Presenting a Clustering Method and Improved Firefly Algorithm for Load Balancing in Cloud Systems

Due to the increasing use of computer systems and the expansion of the use of cloud services, one of the main challenges is resource management. Resources should be managed in such a way that the input load is distributed evenly between different systems in order to reduce energy consumption and prevent service quality violations by balancing the servers in the cloud. The proposed algorithm is a combination of clustering method and a firefly algorithm. In the first step, Bitonic sorting is used to cluster the requests to create balanced clusters, and in the next step, by placing these clusters in the firefly algorithm, the tasks are assigned to the servers. The results of the experiments show that the proposed method improve the average energy consumption by ۱۹%, the task run-time by ۱۷% and the SLA violation rate by ۵% compared to the meta-heuristic algorithms such as the ant colony, genetic, firefly and cuckoo search

کلیدواژه ها:

load balance ، Cloud Computing ، Clustering ، Firefly algorithm ، Bitonic sorting