Anonymous remote computing: A paradigm for parallel programming on interconnected workstations

Parallel computing on interconnected workstations is becoming a viable and attractive proposition due to the rapid growth in speeds of interconnection networks and processors. In the case of workstation clusters, there is alw ays a considerable amount of unused computing capacity available in the net work. However, heterogeneity in architectures and operating systems, load v ariations on machines, variations in machine availability, and failure susc eptibility of networks and workstations complicate the situation for the pr ogrammer. In this context, new programming paradigms that reduce the burden involved in programming for distribution, load adaptability, heterogeneity , and fault tolerance gain importance. This paper identifies the issues inv olved in parallel computing on a network of workstations. The Anonymous Rem ote Computing (ARC) paradigm is proposed to address the issues specific to parallel programming on workstation systems. ARC differs from the conventio nal communicating process model by treating a program as one single entity consisting of several loosely coupled remote instruction blocks instead of treating it as a collection of processes. The ARC approach results in distr ibution transparency and heterogeneity transparency. At the same time, it p rovides fault tolerance and load adaptability to parallel programs on works tations. ARC is developed in a two-tiered architecture consisting of high l evel language constructs and low level ARC primitives. The paper describes an implementation of the ARC kernel supporting ARC primitives.