A Study of Pipelining in Computing Arrays
- 1 May 1986
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. C-35 (5) , 431-440
- https://doi.org/10.1109/tc.1986.1676785
Abstract
In this paper, we take a hard look at scheduling considerations in computing arrays. A simple sufficient condition is developed for determining whether a computing array can be pipelined. If the array cannot be pipelined in the form given, the condition also indicates the direction in which to proceed to make it pipelineable. The overall framework and methodology take a good part of the load off the logical architect of the array, and make the translation from the logical to the physical architecture a mechanical process.Keywords
This publication has 10 references indexed in Scilit:
- A graph theoretic technique for the generation of systolic implementations for shift-invariant flow graphsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- A parallel architecture for Kalman filter measurement update and parameter estimationAutomatica, 1986
- On supercomputing with systolic/wavefront array processorsProceedings of the IEEE, 1984
- Optimizing Synchronous Circuitry by Retiming (Preliminary Version)Published by Springer Nature ,1983
- Wavefront Array Processor: Language, Architecture, and ApplicationsIEEE Transactions on Computers, 1982
- Why systolic architectures?Computer, 1982
- Optimizing synchronous systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1981
- The maximum sampling rate of digital filters under hardware speed constraintsIEEE Transactions on Circuits and Systems, 1981
- A Matrix Data Flow Language/Architecture for Parallel Matrix Operations Based on Computational Wavefront ConceptPublished by Springer Nature ,1981
- Efficient Algorithms for Shortest Paths in Sparse NetworksJournal of the ACM, 1977