Research and implementation of asynchronous concurrent scheduling model for online system

Abstract

Considering the dynamically changing characteristic of the online system's scale and execution time, firstly, this paper proposes a multi-application asynchronous concurrent scheduling model. It is based on the distributed computing parallel scheduling model of coarse grain tasks and aimed at balancing load and reducing communication time. Then the paper introduces how to implement the model. On the case of multi-application concurrent executing, the idle time of the cluster nodes is lessened by using asynchronous concurrent mechanism. The corresponding and dispatching time is decreased via optimizing the grain of every application. Finally, a power system computation example is used to verify the effectiveness of the developed model. © 2011 Springer-Verlag.