Optimisation and Performance Evaluation of Mechanisms for Latency Tolerance in Remote Memory Access Communication on Clusters
This paper describes the design and performance evaluation of mechanisms for latency tolerance in the remote memory access communication on clusters equipped with high-performance networks such as Myrinet. It discusses strategies that bridge the gap between user-level requirements and network-specific communication interfaces while attempting to increase opportunities for latency hiding. Mechanisms for overlapping communication with computation and coalescing small messages (trading latency for bandwidth) are explored. The effectiveness of these techniques is evaluated using microbenchmarks and application kernels including the NAS parallel benchmark suite. The microbenchmark results showed a much better degree of overlap for nonblocking operations in ARMCI as compared to MPI. Application results showed up 30% to 45% improvement over MPI on using nonblocking operations. The aggregation of small messages yielded performance improvement of up to 78% over non-aggregated communication.
- Research Organization:
- Pacific Northwest National Laboratory (PNNL), Richland, WA (US)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC05-76RL01830
- OSTI ID:
- 993675
- Report Number(s):
- PNNL-SA-41796; KJ0101030
- Journal Information:
- International Journal of High Performance Computing and Networking, 2(2-4):198-209, Journal Name: International Journal of High Performance Computing and Networking, 2(2-4):198-209 Journal Issue: 2-4 Vol. 2
- Country of Publication:
- United States
- Language:
- English
Similar Records
Evaluation of Remote Memory Access and Global Arrays Programming Model on the Blue Gene/P Supercomputer
High Performance Remote Memory Access Communication: The ARMCI Approach