Freigeben über


ASP.NET Thread Usage on IIS 7.5, IIS 7.0, and IIS 6.0

I'd like to briefly explain how ASP.NET uses threads when hosted on IIS 7.5, IIS 7.0 and IIS 6.0, as well as the configuration changes that you can make to alter the defaults. Please take a quick look at the “Threading Explained” section in Chapter 6 of “Improving .NET Application Performance and Scalability”. Prior to v2.0 of the .NET Framework, it was necessary to tweak the processModel/maxWorkerThreads, processModel/maxIoThreads, httpRuntime/minFreeThreads, httpRuntime/minLocalRequestFreeThreads, and connectionManagement/maxconnection configuration. The v2.0 .NET Framework attempted to simplify this by adding a new processModel/autoConfig configuration, which made the changes for you at runtime. With the introduction of IIS 7.0 and the ASP.NET integrated pipeline, we've introduced another element to the mix, a registry key named MaxConcurrentRequestsPerCPU. Lets start with a discussion of how things worked on IIS 6.0 before discussing the changes made in IIS 7.0.

When ASP.NET is hosted on IIS 6.0, the request is handed over to ASP.NET on an IIS I/O thread. ASP.NET immediately posts the request to the CLR ThreadPool and returns HSE_STATUS_PENDING to IIS. This frees up IIS threads, enabling IIS to serve other requests, such as static files. Posting the request to the CLR Threadpool also acts as a queue. The CLR Threadpool automatically adjusts the number of threads according to the workload, so that if the requests are high throughput there will only be 1 or 2 threads per CPU, and if the requests are high latency there will be potentially far more concurrently executing requests than 1 or 2 per CPU. The queuing provided by the CLR Threadpool is very useful, because while the requests are in the queue there is only a very small amount of memory allocated for the request, and it is all native memory. It’s not until a thread picks up the request and begins to execute that we enter managed code and allocate managed memory.

The CLR Threadpool is not the only queue used by ASP.NET when hosted in IIS 6.0. There are also queues at the application level, within each AppDomain. If there is a lot of latency, the CLR Threadpool will grow and inject more active threads. At some point we would either run out of threads, not have enough threads left over for performing other tasks, or the memory associated with all the concurrently executing requests would be too much, so ASP.NET imposes a cap on the number of threads concurrently executing requests. This is controlled by the httpRuntime/minFreeThreads and httpRuntime/minLocalRequestFreeThreads settings. If the cap is exceeded, the request is queued in the application-level queue, and executed later when the concurrency falls back down below the limit. The performance of these application-level queues is really quite miserable. If you observe that the “ASP.NET Applications\Requests in Application Queue” performance counter is non-zero, you definitely have a performance problem. These queues were implemented to prevent thread exhaustion and contention related to web service requests. The problem was first described in KB 821268, which I had published many years ago. The KB article has been re-written a few times since it was originally published, and I hope nothing has been lost during the translations.

For most usage scenarios, the changes recommended in the KB article are not necessary because v2.0 introduced processModel/autoConfig. However, the autoConfig setting may not work for everyone--it limits the number of concurrently executing requests per CPU to 12. An application with high latency may want to allow higher concurrency than this, in which case you can disable autoConfig and make the changes yourself. If you do allow higher concurrency, keep an eye on your working set. I believe the default works for about 90% of the applications out there. I do wish we had the foresight to name that setting maxConcurrentRequestsPerCPU, and allow it to be used to control concurrency, since that would be much easier to configure. I guess this is just another example of when business was just a little bit faster than the speed of thought.

When ASP.NET is hosted on IIS 7.5 and 7.0 in integrated mode, the use of threads is a bit different. First of all, the application-level queues are no more. Their performance was always really bad, there was no hope in fixing this, and so we got rid of them. But perhaps the biggest difference is that in IIS 6.0, or ISAPI mode, ASP.NET restricts the number of threads concurrently executing requests, but in IIS 7.5 and 7.0 integrated mode, ASP.NET restricts the number of concurrently executing requests. The difference only matters when the requests are asynchronous (the request either has an asynchronous handler or a module in the pipeline completes asynchronously). Obviously if the reqeusts are synchronous, then the number of concurrently executing requests is the same as the number of threads concurrently executing requests, but if the requests are asynchronous then these two numbers can be quite different as you could have far more reqeusts than threads. So how do things work, exactly, in integrated mode? Similar to IIS 6.0 (classic mode, a.k.a. ISAPI mode), the request is still handed over to ASP.NET on an IIS I/O thread. And ASP.NET immediately posts the request to the CLR Threadpool and returns pending. We found this thread switch was still necessary to maintain optimal performance for static file requests. So although you will take a performance hit if you’re only executing ASP.NET requests, if you have a mix of dynamic and static files, as we see with many large corporate workloads, this thread switch will actually free up threads for retrieving the static files. Finally, once the request is picked up by a thread from the CLR Threadpool, we check to see how many requests are currently executing. If the count is too high, the request is queued in a global (process-wide) queue. This global, native queue performs much better than the application-level queues used when we’re running in ISAPI mode (same as on IIS 6.0). There is very little memory associated with a queued request, and we have not entered managed code yet so there is no managed memory associated with it. And we respect the FIFO aspect of a queue, something we didn’t do with the application-level queues--if there was more than one application, there was no simple way to globally manage the individual queues. We did however have a difficult time trying to come up with a good configuration story for the IIS 7.0 changes.

