MWAMLB: Modified Weighted Active Load Balancing Algorithm

Abstract

With the fast growth of technology and users, cloud computing has become an important IT paradigm where the resources are available online and on fly. Cloud computing is known for handling large amount of storage and computation data. In the cloud environment, the distinguishing feature of easy availability of resources makes their management a challenging task. One of the most important tasks is the balancing of the load among different virtual machines which in turn leads to proper utilization of resources and good response time. Many researchers have addressed the problem of resource provisioning, but the proactive approach has been gaining a lot of attention in recent years. The resource provisioning can be achieved either by allocating the resources judiciously or by predicting the demand in advance. The traditional methods make use of random selection of virtual machines(VMs) for load balancing. In this research work, a Modified Weighted Active Load Balancing framework (MWAMLB) has been offered with the emergence of cloud computing. The main objective of the MWAMLB framework is to improve the response time of the VM by selecting the virtual machine with maximum weight (W). The weight factor is being calculated on the basis of the availability of RAM, bandwidth and MIPS. The MWAMLB framework have been proposed, implemented and validated in this research paper.