Cluster Computing for Web-Scale Data Processing

Authors

  • M Venkataratnam
  • M. Basha
  • L. Hari Prasad

DOI:

https://doi.org/10.46243/jst.2022.v7.i09.pp199-211

Keywords:

MPEG-4, real-time, power-eflcient, motion estimation, content-based,

Abstract

Power efficiency and real-time processing capability are two major issues in today‘s mobile video applications. We proposed a novel Motion Estimation (ME) engine for power-efficient real-time MPEG- 4 video coding based on our previously proposed content-based ME al- gorithm [8], [13]. By adopting Full Search (FS) and Three Step Search (TSS) alternatively according to the nature of video contents, this algo- rithm keeps the visual quality very close to that of FS with only 3% of its computational power. We designed a flexible Block Matching (BM) Unit with 16-PE SIMD data path so that the adaptive ME can be performed at a much lower clock frequency and hardware cost as compared with previousFS based work. To reduce the energy cost caused by excessive external memory access, on-chip SRAM is also utilized and optimized for paral- lel processing in the BM Unit. The ME engine is fabricated with TSMC

0.18 µm technology. When processing QCIF (15 fps) video, the estimated power is 2.88 mW@4.16 MHz (supply voltage: 1.62 V). It is believed to bea favorable contribution to the video encoder LSI design for mobile appli-cations.

 

 

Downloads

Published

2022-11-30

How to Cite

M Venkataratnam, M. Basha, & L. Hari Prasad. (2022). Cluster Computing for Web-Scale Data Processing. Journal of Science & Technology (JST), 7(9), 199–211. https://doi.org/10.46243/jst.2022.v7.i09.pp199-211