DECOMPOSER: a synthesizer for systolic systems

A tool for synthesizing systolic systems is introduced. Given a hierarchical specification of the computations to be performed and hints as to how, this tool generates an analysis of the hardware required to the computations. The computations are specified as directed acyclic graphs, and the hints provide the temporal and topological relationships of each computation. The systolic system is synthesized by traversing the graph and marking each computation with a processor name and a time stamp. Its output can subsequently be fed to the remaining tools in the tool set to generate a VLSI fabrication description of the systolic system.<<ETX>>

[1]  Mary Jane Irwin,et al.  Being Stingy with Multipliers , 1990, IEEE Trans. Computers.

[2]  Peter R. Cappello,et al.  Unifying VLSI Array Designs with Geometric Transformations , 1983, International Conference on Parallel Processing.

[3]  Marina C. Chen,et al.  The Generation of a Class of Multipliers: Synthesizing Highly Parallel Algorithms in VLSI , 1988, IEEE Trans. Computers.

[4]  Marina C. Chen,et al.  A Design Methodology for Synthesizing Parallel Algorithms and Architectures , 1986, J. Parallel Distributed Comput..

[5]  Mary Jane Irwin,et al.  The Arithmetic Cube , 1987, IEEE Transactions on Computers.

[6]  Benjamin W. Wah,et al.  The Design of Optimal Systolic Arrays , 1985, IEEE Transactions on Computers.

[7]  Mary Jane Irwin,et al.  Digit-Pipelined Arnthmetic as Illustrated by the Paste-Up System: A Tutorial , 1987, Computer.

[8]  Mary Jane Irwin,et al.  A Digit Pipelined Dynamic Time Warp Processor , 1986 .

[9]  Patrice Quinton.  Automatic synthesis of systolic arrays from uniform recurrent equations , 1984, ISCA '84.

[10]  Mary Jane Irwin,et al.  An Overview of the Penn State Design System , 1987, 24th ACM/IEEE Design Automation Conference.