Research on Predictable Scheduling Strategy of Virtual Computing Systems Based on Lookahead

Processor scheduling in computer science has been one of the most important research fields, and the scheduling of multiprocessor system is the most important extensively studied in depth. Lookahead computation is an effective method that improves the distributed simulation performance of stochastic petri nets (SPN). It enables us to analyze the structure of nets to reduce the delay time for partitioning the nets blindly. The M/M/n/k model with churn of virtual computing system is given by taking advantage of SPN. The predictable scheduling strategy based on lookahead is presented. The approach based on a part of optimism computed on the prediction time can determine advancement of each user task. The theorem shows that if every task is distributed to virtual machine for running along transmission path with least lookahead, the whole delay time of all tasks in buffer is the shortest.