttering on videos. I was using ubuntu 11.10 now ubuntu 12.10. Looking at syslog I have found that periodically (I notice only on videos but it happens always) this thing happens: Mar 6 23:36:42 virtual1 kernel: [28564.375548] smpboot: CPU 1 is now offline Mar 6 23:36:42 virtual1 kernel: [28564.380751] smpboot: CPU 2 is now offline Mar 6 23:36:42 virtual1 kernel: [28564.394947] smpboot: CPU 3 is now offline Mar 6 23:36:48 virtual1 kernel: [28569.917021] smpboot: Booting Node 0 Processor 1 APIC 0x1 Mar 6 23:36:48 virtual1 kernel: [28569.928015] LVT offset 0 assigned for vector 0xf9 Mar 6 23:36:48 virtual1 kernel: [28569.928372] [Firmware Bug]: cpu 1, try to use APIC500 (LVT offset 0) for vector 0x400, but the register is already in use for vector 0xf9 on another cpu Mar 6 23:36:48 virtual1 kernel: [28569.928378] perf: IBS APIC setup failed on cpu #1 Mar 6 23:36:48 virtual1 kernel: [28569.931305] process: Switch to broadcast mode on CPU1 Mar 6 23:36:48 virtual1 kernel: [28569.934255] smpboot: Booting Node 0 Processor 2 APIC 0x2 Mar 6 23:36:48 virtual1 kernel: [28569.945554] [Firmware Bug]: cpu 2, try to use APIC500 (LVT offset 0) for vector 0x400, but the register is already in use for vector 0xf9 on another cpu Mar 6 23:36:48 virtual1 kernel: [28569.945558] perf: IBS APIC setup failed on cpu #2 Mar 6 23:36:48 virtual1 kernel: [28569.948124] process: Switch to broadcast mode on CPU2 Mar 6 23:36:48 virtual1 kernel: [28569.949644] smpboot: Booting Node 0 Processor 3 APIC 0x3 Mar 6 23:36:48 virtual1 kernel: [28569.960838] [Firmware Bug]: cpu 3, try to use APIC500 (LVT offset 0) for vector 0x400, but the register is already in use for vector 0xf9 on another cpu Mar 6 23:36:48 virtual1 kernel: [28569.960840] perf: IBS APIC setup failed on cpu #3 Mar 6 23:36:48 virtual1 kernel: [28569.962953] process: Switch to broadcast mode on CPU3 I have: updated bios; tried all (really) bios options; changed ram; changed psu and cpu cooler; tried 3.8.1 kernel. What can I do now? Please help me! Thanks, Mario

    Read the article

  • PerformanceCounterCategory.Exists is very slow if category doesn't exists

    - by Shrike
    I have kind of library which uses a bunch of its own perf counters. But I want my library works fine even if that perf counters weren't installed. So I've created wrappers aroung PerformanceCounter and on the first use check if PerfCounter exists or not. If they exists then I'm using native PerformanceCounter instead I'm using a wrapper which do nothing. So to check perf counter existence I use PerformanceCounterCategory.Exists The problem is that if there isn't such category then PerformanceCounterCategory.Exists call takes (on my machine) about 10 seconds! Needless to say it's too slow. What can I do? Code to try it by youself: using System; using System.Diagnostics; class Program { static void Main(string[] args) { var ts = Stopwatch.StartNew(); var res = PerformanceCounterCategory.Exists("XYZ"); Console.WriteLine(ts.ElapsedMilliseconds); Console.WriteLine("result:" + res); } }

    Read the article

  • Dereferencing confusion? What is this example here called?

    - by perf
    Considering this PHP example: class A { public function getB( ) { return new B(); } } class B { public function test( ) { echo "Hello"; } } I could use this: $a = new A( ); $b = $a->getB( ); $b->test( ); // Hello Or this: $a = new A(); $a->getB( )->test( ); // Hello Taking a closer look at the second example... What is the name of this form of expression? Does this have something to do with dereferencing? In which programming languages is this available? What other forms of this exist?