When I discuss how to configure thread usage for ASP.NET/IIS 7.0 integrated mode, please remember that we have a lot of pre-existing code and configuration, and you can’t just create something new the way you would like to without introducing backward compatibility issues. In this new mode, the CLR Threadpool is still controlled by the processModel configuration settings (autoConfig, maxWorkerThreads, maxIoThreads, minWorkerThreads, and minIoThreads). And autoConfig is still enabled, but its modifications to httpRuntime/minFreeThreads and httpRuntime/minLocalRequestFreeThreads do nothing, since the application-level queues do not exist. Perhaps we should have tried to use them to configure the global (process-wide) queue limits, but they have application scope (httpRuntime configuration is application specific), not process scope, not to mention being too difficult to understand. And because of some issues with using the configuration system that I won’t go into right now, we decided to use a registry key to control concurrency. So for IIS 7.0 integrated mode, a DWORD named MaxConcurrentRequestsPerCPU within HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\ASP.NET\2.0.50727.0 determines the number of concurrent requests per CPU. By default, it does not exist and the number of requests per CPU is limited to 12. If you’re curious to see how much faster ASP.NET requests execute without the thread switch, you can set the value to 0. This will cause the request to execute on the IIS I/O thread, without switching to a CLR Threadpool thread. I don’t recommend this primarily because dynamic requests take a long time to execute relative to static requests, and I believe the overall performance of the system is better with the thread switch. However, and this is important, if your application consists of primarily or entirely asynchronous requests, the default MaxConcurrentReqeustsPerCPU limit of 12 will be too restrictive for you, especially if the requests are very long running. In this case, I do recommend setting MaxConcurrentRequestsPerCPU to a very high number.  In fact, in v4.0, we have changed the default for MaxConcurrentRequestsPerCPU to 5000.  There's nothing special about 5000, other than it is a very large number, and will therefore allow plenty of async requests to execute concurrently.  One thing to watch out for is that when concurrency increases, your application will use more memory simply because there are more requests executing in managed code.  The CLR ThreadPool will still do a great job maintaining the number of threads in the ThreadPool, so there should be no concern about this adversly impacting synchronous requests.  I know there are people using ASP.NET 2.0 and developing Comet or Comet-like applications on WS08 x64 servers, and they set MaxConcurrentRequestsPerCPU to 5000 and increase the HTTP.sys kernel queue limit to 10,000 (it has a default of 1000).  The HTTP.sys kernel queue limit is controlled by IIS.  You can change it by opening IIS Manager and opening the Advanced Settings for your application pool and changing the value of "Queue Length".

As a final remark, please note that the processModel/requestQueueLimit configuration limits the maximum number of requests in the ASP.NET system for IIS 6.0, IIS 7.0, and IIS 7.5. This number is exposed by the "ASP.NET/Requests Current" performance counter, and when it exceeds the limit (default is 5000) we reject requests with a 503 status (Server Too Busy).

-Thomas

 

UPDATE (Aug-18-2008) : .NET Framework v3.5 SP1 released earlier this week and it includes an update to the v2.0 binaries that supports configuring IIS application pools via the aspnet.config file.  The aspnet.config file is not very well known.  It is the CLR Hosting configuration file, and ASP.NET/IIS pass it to the CLR when the CLR is loaded.  The host configuration file (aspnet.config) applies configuration at the process-level, as opposed to the application-level like web.config.  There is a new system.web/applicationPool configuration section which applies to integrated mode only (Classic/ISAPI mode ignores these settings). The new config section with default values is:

    <system.web>
        <applicationPool maxConcurrentRequestsPerCPU="12" maxConcurrentThreadsPerCPU="0" requestQueueLimit="5000"/>
    </system.web>

There is a corresponding IIS 7.5 change (Windows Server 2008 R2 only) which allows different aspnet.config files to be specified for each application pool (this change has not been ported to IIS 7.0). With this, you can configure each application pool differently.  The maxConcurrentRequestsPerCPU setting is the same as the registry key described above, except that the setting in aspnet.config will override the registry key value.  The maxConcurrentThreadsPerCPU setting is new, and allows concurrency to be gated by the number of threads, similar to the way it was done in Classic/ISAPI mode.  By default maxConcurrentThreadsPerCPU is disabled (has a value of 0), in favor of gating concurrency by the number of requests, primarily because maxConcurrentRequestsPerCPU performs better (gating the number of threads is more complicated/costly to implement).  Normally you'll use request gating, but you now have the option of disabling it (set maxConccurrentRequestsPerCPU=0) and enabling maxConccurentThreadsPerCPU instead.  You can also enable both request and thread gating at the same time, and ASP.NET will ensure both requirements are met.  The requestQueueLimit setting is the same as processModel/requestQueueLimit, except that the setting in aspnet.config will override the machine.config setting.  All of this may be a little confusing, but for nearly everyone, my recommendation is that for ASP.NET 2.0 you should use the same settings as the defaults in ASP.NET v4.0; that is, set maxConcurrentRequestsPerCPU = "5000" and maxConcurrentThreadsPerCPU="0".

