The impact of communication locality on large-scale multiprocessor performance
ISCA '92 Proceedings of the 19th annual international symposium on Computer architecture
Analysis of communications and overhead reduction in multithreaded execution
PACT '95 Proceedings of the IFIP WG10.3 working conference on Parallel architectures and compilation techniques
| Hi-index | 0.00 |
As multiprocessor sizes scale and computer architects turn to interconnection networks with non-uniform communication latencies, the lure of exploiting to increase performance becomes inevitable. Models that accurately quantify locality effects provide invaluable insight into the importance of exploiting locality as machine sizes and features change. This paper presents a framework for modeling the impact of communication locality on system performance. The framework provides a means for combining simple models of application, processor, and network behavior to obtain a combined model which accurately reflects feedback effects between processors and networks. We introduce a model that characterized application behavior with three parameters which capture computation grain, sensitivity to communication latency, and amount of locality present at execution time. The combined model is validated with measurements taken from a detailed simulator for a complete multiprocessor system. Using the combined model, we show that when the number of outstanding transactions per processor is bounded, exploiting communication locality provides gains which are at most linear in the factor by which average communication distance is reduced. The combined model is also used to obtain rough upper bounds on the performance improvement from exploiting locality to minimized communication distance.