Exploiting locality on parallel irregular problem computations

By now, irregular problems are difficult to parallelize in an automatic way because of their lack of regularity in data access patterns. Most times, programmers must hand-write a particular solution for each problem separately. In this paper we present two pseudo-regular distributions which can be applied to partition most problems achieving very good average case distributions. Also, we have designed a run-time library (DDLY) which provides a fast way to apply data distribution on message passing environments. This library also is able of performing parallel input/output.<
>