Priority based resource allocation and scheduling using artificial bee colony (ABC) optimization for cloud computing systems

Abstract

In the Cloud computing systems, the existing works fails to address the failure rate of servers and no mechanism has been provided for fault recovery. In this paper, we propose a Priority based Resource Allocation & Scheduling Technique using Artificial Bee Colony (ABC) Optimization (PRAS-ABC) for cloud environment. Initially, the work load of server and resource requirements of users are predicted by the scout bees by monitoring the past resource utilizations and size of the allocated VM. With this predicted workload, the expected completion time of each server is estimated. Then the tasks requesting the resources are categorized based on the deadline and resource requirements. Then based on the work load and expected completed time, the servers are categorized. Then each category of task will be allocated to the respective category of servers. The proposed approach is implemented in CloudSim environment of Java and compared with existing techniques in terms of resource utilization, percentage of resources successfully allocated, percentage of missed deadlines, average work load of server etc.