A dynamic load balancing method based on network delay for large distributed systems

The authors propose two dynamic load balancing methods which give considera tion to network delays in a large-scale, wide-area distributed system. In g eneral in load balancing, the number of candidate computers for transfers i ncreases as the number of computers being used increases, and so efforts ar e made to improve response time. There is, however, a concern that performa nce will drop due to delay overhead when candidate destinations for transfe rs include computers with large delays. Therefore, in this research the aut hors configure a collection of computers with low network delays as a duste r. They then propose a clustered dynamic load balancing method that emphasi zes load information within a cluster and delay information when transfers are performed between clusters, and a threshold dynamic load balancing meth od which restricts process destinations by creating a threshold value with respect to the ratio between the delay required for a process transfer and the average processing request time for a process. The results of a perform ance evaluation performed using simulations show that each proposed method can greatly improve the average response time compared to methods that do n ot consider delays. In addition, each method has an environment in which it is thought to be particularly effective given its operational characterist ics. (C) 2001 Scripta Technica.