Affiliation:
1. Center for Supercomputing Research and Development, University of Illinois at Urbana-Champaign, Urbana, Illinois
Abstract
In this paper, we study the impact of synchronization and granularity on the performance of parallel systems using an execution-driven simulation technique. We find that even though there can be a lot of parallelism at the fine grain level, synchronization and scheduling strategies determine the ultimate performance of the system. Loop-iteration level parallelism seems to be a more appropriate level when those factors are considered. We also study barrier synchronization and data synchronization at the loop iteration level and found both schemes are needed for a better performance.
Publisher
Association for Computing Machinery (ACM)
Reference25 articles.
1. Automatic decomposition of scientific programs for parallel execution
2. Alliant. FX/Series Architecture Manual. Alliant Computer Systems Corp. Jan. 1986. Alliant. FX/Series Architecture Manual. Alliant Computer Systems Corp. Jan. 1986.
3. D.-K. Chen. MaxPar: An Execution Driven Simulator for Studying Parallel Systems. CSRD T&917 Center for Supercomputing Research and Development Univ. of lllinois at Urbana-Champaign Sep. 1989. D.-K. Chen. MaxPar: An Execution Driven Simulator for Studying Parallel Systems. CSRD T&917 Center for Supercomputing Research and Development Univ. of lllinois at Urbana-Champaign Sep. 1989.
Cited by
7 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
1. Compiling Loop-Based Nested Parallelism for Irregular Workloads;Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2;2024-04-27
2. Exploiting Coarse-Grained Parallelism Using Cloud Computing in Massive Power Flow Computation;Energies;2018-08-29
3. Communication-avoiding parallel minimum cuts and connected components;ACM SIGPLAN Notices;2018-03-23
4. Communication-avoiding parallel minimum cuts and connected components;Proceedings of the 23rd ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming;2018-02-10
5. MHD code using multi graphical processing units: SMAUG+;Advances in Space Research;2018-01