Synchronous and Asynchronous Sockets  

One of the first issues that you'll encounter when developing your application is the difference between synchronous (blocking) and asynchronous (non-blocking) connections. Whenever you perform some operation on a socket, it may not be able to complete immediately and return control back to your program. For example, a read on a socket cannot complete until some data has been sent by the server. If there is no data waiting to be read, one of two things can happen: the function can wait until some data has been written on the socket, or it can return immediately with an error that indicates that there is no data to be read.

The first case is called a synchronous or blocking socket. In other words, the program is "blocked" until the request for data has been satisfied. When the server does write some data on the socket, the read operation will complete and execution of the program will resume. The second case is called an asynchronous or non-blocking socket, and requires that the application recognize the error condition and handle the situation appropriately.

Programs that use asynchronous sockets typically use one of two methods when sending and receiving data. The first method is called polling and the program periodically attempts to read or write data from the socket, typically using a timer. The second method is to use what is called asynchronous event notification. This means that the program is notified whenever a socket event takes place, and in turn can respond to that event. For example, if the remote program writes some data to the socket, an event is generated so that program knows it can read the data from the socket at that point. Events can be in the form of Windows messages posted to the application's message queue, or as callback functions. With the ActiveX control, standard COM events call any event handlers that have been written.

Synchronous Sockets

For historical reasons, the default behavior is for sockets to function synchronously and not return until the operation has completed. However, blocking sockets in Windows can introduce some special problems in single-threaded applications. To prevent the program from becoming non-responsive, the blocking function will enter what is called a "message loop" where it continues to process messages sent to it by Windows and other applications. Because messages are being processed, this means that the program can be re-entered at a different point with the blocked operation parked on the program's stack. For example, consider a program that attempts to read some data from the socket when a button is pressed. Because no data has been written yet, it blocks and the program goes into a message loop. The user then presses a different button, which causes code to be executed, which in turn attempts to read data from the socket, and so on.

To resolve the general problems with blocking sockets, the Windows Sockets standard states that there may only be one outstanding blocked call per thread of execution. This means that applications that are re-entered (as in the example above) will encounter errors whenever they try to take some action while a blocking function is already in progress. If the language supports the creation of threads, it is strongly recommended that the program create worker threads to perform any socket I/O.

There are significant advantages to using blocking sockets. In most cases, the application design and implementation is simpler, and raw throughput (the rate at which data is sent and received) is generally higher with blocking sockets because it does not have to go through an event mechanism to notify the application of a change in status. If you are using a programming language that supports multithreading, then the use of synchronous sockets is typically the best choice. However, if your are using an older language that does not provide support for multithreading, such as Visual Basic 6.0, and your program needs to establish multiple simultaneous connections, then an asynchronous, event-driven design is more appropriate.

Asynchronous Sockets

SocketWrench facilitates the use of asynchronous sockets by generating events when appropriate. For example, an OnRead event occurs whenever the server writes on the socket, which tells your application that there is data waiting to be read. The use of non-blocking sockets will be demonstrated in the next section, and is one of the key areas in which an ActiveX control has a distinct advantage over coding directly against the Windows Sockets API.

In general, the use asynchronous sockets is preferred when you have a single-threaded application that must establish multiple, simultaneous connections with a server. In that situation, the use of non-blocking sockets avoids the restriction that prevents more than one outstanding socket operation in the thread and can enable the program to remain more responsive to the user. Practically speaking, there are few languages today that do not support multithreading, so this limitation tends to apply more to the legacy languages such as Visual Basic 6.0.

Best Practices

If your programming language of choice does support multithreading, it is recommended that you create worker threads to manage the sockets in your program. This leaves the main thread responsible for handling the user interface, and the worker threads can handle the network communications. There are some significant advantages to this approach:

  • The networking code is generally isolated from the user interface, only requiring that the main UI thread be notified of the progress of the operation. For example, updating a progress bar control as the contents of a file is being downloaded. This tends to minimize any clutter in the UI code and creates a clear separation of functionality that will make the program easier to modify and maintain.
  • Isolating the networking code in a worker thread ensures that there are no conflicts between other threads, including the main UI thread. Each thread effectively owns the sockets that it creates, and those sockets can be used independently of one another without concern about potential conflicts.
  • Code written using synchronous sockets is typically easier to update, maintain and debug. The coding style lends itself to a more straight forward, top-down structure and logical errors are usually easier to find than with code written using asynchronous sockets.
  • There is less overhead associated with synchronous sockets because no event mechanism is used, and handlers don't have to be implemented in callback functions. Event notifications that post messages to hidden window, as is the case with the ActiveX control, have to be processed through the message queue which is typically shared by the UI thread.
  • Polling an asynchronous socket can cause spikes in CPU utilization and is generally not recommended. Applications which attempt to simulate blocking sockets by creating an asynchronous socket and then polling it can negatively impact the performance of the application, and in some cases the overall system.

In summary, there are three general approaches that can be taken when building an application with regard to blocking or non-blocking sockets:

  • Use a synchronous (blocking) socket. In this mode, the program will not resume execution until the socket operation has completed. In a single-threaded application, blocking socket operations can cause code to be re-entered at a different point, leading to complex interactions (and difficult debugging) if there are multiple active connections in use by the application. If the programming language supports multithreading, it is recommended that each connection be isolated within its own worker thread.
  • Use an asynchronous (non-blocking) socket, which allows your application to respond to events. For example, when the server writes data to the socket, an OnRead event is generated for the ActiveX control. Your application can respond by reading the data from the socket, and perhaps send some data back, depending on the context of the data received. The code required for managing asynchronous sockets can be more complex, however it is the best solution for single-threaded applications that must establish simultaneous connections.
  • Use a combination of synchronous and asynchronous socket operations. The ability to switch between blocking and non-blocking modes "on the fly" provides a powerful and convenient way to perform socket operations under some circumstances. However, switching between blocking and non-blocking mode can make the application more complex and difficult to debug. It is important to note that the warning regarding blocking sockets also applies here.

If you decide to use asynchronous sockets in your application, it's important to keep in mind that you must check the return value from every read and write operation. It is possible that your may not be able to send or receive all of the data specified at that time. Frequently, developers encounter problems when they write a program that assumes a given number of bytes can always be written to or read from the socket. In many cases, the program works as expected when developed and tested on a local area network, but fails unpredictably when the program is released to a user that has a slower network connection (such as a serial dial-up connection to the Internet). By always checking the return values of these operations, you insure that your program will work correctly, regardless of the speed or configuration of the network.