Performance assessment of multithreaded quicksort algorithm on simultaneous multithreaded architecture

  • Authors:
  • Basel A. Mahafzah

  • Affiliations:
  • King Abdullah II School for Information Technology, The University of Jordan, Amman, Jordan 11942

  • Venue:
  • The Journal of Supercomputing
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Sorting huge amounts of datasets have become essential in many computer applications, such as search engines, database and web-based applications, in order to improve searching performance. Moreover, due to the witnessed prevalence of the commercial Simultaneous Multithreaded architecture (SMT), parallel programming using multithreading becomes a dire need for efficiently using all available hardware resources for one application. In this paper, one of the efficient and quick algorithms, the Quicksort, is applied as a parallel multithreaded algorithm on SMT architecture, where virtual parallelization has been achieved using the POSIX threads (Pthreads) library. The proposed algorithm is evaluated and compared with its sequential counterpart. The obtained analytical and experimental results reveal that multithreading is a viable technique for implementing the parallel Quicksort algorithm efficiently on SMT architecture, where it has been shown both analytically and experimentally that the parallel multithreaded Quicksort algorithm outperforms the sequential Quicksort algorithm in terms of various performance metrics including; time complexity and speedup.