Task Oriented Joint Multiple Resource Allocation Method for Cloud Computing Environment

Abstract

Cloud is a collection of interconnected computers with parallel and distributed system that are energetically presented as one or more unified computing resources. It is based on service-level agreements formed through negotiation between the service provider and consumers. This project work aims at developing  an Task Oriented Joint Multiple (TOJM) best fit method of resource allocation for cloud computing environments, assuming that more than one resources are assigned concurrently to each service request, after ranking the tasks to be scheduled. When the cloud computing service provider accepts tasks from users, using comparison matrix. The tasks are then ranked and based on weights assigned to various criteria of resource requirement using a process namely analysis hierarchy process. In this project, the Joint Multiple best fit resource allocation strategy is used for allocating the resources to the ranked tasks. This allocation strategy is based on the best-fit approach. The Joint Multiple strategy first identifies a resource whose requirement is maximum compared to the other resources required for that particular job and designates it as the identified resource. The scheme then finds a virtual machine which best suits the identified resource and checks if all other resources required for the task is also available in that Virtual Machine (VM). If so, all the resources are allocated jointly from that virtual machine. Otherwise, the scheme checks for the next virtual machine. If the available VMs do not satisfy the resource requirements, the task is rejected.