OpenMP versus MPI for PDE Solvers Based on Regular Sparse Numerical Operators

Two parallel programming models represeted by OpenMP and MPI are compared for PDE solvers based on regular sparse numerical operators. As a typical representative of such an application, the Euler equations for fluid flow are considered.The comparison of programming models is made with regard to UMA, NUMA, and self optimizing NUMA (NUMA-opt) computer architectures. By NUMA-opt, we mean NUMA systems extended with self optimizations algorithms, in order to reduce the non-uniformity of the memory access time.The main conclusions of the study are: (1) that OpenMP is a viable alternative to MPI on UMA and NUMA-opt artitectures, (2) that OpenMP is not competitive on NUMA platforms, unless special care is taken to get an initial data placement that matches the algorithm, and (3) that for OpenMP to be competitive in the NUMA-opt case, it is not necessary to extend the OpenMP model with additional data distribution directives, nor to include user-level access to the page migration library.

[1]  Erik Hagersten,et al.  Simple COMA node implementations , 1994, 1994 Proceedings of the Twenty-Seventh Hawaii International Conference on System Sciences.

[2]  Lisa Noordergraaf,et al.  Performance experiences on Sun's Wildfire prototype , 1999, SC '99.

[3]  E. Ayguade,et al.  Scaling Irregular Parallel Codes with Minimal Programming Effort , 2001, ACM/IEEE SC 2001 Conference (SC'01).

[4]  Jonathan Harris,et al.  Extending OpenMP For NUMA Machines , 2000 .

[5]  William Gropp,et al.  High-performance parallel implicit CFD , 2001, Parallel Comput..

[6]  Erik Hagersten,et al.  WildFire: a scalable path for SMPs , 1999, Proceedings Fifth International Symposium on High-Performance Computer Architecture.

[7]  Sverker Holmgren,et al.  Performance of High-Accuracy PDE Solvers on a Self-Optimizing NUMA Architecture , 2001, Euro-Par.

[8]  Eduard Ayguadé,et al.  Is Data Distribution Necessary in OpenMP? , 2000, ACM/IEEE SC 2000 Conference (SC'00).