Improved GRASP Technique based Resource Allocation in the Cloud

In the era of cloud computing, everyone is somehow using cloud resources. However, the resources are limited in the Cloud. Cloud vendors look for enhanced returns on investments. Promising return on investment is possible only when the cloud resources are scheduled efficiently to execute jobs within the stipulated time. However, brute force methods require exponential time to produce a schedule. Heuristic and meta-heuristic algorithms have been proposed in the literature to allocate resources to the jobs. These algorithms still suffer from slow convergence. To overcome this problem, researchers clubbed various heuristics and meta-heuristic to form a new hybrid algorithm. With the same motive, this paper explores the limitations of greedy random adaptive search and shows that learning through a fixed set search enhances efficiency. Based on the results, it can be concluded that the proposed algorithm is on par with existing hybrid meta-heuristic algorithms.