Performance Comparison of Decentralized Workload Control through Code Migration in Distributed Desktop Computing Grids

Large scale loosely coupled PCs can organize a cluster and form desktop computing grids on sharing each processing power; power of PCs, transaction distributions and load balancing characterize the performance of the computing grids. This paper describes performance comparison of load balancing control through code migration in distributed desktop computing grids. A prototype of the middleware which migrates object code for computing was implemented, and the trade-off of code migration was observed. Based on the prototype experiment, several centralized and decentralized algorithms for location policy were examined on simulations, and the methodology to control code migration to have suitable response time in the system is derived. The result will be applied to design the grid computing systems and the peer to peer computing systems to maintain the proper performance.