Experiments with Dataflow on a General-Purpose Parallel Computer.
Abstract
The MIT J-Machine, a massively-parallel computer, is an experiment in providing general-purpose mechanisms for communication, synchronization, and naming that will support a wide variety of parallel models of computation. Having universal mechanisms allows the separation of issues of language design and machine organization. We have developed two experimental dataflow programming systems for the J-Machine. For the first system, we adapted Papadopoulos' explicit token store to implement static and then dynamic dataflow. Each node in a dataflow graph is expanded into a sequence of code, each of which is scheduled individually at runtime. For a later system, we made use of Iannucci's hybrid execution model to combine several dataflow graph nodes into a single sequence, decreasing scheduling overhead. By combining the strengths of the two systems, it is possible to produce a system with competitive performance. We have demonstrated the feasibility of efficiently executing dataflow programs on a general purpose parallel computer.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1991
- Accession Number
- ADA235521
Entities
People
- Bill Dally
- Ellen Spertus
Organizations
- Massachusetts Institute of Technology