Abstract
Parallelism is one of the main sources for performance improvement in modern computing environment, but the efficient exploitation of the available parallelism depends on a number of parameters. Determining the optimum number of threads for a given data parallel loop, for example, is a difficult problem and dependent on the specific parallel platform. This paper presents a learning-based approach to parallel workload allocation in a cost-aware manner. This approach uses static program features to classify programs, before deciding the best workload allocation scheme based on its prior experience with similar programs. Experimental results on 12 Java benchmarks (76 test cases with different workloads in total) show that it can efficiently allocate the parallel workload among Java threads and achieve an efficiency of 86% on average.
Original language | English |
---|---|
Title of host publication | Network and Parallel Computing |
Subtitle of host publication | IFIP International Conference, NPC 2007, Dalian, China, September 18-21, 2007. Proceedings |
Editors | Keqiu Li, Chris Jesshope, Hai Jin, Jean-Luc Gaudiot |
Publisher | Springer |
Pages | 506-515 |
Number of pages | 10 |
ISBN (Electronic) | 978-3-540-74784-0 |
ISBN (Print) | 978-3-540-74783-3 |
DOIs | |
Publication status | Published - Sept 2007 |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Publisher | Springer Berlin Heidelberg |
Volume | 4672 |
Keywords / Materials (for Non-textual outputs)
- parallelism
- workload allocation
- cost
- instance-based learning