Quality of service for wide area clusters

Abstract

Quality of Service (QoS) is virtually synonymous with networked multimedia but could other types of distributed system benefit from QoS? This position paper examines the potential use of QoS in wide area cluster computing. The class of parallel applications that can benefit from clusters is currently restricted due to their relatively high latency and low throughput consequences of conventional networking. LANs offer the best performance but also limit the scope for effective clustering to a single room or building. Reliability is another key difference between conventional rnultiprocessors and clusters: the former can reasonably be programmed with the error-free assumption but applications cannot be run on distributed clusters without programming against the potential for remote faults. Emergent wide area high speed switched networks have the potential to reduce latency, increase bandwidth, improve reliability and remove some of the geographical constraints from wide area clustering. Can QoS parameters be used or defined to exploit these capabilities ?