Runtime performance of parallel array assignment
Lei Wang, James M. Stichnoth, Siddhartha Chatterjee · 1996
Compiling the array assignment statement of High Performance Fortran in the presence of block-cyclic distributions of data arrays is considered difficult, and several algorithms have been published to solve this problem. We present a comprehensive study of the performance of these algorithms. We classify these algorithms into several families and identify several issues of interest in the compilation process, and present experimental performance data for the various algorithms. We demonstrate that block-cyclic distributions can be compiled almost as efficiently as block and cyclic distributions.