We previously wrote a newsletter about the TCP window size (see link below). The responses we received to that newsletter convinced us that we should occasionally devote future newsletters to detailing other WAN optimization technologies. 

With that in mind, the next three newsletters will discuss some flow and congestion control mechanisms that are used within TCP. One of the reasons that these mechanisms are important is that they directly influence network throughput.

In a perfect world, there would be no need for flow or congestion control mechanisms. The sending device would transmit packets as quickly as it could, the packets would all be delivered in order and without unexpected delay, and the receiving device could quickly process as many packets as the sending device could send.

However, since that is not a realistic situation, TCP has mechanisms that account for two of the key factors that determine the performance of a TCP/IP network. Those factors being the ability of the receiving device to accept data, and the amount of congestion that exists within the network.

In order to account for the ability of the receiving device to accept data, TCP has a flow control mechanism for each connection. Each segment header contains a field called "advertised window" that indicates how many additional bytes of data the receiver can accept. The sending device is not allowed to send more bytes than the advertised window.

The next newsletter will outline some of TCP's congestion control mechanisms and will discuss their relationship to flow control.

Steve Taylor is president of Distributed Networking Associates and publisher/editor-in-chief of Webtorials. Jim Metzler is vice president of Ashton, Metzler & Associates.