Practical and effective domain-specific function unit design for CGRA

Authors:
Ming Yan;Ziyu Yang;Liu Yang;Lei Liu;Sikun Li
Affiliations:
School of Computer, National University of Defense Technology, Changsha, Hunan, China;School of Computer, National University of Defense Technology, Changsha, Hunan, China;College of Computer and Communication, Hunan University, Changsha, Hunan, China,;School of Computer, National University of Defense Technology, Changsha, Hunan, China;School of Computer, National University of Defense Technology, Changsha, Hunan, China
Venue:
ICCSA'11 Proceedings of the 2011 international conference on Computational science and Its applications - Volume Part V
Year:
2011

Citing 15
Cited 0

PipeRench: a co/processor for streaming multimedia acceleration

ISCA '99 Proceedings of the 26th annual international symposium on Computer architecture
Design and Implementation of the MorphoSys Reconfigurable ComputingProcessor

Journal of VLSI Signal Processing Systems - Special issue on VLSI on custom computing technology
Instruction generation for hybrid reconfigurable systems

ACM Transactions on Design Automation of Electronic Systems (TODAES)
Automatic application-specific instruction-set extensions under microarchitectural constraints

Proceedings of the 40th annual Design Automation Conference
Automated Custom Instruction Generation for Domain-Specific Processor Acceleration

IEEE Transactions on Computers
Application Specific Datapath Extension with Distributed I/O Functional Units

VLSID '07 Proceedings of the 20th International Conference on VLSI Design held jointly with 6th International Conference: Embedded Systems
Polynomial-time subgraph enumeration for automated instruction set extension

Proceedings of the conference on Design, automation and test in Europe
Fast, quasi-optimal, and pipelined instruction-set extensions

Proceedings of the 2008 Asia and South Pacific Design Automation Conference
The reconfigurable instruction cell array

IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Rapid design of area-efficient custom instructions for reconfigurable embedded processing

Journal of Systems Architecture: the EUROMICRO Journal
Fast custom instruction identification by convex subgraph enumeration

ASAP '08 Proceedings of the 2008 International Conference on Application-Specific Systems, Architectures and Processors
Fast enumeration of maximal valid subgraphs for custom-instruction identification

CASES '09 Proceedings of the 2009 international conference on Compilers, architecture, and synthesis for embedded systems
Exact and approximate algorithms for the extension of embedded processor instruction sets

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Fast Identification of Custom Instructions for Extensible Processors

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
CHIPS: Custom Hardware Instruction Processor Synthesis

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Coarse-grained reconfigurable architecture (CGRA) brings more powerful performance than the traditional CPU-like architecture. In this paper, we proposed a fast and effective domain-specific design method for function units of a CGRA named ProDFA. The proposed design method mainly concludes a top-down subgraph enumeration algorithm and a heuristic subgraph identification process based on topological searching. We used a clustering technique to accelerate the maximal valid subgraph enumeration (MVSE), and for the first time the top-down MVSE is combined with the identification process through the topological ordering. Candidate convex subgraphs are finally grouped into function sets according to their isomorphism. Experimental result shows that the performance of subgraph enumeration is improved in most cases, and a small number of candidate function sets are identified.