Resource Allocation with Efficient Load Balancing in Cloud Environment
Faouzia Zegrari, Abdellah Idrissi, Hajar Rehioui · 2016
Cloud computing is among the most active research fields. It provides users services and applications in distributed mode. It is a concept of deportation of applications and data on a remote infrastructure, but also of abstraction of infrastructure and material resources management to customers. The Datacenters are designed to host computing platforms and applications. The variability in the workload may be induced by the heterogeneity of these applications, which requires reasonably managing and distributing resources. The load distribution problem is a challenge that must be overcome. It would be more judicious to implement a load balancing model to optimize performance by distributing the workload fairly over all nodes in the Datacenter and ensuring more service availability. Thus, resource scheduling in the cloud is a complex problem. This paper presents a hybrid approach combining the Genetic Algorithm and the KÜHN Algorithm on which we develop a task allocation strategy by grouping tasks into clusters and fairly distributing the workload amongst a set of cooperating nodes. The aim of our study is to minimize response time and maximize resource utilization. The results of the experiment show that our approach provides good performance.