 

UPDATE (Sep-12-2011) :  The only relevant change to .NET Framework v4.0 (as compared to 3.5 or 2.0) is that the default for maxConcurrentRequestsPerCPU was increased to 5000.  5000 is also the value you should use in versions 2.0 and 3.5, which have a default of 12.  Also, IIS 7.5 is identical to IIS 7.0 as far as threading is concerned.  The only difference between IIS 7.5 and 7.0 that is relevant to this blog post is the support to configure different aspnet.config files for each application pool.  You do this by setting the CLRConfigFile attribute for the application pool.  You can then use the system.web applicationPool configuration mentioned above to set different values for maxConcurrentRequestsPerCPU, maxConcurrentThreadsPerCPU, and requestQueueLimit, if desired. 

In general, running with default configuration works best.  However, applications that have measurable latency, say latency of 100 milliseconds when communicating with a backend web service, will perform better with a few configuration changes.  Let me tell you what configuration changes you should make on IIS 7.0 and IIS 7.5 in integrated mode in order to handle a large number of concurrent requests to an application that has backend latency. By large number of concurrent reqests, I mean between 12 and 5000 per CPU.

  1. For v2.0 and v3.5 set a DWORD registry value @ HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\ASP.NET\2.0.50727.0\MaxConcurrentRequestsPerCPU = 5000.  Restart IIS
  2. For v3.5, you can alternatively set <system.web><applicationPool maxConcurrentRequestsPerCPU="5000"/></system.web> in the aspnet.config file.  If the value is set in both places, the aspnet.config setting overrides the registry setting.
  3. For v4.0, the default maxConcurrentRequestsPerCPU is 5000, so you don't need to do anything.
  4. Increase the HTTP.sys queue limit, which has a default of 1000.  If the operating system is x64 and you have 2 GB of RAM or more, setting it to 5000 should be fine.  If it is too low, you may see HTTP.sys reject requests with a 503 status.  Open IIS Manager and the Advanced Settings for your Application Pool, then change the value of "Queue Length".
  5. If your ASP.NET application is using web services (WFC or ASMX) or System.Net to communicate with a backend over HTTP you may need to increase connectionManagement/maxconnection.  For ASP.NET applications, this is limited to 12 * #CPUs by the autoConfig feature.  This means that on a quad-proc, you can have at most 12 * 4 = 48 concurrent connections to an IP end point.  Because this is tied to autoConfig, the easiest way to increase maxconnection in an ASP.NET application is to set System.Net.ServicePointManager.DefaultConnectionLimit programatically, from Application_Start, for example.  Set the value to the number of concurrent System.Net connections you expect your application to use.  I've set this to Int32.MaxValue and not had any side effects, so you might try that--this is actually the default used in the native HTTP stack, WinHTTP.  If you're not able to set System.Net.ServicePointManager.DefaultConnectionLimit programmatically, you'll need to disable autoConfig , but that means you also need to set maxWorkerThreads and maxIoThreads.  You won't need to set minFreeThreads or minLocalRequestFreeThreads if you're not using classic/ISAPI mode.
  6. If your application sees a large number of concurrent requests at start-up or has a bursty load, where concurrency increases suddenly, you will need to make the application asynchronous because the CLR ThreadPool does not respond well to these loads.  The CLR ThreadPool injects new threads at a rate of about 2 per second.  This is true for all versions of the CLR (v1.0 thru v4.0) at the time of this writing.  If concurrency is bursty and the request thread blocks (e.g. on a backend with latency), the injection rate of 2 threads per second will make your application respond very poorly to this work load.  The fix is to stop blocking on threads by using asynchronous I/O to communicate with the backend with latency.  If you cannot make the application asynchronous, you will need to increase minWorkerThreads.  I don't like to increase minWorkerThreads.  It has a side effect on high-throughput synchronous requests that don't block on threads, because the thread count is artificially high.

 

-Thomas

Comments

  • Anonymous
    September 19, 2007
    Everyone seems to be coming off vacation and my book is done so now to list the resources to make us

  • Anonymous
    April 20, 2008
    Two Threads per Request In .NET 3.0 and 3.5, there is a special behavior that you would observe for IIS-hosted

  • Anonymous
    August 13, 2008
    Folks were hassling me in the comments for not posting the picosecond that .NET 3.5 SP1 came out (or

  • Anonymous
    August 13, 2008
    (Copied from http://blogs.msdn.com/wenlong/archive/2008/04/21/wcf-request-throttling-and-server-scalability.aspx

  • Anonymous
    August 13, 2008
    (Copied from http://blogs.msdn.com/wenlong/archive/2008/04/21/wcf-request-throttling-and-server-scalability.aspx

  • Anonymous
    August 13, 2008
    Folks were hassling me in the comments for not posting the picosecond that .NET 3.5 SP1 came out (or,

  • Anonymous
    October 25, 2008
    The comment has been removed

  • Anonymous
    October 27, 2008
    Registry key names are not case sensitive, so maxconcurrentrequestspercpu will work as well as MaxConcurrentRequestsPerCPU.  In 3.5 SP1, you can use aspnet.config, as shown in the post.  Be sure aspnet.config is located in the same directory as the framework: x86: %windir%Microsoft.NETFrameworkv2.0.50727aspnet.config x64: %windir%Microsoft.NETFramework64v2.0.50727aspnet.config Also note that <system.web> must be contained in the <configuration> section.

