Communication Primitives for Minimally Synchronous Parallel ML

Abstract

AbstractMinimally Synchronous Parallel ML is a functional parallel language whose execution time can then be estimated and dead-locks and indeterminism are avoided. Programs are written as usual ML programs but using a small set of additional primitives. It follows the cost model of the Message Passing Machine model (MPM). This paper explore two versions of an additional communication function: one uses this small set of primitives, the other one is considered as a primitive and implemented at a lower level.KeywordsExecution TimeParallel MachineCost ModelParallel VectorCommunication PhaseThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.