An effective time based load balancer for an academic cloud environment

The proliferation of Internet and mobile phone usage among students led to the state-of-the-art technology in cloud based e-Learning. An academic cloud based e-Learning at its initial stage will act as a private cloud with certain Service Level Agreements (SLA) and limited resources. Therefore, only limited virtualized resources available in an infrastructure can be utilized by the end-users of an academic institution. In order to provide an effective utilization, a proper balancing of workload is needed. Hence in the adduced work, mapping of the user´s requested Virtual Machine (VM) configuration exactly with host configuration is done using Three Phase Filtering Protocol (3PFP). An efficient time based load balancer (TBLB) algorithm is also developed in order to distribute the load evenly among the host machines based on last virtual machine executed date and time (LVMEDT). The performance of the proposed algorithm is analyzed using the cloud testbed.