PERFORMANCE EVALUATION OF A CLOUD BASED LOAD BALANCER SEVERING PARETO TRAFFIC
Ayman G. Fayoumi, Saudi Arabia · 2011
Cloud resources represent an unforeseeable breakthrough in ITC industry. Load balancer is a key element in resource provisioning for high available cloud solutions, and yet its performance depends on the traffic offered load. We develop a discrete event simulation to evaluate the performance with respect to the different load points. The performance metrics were the average waiting time inside the balance as well as the number of tasks. The performance study includes evaluating the chance of immediate serving or rejecting incoming tasks. Pareto traffic was considered for the offered traffic.