Abstract. Communication set generation significantly influences the performance of parallel programs. However, studies seldom give attention to the problem of communication set generation for irregular applications. In this paper, we propose communication optimization techniques for the situation of irregular array references in nested loops. In our methods, the local array distribution schemes are determined so that the total number of communication messages is minimized. Then, we explain how to support communication set generation at compile-time by introducing some symbolic analysis techniques. In our symbolic analysis, symbolic solutions of a set of symbolic expression are obtained by using certain restrictions. We introduce symbolic analysis algorithms to obtain the solutions in terms of a set of equalities and inequalities. Finally, experimental results on a parallel computer CM-5 are presented to validate our approach.

Abstract—Non-volatile memory technologies promise a vari-ety of advantages for memory architectures of next generation computing systems. However, these capabilities come at the cost of some inefficiencies governing the operation of these memories. The most well understood is the asymmetry of access. In order to most effectively take advantage of the benefits of these memory technologies in terms of density and reduced static power in systems while mitigating access complexity an one-size fits all method is not sufficient for all types of applications. Instead, cross-layer techniques that include the compiler, operating system, and hardware layer can extract characteristics from the application that can be used to deliver the highest possible performance while minimizing power consumption for systems using these memories.