  • Anonymous
    December 10, 2008
    최근 진행된 SCOM/SSRS 관련 Troubleshooting에서 처음으로 64bit Windows 2008에서 작동 중인 IIS 프로세스 디버깅을 시도해었습니다. 물론 Managed

  • Anonymous
    January 21, 2009
    maxConcurrentRequestsPerCPU  setting in registry has to be spelled correclty. I wasted a lot of time in setting it to maxConcurrentRequestsPerCpu and it didn't work. Once I changed the dword to maxConcurrentRequestsPerCPU , all my 50 concurrent requests started coming in, instead of just 12 earlier. I would suggest the author to write a official KB article on this.

  • Anonymous
    February 13, 2009
    When hosting an ASP.net web application on IIS7, and you have the application running in integrated pipeline

  • Anonymous
    February 13, 2009
    When hosting an ASP.net web application on IIS7, and you have the application running in integrated pipeline

  • Anonymous
    March 26, 2009
    When hosting an ASP.net web application on IIS7, and you have the application running in integrated pipeline

  • Anonymous
    April 07, 2009
    any behavior changes in WS08?... was wondering about the "this change has not been ported to WS08" comment.

  • Anonymous
    April 08, 2009
    In reference to IIS allowing different aspnet.config files to be specified for each Application Pool, I found out today that it was not ported to WS08.  The change was only made to Windows 7.

  • Anonymous
    February 01, 2010
    About the aspnet.config per application pool setting:

  1. where is it located under Windows 7?
  2. Was it ported to Windows Server 2008 R2?
  • Anonymous
    February 01, 2010
    Found my answers in this Powerpoint: http://download.microsoft.com/download/8/C/2/8C21BAFE-3432-48D1-962A-F7A9DD54A2AC/Extend%20Your%20Web%20Server%20-%20What's%20New%20in%20IIS%20and%20the%20Microsoft%20Web%20Platform.pptx 2008 R2, ApplicationHost.config: <applicationPools> <add name=”DefaultAppPool” CLRConfigFile=”c:myConfigCLRConfigFile.txt” />

  • Anonymous
    March 02, 2010
    Hi , sorry for my englisch.... I looking to this ppt . but I understand this ... I have webservice  which is called from app 200x . But in the w3wp is still 30 threads . Where I set permit number of threads or connection ? thanx

  • Anonymous
    March 02, 2010
    Marek, the number of connections used by System.Net to make web service requests is limited by maxconnection (see connectionManagement configuration section).  In an ASP.NET application, when autoConfig="true" (see processModel configuration section), maxconnection is set to 12*N, where N is the number of CPUs.  I do not think you should change the default values of these settings. Thanks, Thomas

  • Anonymous
    April 07, 2010
    Hi Thomas, Read your post but the one thing I couldn't figure out is how to change the maximal thread count for Asp.Net 2 with Framework 3.5 SP1 on IIS 6.0. Searched all over the web but everything leads back to this post. Thanks in advance, Jonathan.

  • Anonymous
    April 07, 2010
    In general you should not change the default settings for Asp.Net 2 with Framework 3.5 SP1 on IIS 6.0.  They are automatically set for you, but if you really want to cause yourself some trouble you can disable this by setting processModel/autoConfig to false.  The full list of settings controled by autoConfig are listed on MSDN in the description for that attribute.  Thread counts are controlled by maxWorkerThreads and maxIoThreads.  Read KB 821268.  Best of luck!

  • Anonymous
    April 10, 2010
    Hi Thomas, Thanks for the quick response - in our case I guess that the autoConfig values weren't working in our favor. Playing around with these values actually solved an issue we've had with ASP.Net performance.

  • Anonymous
    April 15, 2010
    Dear Thomas, Thank you for the detailed explanation. I'd like to ask what is the relation between requests/threads, as described in your article, and the ThreadPool, as described in the following link: http://stackoverflow.com/questions/1453283/threadpool-in-iis-context More specifically, you mention about 12 requests/threads per CPU (assuming synchronous operations), while in the link above it says that there is one thread pool per process (w3wp.exe) and this thread pool has a default size of 250 worker threads per available processor. Moreover, for a machine with 2 Intel Xeon E5450 processors, what is the total number of requests allowed by default? 2x12, or 2x(4 cores each)x12? Thank you, Ioannis

