Buffer-safe communication optimization based on data flow analysis and performance prediction
Thomas Fahringer, Eduard Mehofer · 1997
This paper presents a novel approach to reduce communication costs of programs for distributed memory machines. Our techniques are based on uni-directional bit-vector data flow analysis that enable vectorizing and coalescing communication, overlapping communication with computation, eliminating redundant messages and amount of data being transferred both within and across loop nests. Our data flow analysis differs from previous techniques that it does not require to explicitly model balanced communication placement and loops and does not employ interval analysis. Our techniques are based on simple yet highly effective data flow equations which are solved iteratively for arbitrary control flow graphs. Moving communication earlier to hide latency has been shown to dramatically increase communication buffer sizes and can even cause runtime errors. We use P³T, a state-of-the-art performance estimator to create a buffer-safe program. By accurately estimating both the communication buff...