ParaDict, a data parallel library for dictionaries

Abstract

We present ParaDict, a library for data parallel dictionaries written in C* having two different interfaces. The first one is written in C* for data parallel users and the second one in C for users that want to use a parallel library but not to write parallel programs. We have seen that C* is an adequate tool to code theoretical PRAM algorithms into readable programs. We run these programs on a CM 200 with better times than other existing implementations. If we compare with a sequential implementation on a workstation they have much better asymptotic behaviour. Finally, we explore the relationship between data parallelism and vectorization. We transform C* code into C code plus compiler directives and we run the result on a Convex C3480 machine. Even if (almost all) the loops were vectorized, the performances were modest. All these facts allow us to look at the development of other libraries with moderate optimism.