Mobile Jacobi schemes for parallel computation

AbstractParallel computers (such as the distributed array professor and systolic arrays) bring into consideration the Jacobi method where several non-interacting rotations can be performed simultaneously. However, the design of the algorithm is much more crucial in a parallel environment; inefficiencies can lead to considerable organizationa costs. This paper provides a general framework for the description of mobile schemes together with two specific schemes, the better of which reduces the organizational overheads for the jacobi method to zero