A network multi‐processor for experiments in parallelism
- 1 June 1991
- journal article
- Published by Wiley in Concurrency: Practice and Experience
- Vol. 3 (3) , 203-219
- https://doi.org/10.1002/cpe.4330030306
Abstract
Although both shared memory and loosely coupled parallel computing systems are now common, many still do not offer an easy way to design, implement, and test parallel algorithms. Our system provides software tools that make possible a variety of connection structures between processes. These structures are said to form a 'Network Multi‐Processor', which is implemented on a local area network of heterogeneous UNIX‐based timesharing computers, plus a set of processor boards dedicated to an application so that accurate timing measurements can be made. We explain how these tools have been used both to aid parallel algorithm development and to explore the properties of different computer interconnection methods.Keywords
This publication has 19 references indexed in Scilit:
- A template-based approach to the generation of distributed applications using a network of workstationsIEEE Transactions on Parallel and Distributed Systems, 1991
- PVM: A framework for parallel distributed computingConcurrency: Practice and Experience, 1990
- Parallel computing comes of age: Supercomputer level parallel computations at CaltechConcurrency: Practice and Experience, 1989
- Partitioning techniques for large-grained parallelismIEEE Transactions on Computers, 1988
- A debugger for parallel processesSoftware: Practice and Experience, 1988
- Monitoring distributed systemsACM Transactions on Computer Systems, 1987
- The cosmic cubeCommunications of the ACM, 1985
- Parallel Search of Strongly Ordered Game TreesACM Computing Surveys, 1982
- A heterogeneous dual processorSoftware: Practice and Experience, 1980
- The UNIX time-sharing systemCommunications of the ACM, 1974