Parallel Processing of Sorting and Searching Algorithms Comparative Study

Saad AL-Azzam, Mohammad Qatawneh

Abstract


Recently, supercomputers structure and its software optimization have been popular subjects. Much of the software recently consumes a long period of time both to sort and search datasets, and thus optimizing these algorithms becomes a priority. In order to discover the most efficient sorting and searching algorithms for parallel processing units, one can compare CPU runtime as a performance index. In this paper, Quick, Bubble, and Merge sort algorithms have been chosen for comparison, as well as sequential and binary as search algorithms. Each one of the sort and search algorithms was tested in worst, average and best case scenarios. And each scenario was applied using multiple techniques (sequential, multithread, and parallel processing) on a various number of processors to spot differences and calculate speed up factor.

The proposed solution aims to optimize the performance of a supercomputer focusing one-time efficiency; all tests were conducted by The IMAN1 supercomputer which is Jordan's first and fastest supercomputer.


Full Text:

PDF


DOI: https://doi.org/10.5539/mas.v12n4p143

Copyright (c) 2018 Saad AL-Azzam, Mohammad Qatawneh

License URL: http://creativecommons.org/licenses/by/4.0

Modern Applied Science   ISSN 1913-1844 (Print)   ISSN 1913-1852 (Online)  Email: mas@ccsenet.org

Copyright © Canadian Center of Science and Education

To make sure that you can receive messages from us, please add the 'ccsenet.org' domain to your e-mail 'safe list'. If you do not receive e-mail in your 'inbox', check your 'bulk mail' or 'junk mail' folders.