Parallel implementation of evolution strategy optimization algorithm on multicore processors

Abstract

PurposeThe purpose of this paper is to present a parallel implementation of an evolution strategy (ES) algorithm for optimization of electromagnetic devices. It is intended for multi‐core processors and for optimization problems that have objective function representing a numerical simulation of electromagnetic devices. The speed‐up of the optimization is evaluated as a function of the number of processor cores used.Design/methodology/approachTwo parallelization approaches are implemented in the program developed – using multithreaded programming and using OpenMP. Their advantages and drawbacks are discussed. The program is tested on two examples for optimization of electromagnetic devices.FindingsUsing the developed parallel ES algorithm on a quad‐core processor, the optimization time can be reduced 2.4‐3 times, instead of the expected four times. This is due to a number of system processes and programs that run on part of the cores.Originality/valueA new parallel ES optimization algorithm has been developed and investigated. The paper could be useful for researchers aiming to diminish the optimization time by using parallel evolution optimization on multi‐core processors.

Application of progressive quadratic response surface method for an oscillation problem optimization

“…PQRSM, the ES algorithm and PRBFRSM are compared (Table II) algorithm is efficient to find the global optimum. But it is time consuming (Ivanov and Brandisky, 2009). In the paper, ES algorithm is not detailed (Ivanov and Brandisky, 2009).…”

Optimization of a power mosfet and its monolithically integrated self-powering circuits

“…• some genetic algorithms: Evolution Strategy (ES) algorithm [5], Restricted Tournament Selection (RTS) with the self-adaptive Simulated Binary Crossover (vSBX) [6]. The optimization results are presented in Table 1.…”

“…(1) to Eqn. (3), the derivatives of F(a) according to X are defined by equations (4) and (5), where the multiple integrals are also computed by ASM [4]. Such an approach is time consuming, but is more accurate than a finite difference computation [3].…”

“…The paper focuses on the integration constraint of a power MOSFET, which will be partially presented in the next section. Then, the gradient-based optimization algorithms as the sequential quadratic programming (SQP) method [4] and several genetic optimization algorithms [5][6] are compared. Specifically, partial derivative computations are introduced and discussed.…”

Optimization Using Multicore-Processor Parallelization: A Case Study of Tax Processing System in Indonesia

“…There are three level of parallelism, instruction, data, thread, which can be used to improve the performance by using super scalar execution, multi-threading computation, and streaming Single Instruction Multi Data [4]. Beside of using the multithreading method, Open MP is the other method to do parallelization, and the result is not really different with the multi-thread method if we use the method in local parallelization case [5,6]. The way to make efficient processor performance is by exploiting their queuing methods [7].…”