Parallel computing on an Ethernet cluster of workstations: Opportunities and constraints

Parallel computing on clusters of workstations is receiving much attention from the research community. Unfortunately, many aspects of parallel comput ing over this parallel computing engine is not very well understood. Some o f these issues include the workstation architectures, the network protocols , the communication-to-computation ratio, the load balancing strategies, an d the data partitioning schemes. The aim of this paper is to assess the str engths and limitations of a cluster of workstations by capturing the effect s of the above issues. This has been achieved by evaluating the performance of this computing environment in the execution of a parallel ray tracing a pplication through analytical modeling and extensive experimentation. We we re successful in illustrating the effect of major factors on the performanc e and scalability of a cluster of workstations connected by an Ethernet net work. Moreover, our analytical model was accurate enough to agree closely w ith the experimental results. Thus, we feel that such an investigation woul d be helpful in understanding the strengths and weaknesses of an Ethernet c luster of workstation in the execution of parallel applications.