Task scheduling strategy based on similarity in peer to peer network
Xiaoyan Zhang · Computer Integrated Manufacturing Systems · 2007
To eliminate the BUMP phenomenon during task scheduling process and reduce the overhead in Peer to Peer(P2P) network,a task scheduling algorithm based on similarity was proposed which could make best use of the network resources.The nodes which could be directly scheduled were determined by computing similarity between current task and past task.To obtain the ordered nodes set,similarity between task needed resources and resources which could be supplied by nodes were computed and modified by distance factors.Based on it,the task scheduling process was completed according to the proposed algorithm.Experimental results showed that the proposed algorithm could effectively reduce the BUMP phenomenon and shorten tasks' execution time.