A Parallelism Extended Approach for the Enumeration of Orthogonal Arrays
Orthogonal Array plays an important role in Design of Experiment and software testing. The most challenging aspect of Orthogonal array is the enumeration problem, that is for given parameter sets, we want to count exactly how many isomorphism classes exist. Enumeration of orthogonal array usually requires huge effort needed to complete its computation. There are several algorithms that have been proposed for enumeration of orthogonal array, however, there exists ineffective parallelism approach for handling this problem.
In this paper, we present a step-by-step parallelism extending approach for enumeration of orthogonal array. The experiments show that this proposed approach could get an relative speedup efficiency of up to 128 processes and a great dynamic load balancing between computing processes.
KeywordsExecution Time Orthogonal Array Relative Speedup Column Size Chunk Size
Unable to display preview. Download preview PDF.
- [DOS+07]Dinan, J., Olivier, S., Sabin, G., Prins, J., Sadayappan, P., Tseng, C.-W.: Dynamic load balancing of unbalanced computations using message passing. In: IEEE International Parallel and Distributed Processing Symposium, IPDPS 2007, pp. 1–8 (March 2007)Google Scholar
- [Mes97]Message-Passing Interface Forum. MPI-2.0: Extensions to the Message-Passing Interface, ch 9. MPI Forum (June 1997)Google Scholar
- [Ngu05]Nguyen, M.: Computer-algebraic methods for the construction of designs of experiments. Ph.D. Thesis, Technische Universiteit Eindhoven (2005)Google Scholar