Welcome to the InfoSci Platform
Optimizing Techniques for OpenCL Programs on Heterogeneous Platforms

Optimizing Techniques for OpenCL Programs on Heterogeneous Platforms

Slo-Li Chu, Chih-Chieh Hsiao
Copyright: © 2012 |Volume: 4 |Issue: 3 |Pages: 15
ISSN: 1938-0259|EISSN: 1938-0267|EISBN13: 9781466612334|DOI: 10.4018/jghpc.2012070103
Cite Article Cite Article

MLA

Chu, Slo-Li, and Chih-Chieh Hsiao. "Optimizing Techniques for OpenCL Programs on Heterogeneous Platforms." IJGHPC vol.4, no.3 2012: pp.48-62. http://doi.org/10.4018/jghpc.2012070103

APA

Chu, S. & Hsiao, C. (2012). Optimizing Techniques for OpenCL Programs on Heterogeneous Platforms. International Journal of Grid and High Performance Computing (IJGHPC), 4(3), 48-62. http://doi.org/10.4018/jghpc.2012070103

Chicago

Chu, Slo-Li, and Chih-Chieh Hsiao. "Optimizing Techniques for OpenCL Programs on Heterogeneous Platforms," International Journal of Grid and High Performance Computing (IJGHPC) 4, no.3: 48-62. http://doi.org/10.4018/jghpc.2012070103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Heterogeneous platforms that are consisted of CPU and add-on streaming processors are widely used in modern computer systems. These add-on processors provide substantially more computation capability and memory bandwidth than conventional multi-cores platforms. General-purpose computations can also be leveraged onto these add-on processors. In order to utilize their potential performance, programming these streaming processors is challenging because of their diverse underlying architectural characteristics. Several optimization techniques are applied on OpenCL-compatible heterogeneous platforms to achieve thread-level, data-level, and instruction-level parallelism. The architectural implications of these techniques and optimization principles are discussed. Finally, a case study of MRI-Q benchmark will be addressed to illustrate to capabilities of these optimization techniques. The experimental results reveal the speedup from non-optimized to optimized kernel can vary from 8 to 63 on different target platforms.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.