Abstract: Cloud computing is distributed computing, which solves tasks through the network and merges computing results. Specifically, in this environment, the user-submitted jobs are first initialized, then the jobs are inserted into the execution queue according to the user-submitted time and priority scheduling according to the appropriate policy, and finally the jobs selected from the task scheduler are segmented and assigned to each node. The running load of some nodes may exceed their critical value, and the convergence speed is relatively slow in numerical analysis. In order to effectively solve these problems, this paper conducts in-depth research on scheduling algorithm and optimization.

Key words: Cloud computing, Task scheduling, Scheduling optimization, Decision making problems