A study of implementation schemes for vectorized sparse EBE matrix-vector multiplication
Advances in Engineering Software
Lazy release consistency for software distributed shared memory
ISCA '92 Proceedings of the 19th annual international symposium on Computer architecture
Relaxed consistency and coherence granularity in DSM systems: a performance evaluation
PPOPP '97 Proceedings of the sixth ACM SIGPLAN symposium on Principles and practice of parallel programming
Efficient Categorization of Sharing Patterns in Software DSM Systems
IPDPS '01 Proceedings of the 15th International Parallel & Distributed Processing Symposium
TreadMarks: distributed shared memory on standard workstations and operating systems
WTEC'94 Proceedings of the USENIX Winter 1994 Technical Conference on USENIX Winter 1994 Technical Conference
| Hi-index | 0.00 |
In this work we approach the seismic wave propagation problem in two dimensions using the finite element method (FEM). This kind of problem is essential to study the structure of the earth's interior and exploring petroleum reservoirs. Using a representative FEM-based application, we propose and evaluate two parallel algorithms based on the inverse mapping and on the mesh coloring, respectively. The distinguishing feature of our parallel versions is that they were implemented in a distributed shared-memory system (SDSMs), which offers the intuitive shared-memory programming model on a cluster of low-cost high-performance PCs. Our results for several workloads show that the inverse mapping scheme achieved the best speedup at 7.11 out of 8 processors, though the mesh coloring algorithm scales well. Overall, these preliminary results we obtained indicate that cluster-based SDSMs represent a cost-effective friendly-programming platform for developing parallel FEM-based applications.