The Parallelization of Level 2 and 3 BLAS Operations on Distributed Memory Machines
Mokhtar A. Aboelaze, Nikos P. Chrisochoides, Elias N. Houstis, Catherine E. Houstis · Purdue e-Pubs (Purdue University System) · 1991
We present the parallelization of band matrix-vector and band matrix-matrix product operations on distributed memory multiprocessor systems that support a mesh and ring interconnection topology.Our approach eliminates synchronization delay and minimizes the communication overhead among processors.Three of these operations has