    Read the article

  • URL replacing regex

    - by perf
    I'm trying to make an url that adds a / to all hrefs and srcs in a string. It should only add a / to urls that don't have a http:// at their beginning and that don't have / yet also. If we have this: <a href="ABC">... <img src="DEFG">... <a href="/HIJ">... <a href="http://KLMN">... The results should be something like this: <a href="/ABC">... <img src="/DEFG">... <a href="/HIJ">... <a href="http://KLMN">... This is what i've come up till now: &(href|src)="?!(\/|http::\/\/)(.+)" And the replace would be $1="/$2" It isn't working, though. What am I doing wrong? How would the working regex have to look like

    Read the article

  • What’s new in Silverlight 4 RC?

    - by pluginbaby
    I am here in Las Vegas for MIX10 where Scott Guthrie announced today the release of Silverlight 4 RC and the Visual Studio 2010 tools. You can now install VS2010 RC!!! As always, downloads links are here: www.silverlight.net He also said that the final version of Silverlight 4 will come next month (so april)! 4 months ago, I wrote a blog post on the new features of Silverlight 4 beta, so… what’s new in the RC ?   Rich Text · RichTextArea renamed to RichTextBox · Text position and selection APIs · “Xaml” property for serializing text content · XAML clipboard format · FlowDirection support on Runs tag · “Format then type” support when dragging controls to the designer · Thai/Vietnamese/Indic support · UI Automation Text pattern   Networking · UploadProgress support (Client stack) · Caching support (Client stack) · Sockets security restrictions removal (Elevated Trust) · Sockets policy file retrieval via HTTP · Accept-Language header   Out of Browser (Elevated Trust) · XAP signing · Silent install and emulation mode · Custom window chrome · Better support for COM Automation · Cancellable shutdown event · Updated security dialogs   Media · Pinned full-screen mode on secondary display · Webcam/Mic configuration preview · More descriptive MediaSourceStream errors · Content & Output protection updates · Updates to H.264 content protection (ClearNAL) · Digital Constraint Token · CGMS-A · Multicast · Graphics card driver validation & revocation   Graphics and Printing · HW accelerated Perspective Transforms · Ability to query page size and printable area · Memory usage and perf improvements   Data · Entity-level validation support of INotifyDataErrorInfo for DataGrid · XPath support for XML   Parser · New architecture enables future innovation · Performance and stability improvements · XmlnsPrefix & XmlnsDefinition attributes · Support setting order-dependent properties   Globalization & Localization · Support for 31 new languages · Arabic, Hebrew and Thai input on Mac · Indic support   More … · Update to DeepZoom code base with HW acceleration · Support for Private mode browsing · Google Chrome support (Windows) · FrameworkElement.Unloaded event · HTML Hosting accessibility · IsoStore perf improvements · Native hosting perf improvements (e.g., Bing Toolbar) · Consistency with Silverlight for Mobile APIs and Tooling · SDK   - System.Numerics.dll   - Dynamic XAP support (MEF)   - Frame/Navigation refresh support   That’s a lot!   You will find more details on the following links: http://timheuer.com/blog/archive/2010/03/15/whats-new-in-silverlight-4-rc-mix10.aspx http://www.davidpoll.com/2010/03/15/new-in-the-silverlight-4-rc-xaml-features/   Technorati Tags: Silverlight

    Read the article

  • How to modify code so that it adheres to the Law of Demeter

    - by guazz
    public class BigPerformance { public decimal Value {get;set;} } public class Performance { public BigPerformance BigPerf {get; set}; } public class Category { public Performance Perf {get;set; } } If I call: Category cat = new Cateogry(); cat.Perf.BigPerf.Value = 1.0; I assume this this breaks the LoD? If so, how do I remedy this if I have a large number of inner class Properties?

