Modeling Analysis of Key Message Approach on Clusters

摘要: Research on cluster/distributed computing has been fueled by the availability of cost-effective workstations and high performance networks. Many parallel programming languages related models have become widely accepted. However, communication overhead is a major problem running applications in environments. To reduce thus completion time application, this paper introduces analyzes an efficient Key Message (KM) approach to supporting cluster The includes application model model. adopts DAG (directed acyclic graph) task graph abstract application. presents priority-based M/M/l queue analyze delay. Then, with these fundamental assumptions, analytical method presented verify approach. It demonstrates that when network background load increases or computation ratio decreases, analysis results show significant improvement over system which does not use KM