Authors: Imed Eddine Bennour; Monia Ettouil; Rim Zarrouk; Jemai Abderrazak
Addresses: Laboratory of Electronic and Microelectronic, Monastir, Avenue de l'Environnement – 5000 Monastir, Tunisia; National Engineering School at Sousse, 4023, Sousse, Tunisia ' Laboratory LIP2, University of Tunis El Manar, Faculty of Sciences at Tunis, Tunisia; University of Sousse, ISIT'Com, 4011, Hammem Sousse, Tunisia ' Laboratory of Electronic and Microelectronic, Faculty of Sciences at Monastir, 5019, Monastir, Tunisia; Tunisia Polytechnic School, La Marsa Tunisia ' Laboratory LIP2, University of Tunis El Manar, 2092, Tunisia; Faculty of Sciences at Tunis, INSAT, University of Carthage, 1080, Tunis, Tunisia
Abstract: Optimisation meta-heuristics such as particle swarm optimisation (PSO) require high-performance computing (HPC). The use of software parallelism and hardware parallelism is mandatory to achieve HPC. Thread-level parallelism is a common software solution for programming on multicore systems. The Java language, which includes important aspects such as its portability and architecture neutrality, its multithreading facilities and its distributed nature, makes it an interesting language to parallel PSO. However, many factors may impact the runtime performance: the coding styles, the threads-synchronisation levels, the harmony between the software parallelism injected into the code and the available hardware parallelism, the Java networking APIs, etc. This paper analyses the Java runtime performance on handling multithread PSO over general purpose multicore machines and networked machines. Synchronous, asynchronous, single-swarm and multi-swarm PSO variants are considered.
Keywords: high-performance computing; HPC; particle swarm optimisation; PSO; multicore; multithread; performance; simulation.
International Journal of Computational Science and Engineering, 2019 Vol.19 No.4, pp.483 - 493
Received: 04 Jan 2016
Accepted: 17 Jul 2016
Published online: 30 Aug 2019 *