  • Anonymous
    April 16, 2010
    The comment has been removed

  • Anonymous
    April 22, 2010
    The comment has been removed

  • Anonymous
    April 22, 2010
    I do not like web gardens, which is what we call it when you have multiple processes running for the same application pool.  Technically, if you're running out of a resource that has process scope, then adding more processes will give you more of those resources, and perhaps help.  The reason I don't like web gardens is because they require more memory and don't really solve any problem--at best they might work around a problem that should really be fixed in a different way. You need to determine why your web server doesn't scale under load.  I'd start with performance counters.  Is "Processor% Processor Time" at or greater than 90%?  Is ".NET CLR Memory% Time in GC" above 10%?  If the value of "ASP.NET ApplicationsPipeline Instance Count" and/or "ASP.NET ApplicationsRequests Executing" is pegged at 12 * #CPUs, in your case 12 * 8 = 96, then the pages that make synchronous web service requests might be preventing any other requests from running.  Hard to tell for sure without looking in a debugger or digging through the W3SVC logs. I think you should make the web service requests asynchronous.  Also, it is possible that you need to set the TcpTimedWaitDelay registry setting, which has a default of 2 minutes, as well as the MaxUserPort setting, to more appropriate values for your load.  You can determine whether this is necessary by running "netstat -n" and looking to see how many connections are in the TIME_WAIT state.  See KB 328476 for a better description of this. Please do set MaxConcurrentRequestsPerCPU = 5000, which is the new default in v4.0.  The v2.0/3.5 default of 12 is too low.  However, you also need to make your web service requests asynchronous.

  • Anonymous
    April 22, 2010
    Thomas, Thanks for the prompt response. Regarding your questions about performance counters - I cann tell you that memory usage is very low; however processor Time at those times is almost 100%. I do not know about the other counters - we rolled back our code changes, but surely I will collect these counters when the code will be deployed again. We will make calls async, this is on our to do list for some time. Thanks again

  • Anonymous
    July 19, 2010
    The comment has been removed

  • Anonymous
    July 19, 2010
    Pierre, First, you might want to eliminate client issues by applying a load to a very simple Hello World page, just to make sure that you're able to maximize the CPU. Now if you're not having any client issues, then usually you can determine why the CPU is not maximized by applying the load and then breaking into the process with a debugger.  You will often see where the requests are blocked just by looking at stack traces for all the threads.  I would use windbg and sos for this, which you can learn about by doing a search online.  This should be quite easy if the % Processor Time is really low. Alternatively, you could use a call attributed profiler, but this will be harder to setup, and probably cost you some money unless you've already purchased a profiler.  A sampling profiler often won't help in this situation, although in general, I prefer them over call attributed profilers.  Anyhow, when I can't maximize the CPU, I'll just break in with a debugger and see what the threads are doing. Thanks, Thomas

  • Anonymous
    July 19, 2010
    The comment has been removed

  • Anonymous
    July 20, 2010
    Pierre, If all threads are in Sleep and you're only seeing 7 rps on this dual core server, my guess is that your load client is only issuing about ten requests concurrently.  Otherwise, the thread pool work item queue would have entries and the thread count would grow (because the CPU utilization is so low) along with the number of concurrently executing requests.  I have a few questions:

  1. What's the value of the "ASP.NETRequests Current" performance counter when you see 7rps?  
  2. Is the application pool using integrated or classic mode?
  3. What version of ASP.NET is this?
  4. Is the web service ASMX or SVC? Thanks, Thomas
  • Anonymous
    July 20, 2010
    The comment has been removed

  • Anonymous
    July 20, 2010
    Hi, We have server 2008 x64. so we have to add DWORD (32 bit) or QWORD (64bit) for maxConcurrentRequestsPerCPU?

  • Anonymous
    July 21, 2010
    Pierre, Can you set this DWORD registry value, restart IIS, and run the test again? HKLMSOFTWAREMicrosoftASP.NET2.0.50727.0MaxConcurrentRequestsPerCPU = 5000 It's not clear if you have v3.5 SP1, which is required in order to set the value in aspnet.config, so please try the registry change and restart IIS. Thanks, Thomas

  • Anonymous
    July 21, 2010
    Milind, it is a DWORD.

  • Anonymous
    July 21, 2010
    Thomas, I do have .NET 3.5SP1 installed but in any case already had that registry entry during my previous tests (I tried the registry before changing the aspnet.config). Also in one of my tests I tried to prove that the aspnet.config was indeed used by setting the maxConcurrentThreadsPerCPU to 1 and keeping the user count to 500: I immediately saw 500 requests queued when i started the load test, which is a very different behavior then the test I described in my previous post. I know it's a lot to ask and I thank you for keeping up with so far but if you have a load tester handy I would really appreciate if you could try to reproduce. I can send you my test project if needed but it would be almost faster to create from scratch (the extent of the code is this one action I pasted in my previous post) :) Thank you, pierre

  • Anonymous
    July 21, 2010
    I can't help, but I bet you can narrow this down yourself.  You might alternatively try Microsoft Product Support, or one of the forums, like those on www.asp.net or www.iis.net. Thanks, Thomas

  • Anonymous
    September 25, 2010
    Hi Thomas, Does IIS 7 integrated mode ignore maxConnection attrib (under connectionManagement) as well? Is it possible that Pierre's number of outgoing TCP/IP connection is throttled at 2? I am trying to make sure the optimum settings for my app, other than setting the maxConcurrentRequestsPerCPU. Thanks! Randhir

  • Anonymous
    September 27, 2010
    In all ASP.NET applications, the maxConnection attribute is set automatically to 12 * #CPUs when autoConfig="true" in the processModel configuration section of machine.config.  By default, autoConfig="true".  So for example, on a dual core server, it is set to 24 automatically.  

