Treffer: Implementation and testing of parallel PSO to attain speedup on general purpose computer systems.

Many real-world problems in the science and engineering field can be modeled as complex mathematical optimization problem and can be easily solved by swarm intelligence algorithms (SIAs). These optimization issues become more complicated due to using high dimensions and too many decision variables. The numbers of steps inside SIA used for optimization also enhance the complication of computation. These algorithms are iterative since many iterations of the procedure are to be done. The quality of the solution also depends on the number of iterations. So these repetitions or iterations again need more computation time. More computational power is required to solve any optimization problem through SIA, which is only suitable on the workstation or multiple parallel computers with high configurations. It is a little bit difficult to utilize the power of SIA effectively on an ordinary/ general-purpose system. Therefore, we proposed and examined a strategy and its implementation in this paper based on parallel computing, we have selected the most popular SIA known as Particle Swarm Optimization (PSO) algorithm and its parallel version. Along with unalike configurations and five standard benchmark functions, this analysis is performed over three system. The performance metrics such as speedup (S), real speedup (RS), maximum speedup (MS), efficiency (E), and scalability are taken into consideration for evaluation of the entire setup. On the basis of the results, we get to know that system 03 (S3) is finest among all, with an efficiency of 78.32%. [ABSTRACT FROM AUTHOR]

Copyright of Multimedia Tools & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)