Asynchronous Methods using Wait and Async
In IIS Web server .Net Framework maintains a pool of Threads to serve incoming requests. When a request arrives a thread from the pool is dispatched to process that request. If the request is processed synchronously, the thread that processes the request is busy while the request is being processed, and that thread cannot service another request.
This might not be a problem, because the thread pool can be made large enough to accommodate many busy threads. However, the number of threads in the thread pool is limited (the default maximum for .NET 4.5 is 5,000). In large applications with high concurrency of long-running requests, all available threads might be busy. This condition is known as thread starvation. When this condition is reached, the web server queues requests. If the request queue becomes full, the web server rejects requests with an HTTP 503 status (Server Too Busy). The CLR thread pool has limitations on new thread injections. If concurrency is bursty (that is, your web site can suddenly get a large number of requests) and all available request threads are busy because of backend calls with high latency, the limited thread injection rate can make your application respond very poorly. Additionally, each new thread added to the thread pool has overhead (such as 1 MB of stack memory). A web application using synchronous methods to service high latency calls where the thread pool grows to the .NET 4.5 default maximum of 5, 000 threads would consume approximately 5 GB more memory than an application able the service the same requests using asynchronous methods and only 50 threads. When you’re doing asynchronous work, you’re not always using a thread. For example, when you make an asynchronous web service request, ASP.NET will not be using any threads between the async method call and the await. Using the thread pool to service requests with high latency can lead to a large memory footprint and poor utilization of the server hardware.
Processing Asynchronous Requests
In web applications that sees a large number of concurrent requests at
start-up or has a bursty load (where concurrency increases suddenly), making
these web service calls asynchronous will increase the responsiveness of your
application. An asynchronous request takes the same amount of time to process as
a synchronous request. For example, if a request makes a web service call that
requires two seconds to complete, the request takes two seconds whether it is
performed synchronously or asynchronously. However, during an asynchronous call,
a thread is not blocked from responding to other requests while it waits for
the first request to complete. Therefore, asynchronous requests prevent request
queuing and thread pool growth when there are many concurrent requests that
invoke long-running operations.In general, use synchronous methods for the following conditions:
- The operations are simple or short-running.
- Simplicity is more important than efficiency.
- The operations are primarily CPU operations
instead of operations that involve extensive disk or network overhead. Using
asynchronous action methods on CPU-bound operations provides no benefits and
results in more overhead.
- You're calling services that can be consumed through asynchronous methods, and you're using .NET 4.5 or higher.
- The operations are network-bound or I/O-bound instead of CPU-bound.
- Parallelism is more important than simplicity of code.
- You want to provide a mechanism that lets users cancel a long-running request.
- When the benefit of switching threads out weights the cost of the context switch. In general, you should make a method asynchronous if the synchronous method waits on the ASP.NET request thread while doing no work. By making the call asynchronous, the ASP.NET request thread is not stalled doing no work while it waits for the web service request to complete.
- Testing shows that the blocking operations are a bottleneck in site performance and that IIS can service more requests by using asynchronous methods for these blocking calls.
Few applications require all action methods to be asynchronous. Often, converting a few synchronous action methods to asynchronous methods provides the best efficiency increase for the amount of work required.
The Sample Application
You can download the sample application from https://github.com/RickAndMSFT/Async-ASP.NET/ on the GitHub site. The repository consists of three projects:- Mvc4Async: The ASP.NET MVC 4 project that contains the code used in this tutorial. It makes Web API calls to the WebAPIpgw service.
- WebAPIpgw: The ASP.NET MVC 4 Web API project that implements the
Products, Gizmos and Widgetscontrollers. It provides the data for the WebAppAsync project and the Mvc4Async project. - WebAppAsync: The ASP.NET Web Forms project used in another tutorial.