  • Anonymous
    September 29, 2010
    The comment has been removed

  • Anonymous
    October 11, 2010
    Mikhai,

  1. On IIS 6 and in IIS 7 classic mode, each application (AppDomain) has a queue that it uses to maintain the availability of worker threads. The number of requests in this queue increases if the number of available worker threads falls below the limit specified by <httpRuntime minFreeThreads=/>.   When the limit specified by <httpRuntime appRequestQueueLimit=/> is exceeded, the request is rejected with a 503 status code and the client receives an HttpException with the message "Server too busy."   There is also an ASP.NET performance counter, "Requests In Application Queue", that indicates how many requests are in the queue.  Yes, the CLR thread pool is the one exposed by the .NET ThreadPool class.
  2. The requestQueueLimit is poorly named.  It actually limits the maximum number of requests that can be serviced by ASP.NET concurrently.  This includes both requests that are queued and requests that are executing.  If the "Requests Current" performance counter exceeds requestQueueLimit, new incoming requests will be rejected with a 503 status code.
  3. On IIS 7 integrated mode, if the total number of excecuting requests is greater than maxConcurrentRequestsPerCPU * #CPUs, then new incoming requests will be inserted into the queue. Thanks, Thomas
  • Anonymous
    October 11, 2010
    Tomas, thanks for the explanation it's quite clear now. Mikhail Mikheev

  • Anonymous
    October 12, 2010
    Thanks Thomas. So essentially as I understand it, if app A connects to a web service S, then there can not be anymore than 12 calls (per CPU core) in flight from A to S. Other calls to S from A will block on the acquire connection call till either timeout happens or one of those 12 calls completes and closes the connection. Is it the same case if A is calling S asynchronously? Can 13th call be placed on the network while other 12 are waiting on IOCompletion port?

  • Anonymous
    October 28, 2010
    Randhir, It is unfortunate that we have the constraint set by the maxconnection attribute (in connectionManagement section).  It is really a pain to scale with System.Net.HttpWebRequest. To answer your question, ASP.NET sets maxconnection to 12 times the number of CPU cores automatically when processModel/autoConfig is set to true.  It does not matter if you use System.Net.HttpWebRequest synchronously or asynchronously, the number of connections between a client and a server is limited by the value of maxconnection.  To change the config value, you have to disable autoConfig or set System.Net.ServicePointManager.DefaultConnectionLimit programmatically, once, after the ASP.NET AppDomain starts. To scale with System.Net.HttpWebRequest, you will also probably need to change TcpTimedWaitDelay (to 30 seconds) and MaxUserPorts (to 0xFFFF), but before you do that you might want to determine if this is an issue.  See blogs.msdn.com/.../powershell-script-troubleshooting-for-port-exhaustion-using-netstat.aspx for some diagnosis steps. Once you've solved the network bottle necks, you're probably going to run out of threads next.  Keep an eye on the thread count and the ThreadPool thread limits. Finally, if you've removed all these bottle necks, you will very likely run into memory issues.  Keep an eye on % Time in GC and if it is above 5% (it most likely will be), use the CLR Profiler to reduce allocations on a per request basis.  One of my first blog posts explains how to use the CLR Profiler. Making web requests to a backend tends to be so expensive and complicated to configure that I suspect most people scale by adding additional web servers to a web farm. Thanks, Thomas

  • Anonymous
    November 04, 2010
    Thomas, Do you know if  the maxConcurrentRequestsPerCPU settings and/or defaults changed in .Net 4 and/or Server 2008 R2?

  • Anonymous
    November 04, 2010
    John, The default for maxConcurrentRequestsPerCPU in v4.0 is 5000.  With WS08 R2, you can also set the value in the aspnet.config file as described in the blog post. Thanks, Thomas

  • Anonymous
    February 09, 2011
    Hi Thomas. I wanted to verify that I have the correct steps in applying the settings Is this value also needed in the registry? - maxConcurrentThreadsPerCPU="0".   Do I need to add the DWORD in the registry AND update the ‘aspnet.config’ or is the registry the only location needed? Thanks!

  • Anonymous
    June 30, 2011
    Just wanted to thank you for writing this article. This was very enlightening to me and helped me get to the bottom of a couple of troublesome bottlenecks in our code.

  • Anonymous
    July 03, 2011
    Just want to ask two questions. I am using ASP.NET 4.0, IIS 7.0 integrated mode, 1) If I need data from a remote database which takes a lot of time to respond. Whether I use synchronous API or asynchronous API? Is my other request get some benefit If I use asynchronous API for the current request. 2) How do you relate ThreadPool.GetMaxThreads(or ThreadPool.SetMaxThreads) with maxConcurrentRequestsPerCPU? I mean when maxConcurrentRequestsPerCPU=5000 and ThreadPool.GetMaxThreads =100 then what happens?

  • Anonymous
    July 18, 2011
    TomW, sorry for the very long delay, for Windows Server 2008 R2 you can set the registry key or aspnet.config.  aspnet.config will override the registry key.  Settings that you do not specify will assume their default values.

