An optimization method of webp images lossy compression algorithm for fpgas

Yaqian Zhao, Zhenhua Guo, Baoyu Fan, Kai Gao, Li Wang, Fang Cao

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

To improve the efficiency of WebP images lossy compression algorithm for heterogeneous accelerating based on FPGAs in data center server, this paper proposed an optimization method of WebP image lossy compression algorithm for FPGAs. By adopting the optimization method of this paper, the overall throughput performance of WebP heterogeneous acceleration solution based on OpenCL language can be up to 281MPixel/s. The energy efficiency ratio is 2.29x higher than Intel Xeon E5-2690v3 processor. Compare with the industry-leading WebP heterogeneous acceleration solution of Xilinx and CTAccel, the throughput performance of this paper is 2.5x better than Xilinx and 1.8x better than CTAccel. Finally, this paper improves the utilization of RAM resources according to the characteristics of the RAM block in the FPGA.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE Intl Conf on Parallel and Distributed Processing with Applications, Big Data and Cloud Computing, Sustainable Computing and Communications, Social Computing and Networking, ISPA/BDCloud/SustainCom/SocialCom 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages705-711
Number of pages7
ISBN (Electronic)9781728143286
DOIs
Publication statusPublished - Dec 2019
Externally publishedYes
Event17th IEEE International Conference on Parallel and Distributed Processing with Applications, 9th IEEE International Conference on Big Data and Cloud Computing, 9th IEEE International Conference on Sustainable Computing and Communications, 12th IEEE International Conference on Social Computing and Networking, ISPA/BDCloud/SustainCom/SocialCom 2019 - Xiamen, China
Duration: 16 Dec 201918 Dec 2019

Publication series

NameProceedings - 2019 IEEE Intl Conf on Parallel and Distributed Processing with Applications, Big Data and Cloud Computing, Sustainable Computing and Communications, Social Computing and Networking, ISPA/BDCloud/SustainCom/SocialCom 2019

Conference

Conference17th IEEE International Conference on Parallel and Distributed Processing with Applications, 9th IEEE International Conference on Big Data and Cloud Computing, 9th IEEE International Conference on Sustainable Computing and Communications, 12th IEEE International Conference on Social Computing and Networking, ISPA/BDCloud/SustainCom/SocialCom 2019
Country/TerritoryChina
CityXiamen
Period16/12/1918/12/19

Keywords

  • Fpga
  • Image compression
  • Opencl
  • Webp

Fingerprint

Dive into the research topics of 'An optimization method of webp images lossy compression algorithm for fpgas'. Together they form a unique fingerprint.

Cite this