Fault - Tolerant Clusters of Workstations with Single System Image

he computing trend is moving from clustering highend mainframes to clustering desktop computers. This trend is triggered by the widespread use of PCs, workstations, Gigabit networks, and middleware support for clustering. This paper presents new approaches to achieving fault tolerance and single system image (SSI) in a workstation cluster. A multicomputer cluster is a collection of node computers, which are physically connected by local area networks or high-bandwidth switch networks using optical fibres. The workstations in the cluster can work collectively as an integrated computing resource, that is a SSI, or they can operate as individual computers, separately.