    Read the article

  • The WaitForAll Roadshow

    - by adweigert
    OK, so I took for granted some imaginative uses of WaitForAll but lacking that, here is how I am using. First, I have a nice little class called Parallel that allows me to spin together a list of tasks (actions) and then use WaitForAll, so here it is, WaitForAll's 15 minutes of fame ... First Parallel that allows me to spin together several Action delegates to execute, well in parallel.   public static class Parallel { public static ParallelQuery Task(Action action) { return new Action[] { action }.AsParallel(); } public static ParallelQuery> Task(Action action) { return new Action[] { action }.AsParallel(); } public static ParallelQuery Task(this ParallelQuery actions, Action action) { var list = new List(actions); list.Add(action); return list.AsParallel(); } public static ParallelQuery> Task(this ParallelQuery> actions, Action action) { var list = new List>(actions); list.Add(action); return list.AsParallel(); } }   Next, this is an example usage from an app I'm working on that just is rendering some basic computer information via WMI and performance counters. The WMI calls can be expensive given the distance and link speed of some of the computers it will be trying to communicate with. This is the actual MVC action from my controller to return the data for an individual computer.  public PartialViewResult Detail(string computerName) { var computer = this.Computers.Get(computerName); var perf = Factory.GetInstance(); var detail = new ComputerDetailViewModel() { Computer = computer }; try { var work = Parallel .Task(delegate { // Win32_ComputerSystem var key = computer.Name + "_Win32_ComputerSystem"; var system = this.Cache.Get(key); if (system == null) { using (var impersonation = computer.ImpersonateElevatedIdentity()) { system = computer.GetWmiContext().GetInstances().Single(); } this.Cache.Set(key, system); } detail.TotalMemory = system.TotalPhysicalMemory; detail.Manufacturer = system.Manufacturer; detail.Model = system.Model; detail.NumberOfProcessors = system.NumberOfProcessors; }) .Task(delegate { // Win32_OperatingSystem var key = computer.Name + "_Win32_OperatingSystem"; var os = this.Cache.Get(key); if (os == null) { using (var impersonation = computer.ImpersonateElevatedIdentity()) { os = computer.GetWmiContext().GetInstances().Single(); } this.Cache.Set(key, os); } detail.OperatingSystem = os.Caption; detail.OSVersion = os.Version; }) // Performance Counters .Task(delegate { using (var impersonation = computer.ImpersonateElevatedIdentity()) { detail.AvailableBytes = perf.GetSample(computer, "Memory", "Available Bytes"); } }) .Task(delegate { using (var impersonation = computer.ImpersonateElevatedIdentity()) { detail.TotalProcessorUtilization = perf.GetValue(computer, "Processor", "% Processor Time", "_Total"); } }).WithExecutionMode(ParallelExecutionMode.ForceParallelism); if (!work.WaitForAll(TimeSpan.FromSeconds(15), task => task())) { return PartialView("Timeout"); } } catch (Exception ex) { this.LogException(ex); return PartialView("Error.ascx"); } return PartialView(detail); }

    Read the article

  • How do I use PerformanceCounterType AverageTimer32?

    - by Patrick J Collins
    I'm trying to measure the time it takes to execute a piece of code on my production server. I'd like to monitor this information in real time, so I decided to give Performance Analyser a whizz. I understand from MSDN that I need to create both an AverageTimer32 and an AverageBase performance counter, which I duly have. I increment the counter in my program, and I can see the CallCount go up and down, but the AverageTime is always zero. What am I doing wrong? Thanks! Here's a snippit of code : long init_call_time = Environment.TickCount; // *** // Lots and lots of code... // *** // Count number of calls PerformanceCounter perf = new PerformanceCounter("Cat", "CallCount", "Instance", false); perf.Increment(); perf.Close(); // Count execution time PerformanceCounter perf2 = new PerformanceCounter("Cat", "CallTime", "Instance", false); perf2.NextValue(); perf2.IncrementBy(Environment.TickCount - init_call_time); perf2.Close(); // Average base for execution time PerformanceCounter perf3 = new PerformanceCounter("Cat", "CallTimeBase", "Instance", false); perf3.Increment(); perf3.Close(); perf2.NextValue();

    Read the article

  • Overhead of Perfmon -> direct to SQL Database

    - by StuartC
    HI All, First up, I'm a total newb at Performance Monitoring. I'm looking to set up central performance monitoring of some boxes. 2K3 TS ( Monitor General OS Perf & Session Specific Counters ) 2K8 R2 ( XenApp 6 = Monitor General OS Perf & Session Specific Counters ) File Server ( Standard File I/O ) My ultimate aim is to get as many counters/information, without impacting the clients session experience at all. Including counters specific to their sessions. I was thinking it logging directly to a SQL on another server, instead of a two part process of blg file then relog to sql. Would that work ok? Does anyone know the overhead of going straight to SQL from the client? I've searched around a bit, but havent found so much information it can be overwhelming. thanks

