Round Trip Time (RTT) and Time Out
The size and the complexity of computer networks have grown in past years. To achieve an efficient and reliable transmission some protocols inevitably need to handle complicated network traffics and unexpected transmission losses. These problems usually are referred to as flow control and congestion control. The technologies of managing complex computer network need to be more circumspect not only in the sending host and receiving hosts but also the intermediate routes. One of these protocols is the transmission control protocol that has a responsibility of ensuring reliability.
Because TCP guarantees the reliable delivery of data it retransmits each segment if an ACK is not received in a certain period of time. TCP sets this timeout as a function of the round trip time it expects between the two ends of eh connection . unfortunately given the range of possible RTT between any pair of hosts in the internet as well as the variation in RTT between the same two hosts over time. Choosing appropriate timeout value is not very easy. To address this problem TCP uses an adapt retransmission mechanism.
Round trip time (RTT) is an important in determining the behaviour of a TCP connection passively estimating RTT is useful in measuring the congestion window size and retransmission timeout of a connection as well as the available bandwidth on a path. This information can help determine factors that limit data flow rates and can congestion. When known at a network link along the path RTT can also aid efficient queen management and buffer provisioning. Additionally RTT can be used to improve node distribution in peer to peer and overlay networks.
The amount of time set at the time is however of great importance. If the time expires too quickly then premature time outs will be generated during usual case and thus unnecessary retransmission will occur. On the other hand if a long time is set for the timer the TCP will slowly respond to the segment loss which means longer delays in the transportation and overall end to end delay. Therefore the TCP time out must be set to be as optimum as possible.