User behavior-based resource scheduling mechanism for cloud computing with feedback control

The ever-increasing quantity and variability of both user requirements and resources poses many challenges to cloud resource scheduling. One challenge is how to identify the real needed resources for users with diverse requirements. In response to this issue, the problem of cloud resource scheduling is studied from the perspective of users is studied, which is based on user behaviors. And a user behavior-based resource scheduling mechanism with feedback control (UBRSM-FC) is further developed. UBRSM-FC integrates the user behaviors information collected from scheduling history into the resource scheduling via the relevance feedback mechanism. UBRSM-FC establishes an active discovery network and feedback network for user requirements to perform continuous feedback on the scheduling process which can make scheduling results closer and closer to the real purpose of users. Simulation results confirm that this user behavior-based scheduling mechanism is much more efficient in satisfying the diverse requirements of users, and also performs well on the resource utilization rate of the cloud system. © 2018, Editorial Office of Systems Engineering and Electronics. All right reserved.