    Read the article

  • CLI-Based monitoring tool for KVM

    - by Pinnacle
    I am developing a scheduler for running VMs on KVM. The scheduling has over-commitment of resources like memory and CPU. For this, I need a CLI-based monitoring tool that keeps me giving information about the resource usage of each VM, because it might be the case that due to over-provisioning of resources, VMs on a particular host are running very slowly depending on the benchmarks/programs each VM is running, and then I need to migrate a VM to another host and so on. I looked into libvirt-based tools like collects, MUNIN, Nagios-vert, etc.( http://libvirt.org/apps.html#monitoring ) I also looked into Ubuntu utility perf-kvm ( http://manpages.ubuntu.com/manpages/maverick/man1/perf-kvm.1.html ) I want to ask which CLI-based would be recommended by the community so that I can make a automated scheduler that takes care of the above situation.

    Read the article

  • Uninstalling Reporting Server 2008 on Windows Server 2008

    - by Piotr Rodak
    Ha. I had quite disputable pleasure of installing and reinstalling and reinstalling and reinstalling – I think about 5 times before it worked – Reporting Server 2008 on Windows Server with the same year number in name. During my struggle I came across an error which seems to be not quite unfamiliar to some more unfortunate developers and admins who happen to uninstall SSRS 2008 from the server. I had the SSRS 2008 installed as named instance, SQL2008. I wanted to uninstall the server and install it to default instance. And this is when it bit me – not the first time and not the last that day . The setup complained that it couldn’t access a DLL: Error message: TITLE: Microsoft SQL Server 2008 Setup ------------------------------ The following error has occurred: Access to the path 'C:\Windows\SysWOW64\perf-ReportServer$SQL2008-rsctr.dll' is denied. For help, click: http://go.microsoft.com/fwlink?LinkID=20476&ProdName=Microsoft+SQL+Server&EvtSrc=setup.rll&EvtID=50000&ProdVer=10.0.1600.22&EvtType=0x60797DC7%25400x84E8D3C0 ------------------------------ BUTTONS: OK This is a screenshot that shows the above error: This issue seems to have a bit of literature dedicated to it and even seemingly a KB article http://support.microsoft.com/kb/956173 and a similar Connect item: http://connect.microsoft.com/SQLServer/feedback/details/363653/error-messages-when-upgrading-from-sql-2008-rc0-to-rtm The article describes issue as following: When you try to uninstall Microsoft SQL Server 2008 Reporting Services from the server, you may receive the following error message: An error has occurred: Access to the path 'Drive_Letter:\WINDOWS\system32\perf-ReportServer-rsctr.dll' is denied. Note Drive_Letter refers to the disc drive into which the SQL Server installation media is inserted. In my case, the Note was not true; the error pointed to a dll that was located in Windows folder on C:\, not where the installation media were. Despite this difference I tried to identify any processes that might be keeping lock on the dll. I downloaded Sysinternals process explorer and ran it to find any processes I could stop. Unfortunately, there was no such process. I tried to rerun the installation, but it failed at the same step. Eventually I decided to remove the dll before the setup was executed. I changed name of the dll to be able to restore it in case of some issues. Interestingly, Windows let me do it, which means that indeed, it was not locked by any process. I ran the setup and this time it uninstalled the instance without any problems:   To summarize my experience I should say – be very careful, don’t leave any leftovers after uninstallation – remove/rename any folders that are left after setup has finished. For some reason, setup doesn’t remove folders and certain files. Installation on Windows Server 2008 requires more attention than on Windows 2003 because of the changed security model, some actions can be executed only by administrator in elevated execution mode. In general, you have to get used to UAC and a bit different experience than with Windows Server 2003. Technorati Tags: SQL Server 2008,Windows Server 2008,SRS,Reporting Services

    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

1 2 3 4 5 6  | Next Page >