Data Distribution Analysis and Optimization for Pointer-Based Distributed Programs

High Performance Fortran (HPF) provides distributed arrays to eeciently support a global name space on distributed memory architectures. The distributed data structures supported by HPF, however , are only limited to array constructs, and do not extend to pointer-based distributed structures. With the support of distributed pointers and class abstractions in parallel C++-like languages, application programmers now can build their own pointer-based aggregate objects to provide a global name space on distributed environments. However, a critical question remains if the compiler can understand the distribution pattern of pointer-based distributed objects built by application programmers, and perform optimization as eeectively as the HPF compiler does with distributed arrays. In this paper , we address this challenging issue. In our work, we rst present a parallel programming model which allows application programmers to build pointer-based distributed objects at application levels. Next, we propose a distribution analysis algorithm which can automatically summarize the distribution pattern of pointer-based distributed objects built by application programmers. Our work, to our best knowledge, is the rst work to attempt to address this open issue. Our distribution analysis framework employs Feautrier's parametric integer programming as the basic solver, and can always obtain precise distribution information from the class of programs written in our parallel programming model with static control. Experimental results done on a 16-node IBM SP-2 machine show that the compiler with the help of distribution analysis algorithm can signiicantly improve the performance of pointer-based distributed programs.