Linköping University: Students Alumni Trade and Industry/Society Internal Search
elear_IWOCL14

Pattern Matching in OpenCL: GPU vs CPU Energy Consumption on Two Mobile Chipsets

Elena Aragon
 
Juan M. Jimenez
Arian Maghazeh
 
Jim Rasmusson
Unmesh D. Bordoloi

2nd International Workshop on OpenCL (IWOCL 2014), Bristol, May 12-14, 2014.

ABSTRACT
Adaptations of the Aho-Corasick (AC) algorithm on high performance graphics processors (also called GPUs) have garnered increasing attention in recent years. However, no results have been reported regarding their implementations on mobile GPUs. In this paper, we show that implement- ing a state-of-the-art Aho-Corasick parallel algorithm on a mobile GPU delivers significant speedups. We study a few implementation optimizations some of which may seem counter-intuitive to standard optimizations for high-end GPUs. More importantly, we focus on measuring the energy con- sumed by different components of the OpenCL application rather than reporting the aggregate. We show that there are considerable energy savings compared to the CPU im- plementation of the AC algorithm.


Related files:
elear_IWOCL14.pdfAdobe Acrobat portable document


[AMMR14] Elena Aragon, Juan M. Jimenez, Arian Maghazeh, Jim Rasmusson, Unmesh D. Bordoloi, "Pattern Matching in OpenCL: GPU vs CPU Energy Consumption on Two Mobile Chipsets", 2nd International Workshop on OpenCL (IWOCL 2014), Bristol, May 12-14, 2014.
( ! ) perl script by Giovanni Squillero with modifications from Gert Jervan   (v3.1, p5.2, September-2002-)