Volume : 6, Issue : 6, June - 2017

ANALYSIS OF VARIOUS SORTING ALGORITHMS IN OPENCL

Tanvi, Shreshtha Sharma

Abstract :

<p>&nbsp;<i style="text-align: justify; font-size: 12px;"><span lang="EN-GB" style="font-size: 10pt; line-height: 99%;">With the availability of multi-core processors and graphics processing units in the market, heterogeneous computing environment with the immense performance capability can be easily constructed. Heterogeneous computing accelerates the performance by transferring the computation intensive code to the GPU and the remaining code runs in the CPU. OpenCL is a standardized framework for the heterogeneous computing. Sorting algorithms are considered as fundamental building blocks for algorithms and has many applications in computer science and technology. In this paper, we implement Selection Sort, Btonic Sort and Radix Sort in OpenCL. The algorithms are designed to exploit the parallelism model available on multi-core GPUs with the help of OpenCL specification. In addition, the comparison between the traditional sequential sorting algorithms and parallel sorting algorithms are made on the Intel &reg; Core&trade; i5-3317U CPU @ 1.70 GHz architecture and AMD Radeon HD 7600 M series GPU.</span></i></p> <p class="MsoNormal" style="margin-bottom:12.0pt;text-align:justify;line-height:&#10;99%;mso-pagination:none;mso-layout-grid-align:none;punctuation-wrap:simple;&#10;text-autospace:none"><i><span lang="EN-GB" style="font-size:10.0pt;&#10;line-height:99%"><o:p></o:p></span></i></p>