Jonathan M.D Hill and David B. Skillicorn.
In Journal of Future Generation Computer Systems, April 1998
Also available in High Perfomance Computing and Networking (HPCN'97), LNCS 1225, pages 762--771, Springer-Verlag, April 97, and Technical Report PRG-TR-16-96, Programming Research Group, Oxford University Computing Laboratory, August 1996.
Abstract:We focus on two criticisms of Bulk Synchronous Parallelism (BSP): that delaying communication until specific points in a program causes poor performance, and that frequent barrier synchronisations are too expensive for high-performance parallel computing. We show that these criticisms are misguided, not just about BSP but about parallel programming in general, because they are based on misconceptions about the origins of poor performance. The main implication for parallel programming is that higher levels of abstraction do not only make software construction easier---they also make high-performance implementation easier.