A comparison of the capabilities of PVM, MPI and JAVA']JAVA for distributed parallel processing

Networked Unix workstations as well as workstations based on Windows 95 and Windows NT are fast becoming the standard for computing environments in many universities and research sites. In order to harness the tremendous potential for computing capability represented by these networks of workstations many new (and not so new) tools are being developed Parallel Virtual Machine (PVM) and Message Passing interface (MPI) have existed on Unix workstations for some time, and are maturing in their capability for handling Distributed Parallel Processing (DPP). Recently, however, JAVA, with all of its followers, has began to make an impact in the DPP arena as well. This paper will explore each of these three vehicles for DPP considering capability, ease of use, and availability. We will explore the programmer interface as well as explore their utility in solving real world parallel processing applications. We will show that each has its advantages. The bottom line is that programming a distributed cluster of workstations is challenging, worthwhile and fun!