The Gizmos Synchronous Action Method
The following code shows theGizmos synchronous action method that
is used to display a list of gizmos. (For this article, a gizmo is a fictional
mechanical device.)
public ActionResult Gizmos() { ViewBag.SyncOrAsync = "Synchronous"; var gizmoService = new GizmoService(); return View("Gizmos", gizmoService.GetGizmos()); }The following code shows the
GetGizmos method of the gizmo
service.public class GizmoService { public async Task<List<Gizmo>> GetGizmosAsync( // Implementation removed. public List<Gizmo> GetGizmos() { var uri = Util.getServiceUri("Gizmos"); using (WebClient webClient = new WebClient()) { return JsonConvert.DeserializeObject<List<Gizmo>>( webClient.DownloadString(uri) ); } } }The
GizmoService GetGizmos method passes a URI to an ASP.NET Web API HTTP
service which returns a list of gizmos data. The WebAPIpgw
project contains the implementation of the Web API gizmos, widget and
product controllers.The following image shows the gizmos view from the sample project.
Creating an Asynchronous Gizmos Action Method
The sample uses the new async and await keywords (available in .NET 4.5 and Visual Studio 2012) to let the compiler be responsible for maintaining the complicated transformations necessary for asynchronous programming. The compiler lets you write code using the C#'s synchronous control flow constructs and the compiler automatically applies the transformations necessary to use callbacks in order to avoid blocking threads.The following code shows the
Gizmos synchronous method and the GizmosAsync
asynchronous method. If your browser supports the
HTML 5 <mark>
element, you'll see the changes in GizmosAsync in yellow
highlight.public ActionResult Gizmos() { ViewBag.SyncOrAsync = "Synchronous"; var gizmoService = new GizmoService(); return View("Gizmos", gizmoService.GetGizmos()); }
public async Task<ActionResult> GizmosAsync() { ViewBag.SyncOrAsync = "Asynchronous"; var gizmoService = new GizmoService(); return View("Gizmos", await gizmoService.GetGizmosAsync()); }The following changes were applied to allow the
GizmosAsync to
be asynchronous.- The
method is marked with the
async
keyword, which tells the compiler to generate callbacks for parts of
the body and to automatically create a
Task<ActionResult>that is returned. - "Async" was appended to the method name. Appending "Async" is not required but is the convention when writing asynchronous methods.
- The return type was changed from
ActionResulttoTask<ActionResult>. The return type ofTask<ActionResult>represents ongoing work and provides callers of the method with a handle through which to wait for the asynchronous operation’s completion. In this case, the caller is the web service.Task<ActionResult>represents ongoing work with a result ofActionResult. - The await keyword was applied to the web service call.
- The asynchronous web service API was called (
GetGizmosAsync).
GetGizmosAsync method body another
asynchronous method, GetGizmosAsync is called. GetGizmosAsync immediately
returns a Task<List<Gizmo>> that will eventually complete when the
data is available. Because you don’t want to do anything else until
you have
the gizmo data, the code awaits the task (using the await keyword). You can
use the await keyword only in methods annotated with the async
keyword. The await keyword does not block the thread until the task is complete. It signs up the rest of the method as a callback on the task, and immediately returns. When the awaited task eventually completes, it will invoke that callback and thus resume the execution of the method right where it left off. For more information on using the await and async keywords and the Task namespace, see the async references section.
The following code shows the
GetGizmos and GetGizmosAsync methods.public List<Gizmo> GetGizmos() { var uri = Util.getServiceUri("Gizmos"); using (WebClient webClient = new WebClient()) { return JsonConvert.DeserializeObject<List<Gizmo>>( webClient.DownloadString(uri) ); } }
public async Task<List<Gizmo>> GetGizmosAsync() { var uri = Util.getServiceUri("Gizmos"); using (HttpClient httpClient = new HttpClient()) { var response = await httpClient.GetAsync(uri); return (await response.Content.ReadAsAsync<List<Gizmo>>()); } }The asynchronous changes are similar to those made to the GizmosAsync above.
-
The method signature was annotated with the async keyword,
the return type was changed to
Task<List<Gizmo>>, and Async was appended to the method name. - The asynchronous HttpClient class is used instead of the WebClient class.
- The await keyword was applied to the HttpClient asynchronous methods.
The browsers presentation of the gizmos data is identical to the view created by the synchronous call. The only difference is the asynchronous version may be more performant under heavy loads.
