Efficient Adaptive Algorithms for Transposing Small and Large Matrices on Symmetric Multiprocessors

Matrix transpose in parallel systems typically involves costly all-to-all communications. In this paper, we provide a comparative characterization of various efficient algorithms for transposing small and large matrices using the popular symmetric multiprocessors (SMP) architecture, which carries a relatively low communication cost due to its large aggregate bandwidth and lowlatency inter-process communication. We conduct analysis on the cost of data sending / receiving and the memory requirement of these matrix-transpose algorithms. We then propose an adaptive algorithm that can minimize the overhead of the matrix transpose operations given the parameters such as the data size, number of processors, start-up time, and the effective communication bandwidth. Key words: matrix transpose, SMP, MPI, all-to-all communication.

• Type: Journal
• Year: 2006
• Where: INFORMATICALT
• Authors: Rami Al Na'mneh, W. David Pan, Seong-Moo Yoo
• Keywords: All-to-all Communication, Communication, INFORMATICALT 2006, Transpose