  • Anonymous
    July 18, 2011
    anonymious - It would be better not to block threads while waiting for a slow backend (more than ~100 milliseconds) to respond.  Asynchronous applications will respond better to backend latency and fluctuations in client load.  Your user experience will be better with an asychronous application if the backend has latency.  The downside is that asynchronous programming is more complex and more difficult to debug and maintain, but once written you don't have to revist until it breaks or you need to update it.  To answer your question about the thread pool, when the application is asynchronous it will not use threads when waiting for the backend to respond, so you can have higher request concurrency and use very few threads.

  • Anonymous
    July 19, 2011
    Have been suffering from a restriction to the number of REQUESTS that will process at any one time against a single WORKER PROCESS. Loading IIS , WORKER PROCESS , VIEW CURRENT REQUESTS i would generally see 0-10 concurrent. I use performance monitor to view CONCURRENT CONNECTIONS against the process WEB SERVICE and when they hit 600+ the number of CURRRENT REQUEST jumps from 0 - 10 too 100s or 1000s slowing the application. (IIS7, ASP.NET, mixture of synchronous and asynchronous requests). Have tired amending machine.config, registry and aspnet.config as below:


HKEY_LOCAL_MACHINESOFTWAREMicrosoftASP.NET2.0.50727.0   MaxConcurrentRequestsPerCPU (SET TO 0)

ASPNET.CONFIG AMENDMENTS: C:WindowsMicrosoft.NETFrameworkv2.0.50727ASPNET.CONFIG C:WindowsMicrosoft.NETFramework64v2.0.50727ASPNET.CONFIG <system.web>   <applicationPool      maxConcurrentRequestsPerCPU="0"      maxConcurrentThreadsPerCPU="0"      requestQueueLimit="500000" /> </system.web>

machine.config = <system.web> <processModel autoConfig="true" maxIoThreads="50000" maxWorkerThreads="50000" minIoThreads="500" minWorkerThreads="500" />         <httpRuntime             minFreeThreads="176"             minLocalRequestFreeThreads="152"         />

But to date these changes have not resolved the issue or have appeared to influence it at all. Have attempted to increase the MAXIMUM WORKER PROCESSS against the app pool to 2 (creating web garden) and now see 2 x app pools as expected. Have enabled all web site features in the hope that this will resolve but reading this blog suggests it probably wont. Thanks for any help.

  • Anonymous
    September 12, 2011
    Hello Thomas: I know you explained the differences of IIS on different .NET framework. My question is, could you shed some light in terms of IIS in the context of different OS? For instance, I just read this blog blog.stevensanderson.com/.../measuring-the-performance-of-asynchronous-controllers and it stated "Don’t even bother trying to load test your asynchronous controllers using IIS on Windows XP, Vista, or 7. Under these operating systems, IIS won’t handle more than 10 concurrent requests anyway, so you certainly won’t observe any benefits." So are you talking about IIS 7 in say Windows Server 2008, but not in the context of say Windows Vista or 7? Thanks a bunch.

  • Anonymous
    September 13, 2011
    Hi Liming,   The server operating systems are configured and tuned to run services, and the client operating systems are configured and tuned for the user application experience.  The client also has limitations on the number of remote connections, such as a maximum of 10 TCP/IP connections.  Yes, you would not load test a client operating system.  Do your scale, performance testing, and production deployment on a server operating system. Thanks, Thomas

  • Anonymous
    November 18, 2011
    Thomas, I have been exercising the concurrency settings and processModel in both IIS 6 and IIS 7 and have learned a lot.  Thank you for a great explanation.  However, I am still unclear about how to view the queue depth in the native global queue. For IIS 7 with an integrated app pool under .NET 2.0, if I set MaxConcurrentRequestsPerCPU to 5000 in the registry I can see I am no longer bound by this concurrency setting but instead by the asp.net processModel.  However, what I've noticed is that in the default processModel config on a single core VM where there is 100 maxWorkerThreads and 1 minWorkerThread (as obtained by System.Threading.ThreadPool.GetMaxThreads and System.Threading.ThreadPool.GetMinThreads, IIS becomes slow to respond as a large amount of traffic goes up (using WCAT) - essentially I'm waiting for the CLR Thread Pool to allocate threads (I know how to set the minWorkerThreads to account for a burst of traffic, but please read on as this is not what I'm confused about)  I'll see ASP.NET Apps v2.0.50727Requests Executing slowly increase but no queuing in ASP.NET v2.0.50727Requests Queued.  I can see the .NET CLR worker threads slowly being allocated using System.Threading.ThreadPool.GetAvailableThreads AND IIS is slow to respond until more threads are allocated, but no queue depth is visible. That said, if I understand correctly, requests are queuing in the native global queue until the .NET CLR can spin up the required worker threads, but I don't know what perfmon counter I can view to see this global native queue depth, if there is one available.  So, until IIS 7 hits max worker threads we're slow but I can't see how many are queued.  (In IIS 6, while threads are slowly being allocated in the .NET CLR Thread Pool I can see the queuing). PS.  If I do the same test with using the MaxConcurrentRequestsPerCPU set to 12 or other low value, once I hit request concurrency, the ASP.NET v2.0.50727Requests Queued.counter starts going up by the expected delta. Hopefully the info provided is clear.  I really want to understand how I can see the queue depth of the global native queue you referred to. Thanks!