Performing Multiple Operations in Parallel
Asynchronous action methods have a significant advantage over synchronous methods when an action must perform several independent operations. In the sample provided, the synchronous methodPWG (for
Products, Widgets and Gizmos) displays the results of three web service calls to
get a list of products, widgets, and gizmos. The
ASP.NET Web API project that provides
these services uses
Task.Delay to simulate latency or slow network calls. When the delay is set
to 500 milliseconds, the asynchronous PWGasync method takes a
little over 500 milliseconds to complete while the synchronous PWG version takes over 1,500 milliseconds. The synchronous
PWG method is shown in the following code.public ActionResult PWG() { ViewBag.SyncType = "Synchronous"; var widgetService = new WidgetService(); var prodService = new ProductService(); var gizmoService = new GizmoService(); var pwgVM = new ProdGizWidgetVM( widgetService.GetWidgets(), prodService.GetProducts(), gizmoService.GetGizmos() ); return View("PWG", pwgVM); }The asynchronous
PWGasync method is shown in the following code.public async Task<ActionResult> PWGasync() { ViewBag.SyncType = "Asynchronous"; var widgetService = new WidgetService(); var prodService = new ProductService(); var gizmoService = new GizmoService(); var widgetTask = widgetService.GetWidgetsAsync(); var prodTask = prodService.GetProductsAsync(); var gizmoTask = gizmoService.GetGizmosAsync(); await Task.WhenAll(widgetTask, prodTask, gizmoTask); var pwgVM = new ProdGizWidgetVM( widgetTask.Result, prodTask.Result, gizmoTask.Result ); return View("PWG", pwgVM); }The following image shows the view returned from the PWGasync method.
Using a Cancellation Token
Asynchronous action methods returningTask<ActionResult>
are cancelable, that is they take a
CancellationToken parameter when one is provided with the
AsyncTimeout attribute. The following code shows the GizmosCancelAsync
method with
a timeout of 150 milliseconds.[AsyncTimeout(150)] [HandleError(ExceptionType = typeof(TimeoutException), View = "TimeoutError")] public async Task<ActionResult> GizmosCancelAsync( CancellationToken cancellationToken ) { ViewBag.SyncOrAsync = "Asynchronous"; var gizmoService = new GizmoService(); return View("Gizmos", await gizmoService.GetGizmosAsync(cancellationToken)); }The following code shows the GetGizmosAsync overload, which takes a CancellationToken parameter.
public async Task<List<Gizmo>> GetGizmosAsync(string uri,In the sample application provided, selecting the Cancellation Token Demo link calls theCancellationToken cancelToken = default(CancellationToken)){ using (HttpClient httpClient = new HttpClient()) { var response = await httpClient.GetAsync(uri,cancelToken); return (await response.Content.ReadAsAsync<List<Gizmo>>()); } }
GizmosCancelAsync method and demonstrates the cancelation
of the asynchronous call.Server Configuration for High Concurrency/High Latency Web Service Calls
To realize the benefits of an asynchronous web application, you might need to make some changes to the default server configuration. Keep the following in mind when configuring and stress testing your asynchronous web application.- Windows 7, Windows Vista and all Windows client operating systems have a maximum of 10 concurrent requests. You'll need a Windows Server operating system to see the benefits of asynchronous methods under high load.
- Register .NET 4.5 with IIS from an elevated command prompt:
%windir%\Microsoft.NET\Framework64\v4.0.30319\aspnet_regiis -i
See ASP.NET IIS Registration Tool (Aspnet_regiis.exe) - You might need to increase the
HTTP.sys queue limit from the
default value of 1,000 to 5,000. If the setting is too low, you may see
HTTP.sys reject requests with a HTTP 503 status. To change the
HTTP.sys queue limit:
- Open IIS manager and navigate to the Application Pools pane.
- Right click on the target application pool and select Advanced
Settings.

- In the Advanced Settings dialog box, change Queue Length from 1,000
to 5,000.
Choosing Asynchronous and Synchronous
Multiple Operations in Parallal
High Configuration and High Latency Call
Thanks
ReplyDelete