FPGA-based vector processing for matrix operations

Yang Hongyan, Sotirios Ziavras, Hu Jie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

A programmable vector processor and its implementation with a field-programmable gate array (FPGA) are presented. This processor is composed of a vector core and a tightly coupled five-stage pipelined RISC scalar unit. It supports the IEEE 754 single-precision floating-point standard and also the efficient implementation of some sparse matrix operations. The processor is implemented on the Xilinx XC2V6000-5 FPGA chip. To test the performance, the W-matrix sparse solver for linear equations is realized. W-matrix was first proposed for power flow analysis and is prone to parallel computing. We show that actual power matrices with up to 1723 nodes can be dealt with in less than 1.1ms on the FPGA. A comparison with a commercial PC indicates that the vector processor is competitive for such computation-intensive problems.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Information Technology-New Generations, ITNG 2007
Pages989-994
Number of pages6
DOIs
StatePublished - Aug 28 2007
Event4th International Conference on Information Technology-New Generations, ITNG 2007 - Las Vegas, NV, United States
Duration: Apr 2 2007Apr 4 2007

Other

Other4th International Conference on Information Technology-New Generations, ITNG 2007
CountryUnited States
CityLas Vegas, NV
Period4/2/074/4/07

Fingerprint

Field programmable gate arrays (FPGA)
Processing
Reduced instruction set computing
Parallel processing systems
Linear equations

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems
  • Computer Science Applications

Cite this

Hongyan, Y., Ziavras, S., & Jie, H. (2007). FPGA-based vector processing for matrix operations. In Proceedings - International Conference on Information Technology-New Generations, ITNG 2007 (pp. 989-994). [4151833] https://doi.org/10.1109/ITNG.2007.95
Hongyan, Yang ; Ziavras, Sotirios ; Jie, Hu. / FPGA-based vector processing for matrix operations. Proceedings - International Conference on Information Technology-New Generations, ITNG 2007. 2007. pp. 989-994
@inproceedings{7fbd26677be1462ab9fd67bdfecad987,
title = "FPGA-based vector processing for matrix operations",
abstract = "A programmable vector processor and its implementation with a field-programmable gate array (FPGA) are presented. This processor is composed of a vector core and a tightly coupled five-stage pipelined RISC scalar unit. It supports the IEEE 754 single-precision floating-point standard and also the efficient implementation of some sparse matrix operations. The processor is implemented on the Xilinx XC2V6000-5 FPGA chip. To test the performance, the W-matrix sparse solver for linear equations is realized. W-matrix was first proposed for power flow analysis and is prone to parallel computing. We show that actual power matrices with up to 1723 nodes can be dealt with in less than 1.1ms on the FPGA. A comparison with a commercial PC indicates that the vector processor is competitive for such computation-intensive problems.",
author = "Yang Hongyan and Sotirios Ziavras and Hu Jie",
year = "2007",
month = "8",
day = "28",
doi = "https://doi.org/10.1109/ITNG.2007.95",
language = "English (US)",
isbn = "0769527760",
pages = "989--994",
booktitle = "Proceedings - International Conference on Information Technology-New Generations, ITNG 2007",

}

Hongyan, Y, Ziavras, S & Jie, H 2007, FPGA-based vector processing for matrix operations. in Proceedings - International Conference on Information Technology-New Generations, ITNG 2007., 4151833, pp. 989-994, 4th International Conference on Information Technology-New Generations, ITNG 2007, Las Vegas, NV, United States, 4/2/07. https://doi.org/10.1109/ITNG.2007.95

FPGA-based vector processing for matrix operations. / Hongyan, Yang; Ziavras, Sotirios; Jie, Hu.

Proceedings - International Conference on Information Technology-New Generations, ITNG 2007. 2007. p. 989-994 4151833.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - FPGA-based vector processing for matrix operations

AU - Hongyan, Yang

AU - Ziavras, Sotirios

AU - Jie, Hu

PY - 2007/8/28

Y1 - 2007/8/28

N2 - A programmable vector processor and its implementation with a field-programmable gate array (FPGA) are presented. This processor is composed of a vector core and a tightly coupled five-stage pipelined RISC scalar unit. It supports the IEEE 754 single-precision floating-point standard and also the efficient implementation of some sparse matrix operations. The processor is implemented on the Xilinx XC2V6000-5 FPGA chip. To test the performance, the W-matrix sparse solver for linear equations is realized. W-matrix was first proposed for power flow analysis and is prone to parallel computing. We show that actual power matrices with up to 1723 nodes can be dealt with in less than 1.1ms on the FPGA. A comparison with a commercial PC indicates that the vector processor is competitive for such computation-intensive problems.

AB - A programmable vector processor and its implementation with a field-programmable gate array (FPGA) are presented. This processor is composed of a vector core and a tightly coupled five-stage pipelined RISC scalar unit. It supports the IEEE 754 single-precision floating-point standard and also the efficient implementation of some sparse matrix operations. The processor is implemented on the Xilinx XC2V6000-5 FPGA chip. To test the performance, the W-matrix sparse solver for linear equations is realized. W-matrix was first proposed for power flow analysis and is prone to parallel computing. We show that actual power matrices with up to 1723 nodes can be dealt with in less than 1.1ms on the FPGA. A comparison with a commercial PC indicates that the vector processor is competitive for such computation-intensive problems.

UR - http://www.scopus.com/inward/record.url?scp=34548138017&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548138017&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/ITNG.2007.95

DO - https://doi.org/10.1109/ITNG.2007.95

M3 - Conference contribution

SN - 0769527760

SN - 9780769527765

SP - 989

EP - 994

BT - Proceedings - International Conference on Information Technology-New Generations, ITNG 2007

ER -

Hongyan Y, Ziavras S, Jie H. FPGA-based vector processing for matrix operations. In Proceedings - International Conference on Information Technology-New Generations, ITNG 2007. 2007. p. 989-994. 4151833 https://doi.org/10.1109/ITNG.2007.95