Although the existing scheduling algorithm of intelligent network disk （ IND） storage system minimizes the weighted time cost of tasks,it may result in a starvation phenomenon because many tasks cannot be executed for a long time. In order to solve this problem,an controllable scheduling algorithm （ CSA） is proposed,which divides the tasks into several step chains,computes the ρ factor of tasks according to the execution time and weight of step chains,and selects the unexecuted step chain according to both the ρ factor and the execution time of the tasks of ρ factor. Experimental results indicate that,as compared with the existing IND algorithm,the proposed CAS is of similar total weighted time cost,lower average turnaround time cost of tasks and completion time cost of short tasks, and that it effectively eliminates the starvation with balanced efficiency and fairness.