VMPP: A Virtual Machine for Parallel Processing

Abstract

The field of parallel processing is young and rapidly evolving. Consequently, there is a great diversity of languages and architectures. To make matters worse these languages and architectures often become obsolete at a rapid pace. In this environment, portability becomes an extremely important issue. Unfortunately, most parallel languages are not portable. This portability problem can be solved using a virtual machine approach. In this approach, front-end translators translate various parallel source languages into code for a virtual machine. Back-end translators translate the virtual machine code into executable codes for a variety of parallel architectures. The Virtual Machine for Parallel Processing (VMPP) is a proposal for just such a virtual machine. VMPP is designed to provide portability for a variety of high-level parallel programming languages without drastically sacrificing performance. It accomplishes this by defining a graphbased intermediate representation and a data-driven execution model. Table of Contents Note: Abstract extracted from PDF text