Enhanced load aware weighted round robin algorithm in cloud

In a large-scale distributed cloud network there are lots of factors which affect the performance of the distributed systems, among this load balancing scheduling has a huge impact. It is recommendable to have a load balancer which equally splits the workload among all the available servers, according to the required parameters. The parameters of each server or a request can be termed as heavy loads or light loads relative to one another. Therefore, in the cloud environment, we need to assess the server capacity to overcome the high traffic and balance the loads properly. Subsequently, there is a need of a dynamic load balancing algorithm which splits and distribute all the loads equally on the basis of different parameters of the servers. The main aim of this research work is to address these requirements by devising a dynamic load aware load balancer for a heterogenous cloud environment. This is used to determine the weights of each queue dynamically based on the current traffic characteristics and static weights assigned to each server. The aim is to improve the average throughput and also to reduce the packet loss in the cloud networks. The experimentation of the proposed algorithm is performed using a simulator and the simulation results prove that there is a better improvement in the performance of the load balancer and also improves the average throughput compared with the existing WRR.