Parallel program execution on a network of workstations

ABSTRACT

In this project we study the parallel execution of programs on a collection of workstations connected by a local area network. So far, our work has focused on developing a number of pilot applications. We intend to use these applications as an experimental testbed for studying issues in problem partitioning, fault tolerance, load sharing, and monitoring and debugging of distributed programs. In this overview we focus our attention on one particular application (a distributed compiler) and on our work in fault-tolerance.