Parallel Implementation of Distributed Global Optimization (DGO)

TL;DR

This paper demonstrates that parallel implementation of distributed global optimization (DGO) significantly improves performance, reducing execution time from quadratic to near-linear scaling on parallel computers, with a maximum speedup of 126.

Contribution

The paper presents a parallel implementation of DGO that achieves near-linear performance scaling, effectively addressing the algorithm's original quadratic time complexity.

Findings

Performance increases approximately linearly with the number of processors.

Maximum speedup of 126 achieved on MP-1 with 128 processing elements.

Parallel implementation reduces execution time from O(n^2) to near O(n) for high-dimensional problems.

Abstract

Parallel implementations of distributed global optimization (DGO) [13] on MP-1 and NCUBE parallel computers revealed an approximate O(n) increase in the performance of this algorithm. Therefore, the implementation of the DGO on parallel processors can remedy the only draw back of this algorithm which is the O(n2) of execution time as the number of the dimensions increase. The speed up factor of the parallel implementations of DGO is measured with respect to the sequential execution time of the identical problem on SPARC IV computer. The best speed up was achieved by the SIMD implementation of the algorithm on the MP-1 with the total speedup of 126 for an optimization problem with n = 9. This optimization problem was distributed across 128 PEs of Mas-Par.