Authors: Jia-Hwa Wu, Chih-Ping Chu
Addresses: Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan 701, ROC. ' Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan 701, ROC
Abstract: Two-dimensional arrays occur quite frequently in real programmes. In general, for multi-dimensional arrays under constant bounds, the Lambda test is an efficient data dependence method to check whether there exist real solutions. In this paper, we propose a multi-dimensional Interval Reduction (IR) test. The multi-dimensional IR test can be applied towards testing whether there are integer solutions for multi-dimensional arrays under constant limits, increasing the testing precision and exploiting the degree of loop parallelisation and vectorisation. Experiments with benchmarks showing the effects of the multi-dimensional IR test are also presented.
Keywords: parallelising compilers; data dependence analysis; loop parallelisation; vectorisation; high performance computing; interval reduction tests; multi-dimensional arrays.
International Journal of High Performance Computing and Networking, 2006 Vol.4 No.5/6, pp.331 - 337
Available online: 01 May 2007 *Full-text access for editors Access for subscribers Purchase this article Comment on this article