Although the Internet and World Wide Web (WWW) are popular as tools for convenient exchange of information, it is not easy to utilise the Internet for time-critical applications such as on-line remote diagnosis in telemedicine. It is a wish of the United Nations to bring e-health to every corner of the world via the Internet. This is easier said than done because the sheer size of the Internet implies unpredictable faults of all kinds. These faults are physically translated into communication and computation delays. Since these faults and delays have many contributing factors that can change suddenly, it is impractical to monitor them all for the sake of fault tolerance. For this reason the new concept of interpreting the channel dynamics by gauging its end-to-end behaviour has emerged. The aim is to measure the changes of the average service roundtrip time (RTT) over time and interpret the possible signs of faults from these changes. If the length of the average service RTT is suddenly increased in an exponential manner, network congestion and widespread retransmission are indicated.
Then, the Internet and/or the applications running on it should invoke fault tolerance measures to prevent system breakdown and partial failures. This concept of gauging the channel dynamics to prevent system failure is generally known as Internet End-to-End Performance Measurement (IEPM). The purpose of the book is to shed light on some of the novel practical fault tolerance techniques that can help shorten the end-to-end service roundtrip (RTT) time of a logical Internet channel. As a result the Internet can be harnessed for serious time-critical applications. Several practical cases are presented to demonstrate how the effective harnessing can be achieved.