Abstract
Cloud computing system is a progression of distributed system that has been adopted by worldwide scientifically and commercially. For optimal utilization of cloud’s potential power, effective and efficient algorithms are expected, which will select best resources from available cloud resources for different applications. This allocation of user requests to the cloud resources can optimize several parameters like energy consumption, makespan, throughput, etc. In this paper, we have proposed a learning automata based algorithm to minimize the makespan of the cloud system and also to increase the resource utilization that holds secured resource allocation. We have simulated our algorithm, ALOLA with the help of CloudSim simulator in a heterogeneous environment. During the comparison of the algorithm, we provide a finite set of tasks to the ALOLA algorithm once and estimate the makespan of the system. We have compared our proposed technique (ALOLA), i.e., with learning automata and without learning automata (random allocation algorithm), and show the system performance.