  • Anonymous
    December 08, 2011
    The comment has been removed

  • Anonymous
    January 19, 2012
    very helpful

  • Anonymous
    January 08, 2013
    The comment has been removed

  • Anonymous
    January 08, 2013
    The comment has been removed

  • Anonymous
    January 08, 2013
    The comment has been removed

  • Anonymous
    January 09, 2013
    I have reposted the three comments above with nice markup in the IIS forum here: forums.iis.net/.../1. If I find any answers on my own, I'll write them there instead.

  • Anonymous
    January 09, 2013
    Tobias, in response to the post you gave on the IIS forum:

  1. Yes, appRequestQueueLimit only applies to IIS 6 (also 7 when running in classic mode). 2a) IIS 7 and later have the queues that you mention. 2b) The HTTP.sys kernel queue is not the same as the ASP.NET process-wide queue. 2c) The HTTP.sys kernel queue is essentially a completion port on which user-mode (IIS) receives requests from kernel-mode (HTTP.sys).  It has a queue limit, and when that is exceeded you will receive a 503 status code.  The HTTPErr log will also indicate that this happened by logging a 503 status and QueueFull. 2d) I do not know the details of how HttpClient or HttpWebRequest are implemented.  You need to ensure that you are closing/disposing all System.Net objects properly.  You likely need to increase connectionManagment/maxconnection in the config file or increase it programmatically via ServicePointManager.DefaultConnectionLimit.  You may also need to modify the default registry values for TcpTimedWaitDelay and MaxUserPorts if your connections are sitting in the TIME_WAIT state or you do not have enough ports available.  Be careful with these registry values--you need to know what you're doing, and why you're doing it.  Perhaps the System.Net folks have a forum? 2e) "Process(w3wp)Thread Count" and the ".NET CLR LocksAndThreads" performance counters will help a little, but ultimately you will need to resort to the debugger (windbg) and the sos.dll debugger extension.  It has a !ThreadPool command that will tell you how many threads are active in the pool and what the maximum limits are. 2f) "Web ServiceCurrent Connections" is the number of connections to IIS.  This has nothing to do with your outbound System.Net connections.
  2. ASP.NET v4.5 has a performance counter in the "ASP.NET" category specifically for the native queue.  This is new to v4.5. 4a) Perhaps, but I'm not familiar with the "TCPv4Connection Established" performance counter. 4b) Yes, I would be careful about changing TcpTimedWaitDelay and/or MaxUserPort.  You need to know what you're doing, and why you're doing it. Thanks, Thomas PS. I moved away from the ASP.NET team several years ago and am not actively posting to this blog.  I recommend that you post questions on the ASP.NET forums, which is actively used by many ASP.NET folks.
  • Anonymous
    May 31, 2013
    Hi, When we talk about any of this setting which needs to be applied per CPU, is it physical or logical CPU? for e.g. minfreethreads = 88 * # CPU Here # CPU is physical or logical?

  • Anonymous
    April 16, 2014
    The number of CPU's recognized by .NET, and by extension ASP.NET, is the number of logical CPU's

  • Anonymous
    June 05, 2014
    Hello, I am having a application running on IIS, i want to make it capable of handling 50,000 request per minute. Currently when the number of request goes above 30 thousand request per minute i am getting error 503.2 service unavailable which is due to request limit exceeded.   I am having a HP DL series server with 2 processors with 12 core each. Can any one suggest me the best IIS settings for this application. Thanks in advance.

  • Anonymous
    November 25, 2014
    Greetings I would like to know what the  HTTP.sys queue limit should be running a 32-bit application pool on IIS 7.0 with Server 2008 running SQL server 2008. Any advice will be appreciated. Thanks in advance.

  • Anonymous
    January 29, 2015
    Thank you very much for this great article. It gave my team and me the opportunity to fix a production issue we were facing for almost one year.

  • Anonymous
    April 23, 2015
    thanks for writing this good article. I am setting following configuration and http.sys queue length to  10.    <applicationPool        maxConcurrentRequestsPerCPU="13"        maxConcurrentThreadsPerCPU="0"        requestQueueLimit="15" /> I was expecting to get 503 after 25 or 35 requests. but it is able to accept more than 1000 requests "requests current ' perfcounter was showing  more than 1000. I also modified  process model element of machine.config but it did not work. Acutally maxConcurrentRequestsPerCPU and maxConcurrentThreadsPerCPU are working but requestQueueLimit is not being honored.  Am I doing anything wrong. any suggestion to debug.

  • Anonymous
    May 01, 2015
    Ignore my above comment, i was able to verify it. It was accepting more than queuelenght for the first time, later on everything went fine.

  • Anonymous
    July 12, 2015
    Note: Max connections for aspnet under autoconfig changed for net 45 - go to the MS referencesource website (for .net45 source code) and find "SetAutoConfigLimits" private function of HttpRuntime class, System.Web. See the maxconnections limit under auto config is now defaulted higher to Max Int (previously 12*cores as per update above)

  • Anonymous
    November 10, 2015
    Sometimes, in IIS 7.5, Windows Server 2008 R2 in WebSite for ASP.NET 4.5.1, in AppPool Classic, CLR 4.0, I get "Server Too Busy" Which is good patterns and practices about it ? web.config configuration without performance issues? Thx a lot.