A novel heuristic resolving deadline-oriented task scheduling in cloud

Cloud computing is a framework that allows easy network access to a shared pool of customizable computing resources, whenever required. It uses virtualization technology to provide computing resources to the clients which is why scheduling and resource allocation are a piece of important research issue in cloud computing. In order to exploit the ability of cloud computing, we need an effective task scheduling algorithm. A Scheduler should involve in scheduling client requests so as to achieve greater degree of load balancing and resource utilization. In this paper, an approach is portrayed, in which the basic computing elements are the virtual machines (VMs), termed as resources, which may be of varying sizes. The performance requirements are specified by the clients in terms of deadlines to the submitted tasks. Therefore, the goal is to ensure that all the assigned tasks are completed within their deadline and also that the tasks are evenly allocated across all the resources so that the utilization is effective. The algorithm is designed based on the striking features of the Max-Min algorithm and also on a novel concept termed as flexibility time. The objective of this determination is to harness the utilization of the internal data center and to minimize outsourcing tasks in the cloud, while attaining the applications´ quality of service(QOS) constraints.