Evaluation of PGAS Communication Paradigms with Geometric Multigrid

Abstract

Partitioned Global Address Space (PGAS) languages and one-sided communication enable application developers to select the communication paradigm that balances the performance needs of applications with the productivity desires of programmers. In this paper, we evaluate three different one-sided communication paradigms in the context of geometric multigrid using the miniGMG benchmark. Although miniGMG's static, regular, and predictable communication does not exploit the ultimate potential of PGAS models, multigrid solvers appear in many contemporary applications and represent one of the most important communication patterns. We use UPC++, a PGAS extension of C++, as the vehicle for our evaluation, though our work is applicable to any of the existing PGAS languages and models. We compare performance with the highly tuned MPI baseline, and the results indicate that the most promising approach towards achieving performance and ease of programming is to use high-level abstractions, such as the multidimensional arrays provided by UPC++, that hide data aggregation and messaging in the runtime library.

Implementing High-Performance Geometric Multigrid Solver with Naturally Grained Messages

“…In our prior work [14], we developed a naturally grained version of the miniGMG benchmark. However, we did not attempt to optimize performance of the naive implementation.…”

“…HPGMG-FV solves the variable-coefficient PDE b∇·β∇u = f using the finite-volume method on a structured grid. As such, it is superficially similar to miniGMG [13], [18] which we used as a basis for our previous work [14]. However, unlike miniGMG, HPGMG uses Full Multigrid (FMG) and a distributed V-Cycle.…”

“…The strategy used by the GASNet VIS implementation is very similar to the array-based code described in our previous work [14]. We chose to use GASNet VIS rather than higher-level arrays to minimize the changes required to the naturally grained implementation of HPGMG, and because we did not require the generality provided by UPC ++ multidimensional arrays.…”