Author(s): Said Fathy El-Zoghdy
Journal: International Journal of Computer Networks & Communications
ISSN 0975-2293
Volume: 4;
Issue: 1;
Start page: 113;
Date: 2012;
VIEW PDF
DOWNLOAD PDF
Original page
Keywords: Grid Computing | Resource Management | Load Balancing | Performance Evaluation | Queuing Theory | Simulation Models
ABSTRACT
This paper addresses the problem of scheduling and load balancing in heterogeneous computationalgrids. We proposed a two-level load balancing policy for the multi-cluster grid environment wherecomputational resources are dispersed in different administrative domains or clusters which are locatedin different local area networks. The proposed load balancing policy takes into account the heterogeneityof the computational resources. It distributes the system workload based on the processing elementscapacity which leads to minimize the overall job mean response time and maximize the system utilizationand throughput at the steady state. To evaluate the performance of the proposed load balancing policy,an analytical model is developed. The results obtained analytically are validated by simulating the modelusing Arena simulation package. The results show that the overall mean job response time obtained bysimulation is very close to that obtained analytically. Also, the simulation results show that theperformance of the proposed load balancing policy outperforms that of the Random and Uniformdistribution load balancing policies in terms of mean job response time. The maximum improvement ratioobtained when the system workload is low. It decreases slowly as the system workload graduallyincreases and it decreases rapidly when the system arrival rate λ approaches the system processing rateμ because the system gradually approaches its saturation point.
Journal: International Journal of Computer Networks & Communications
ISSN 0975-2293
Volume: 4;
Issue: 1;
Start page: 113;
Date: 2012;
VIEW PDF


Keywords: Grid Computing | Resource Management | Load Balancing | Performance Evaluation | Queuing Theory | Simulation Models
ABSTRACT
This paper addresses the problem of scheduling and load balancing in heterogeneous computationalgrids. We proposed a two-level load balancing policy for the multi-cluster grid environment wherecomputational resources are dispersed in different administrative domains or clusters which are locatedin different local area networks. The proposed load balancing policy takes into account the heterogeneityof the computational resources. It distributes the system workload based on the processing elementscapacity which leads to minimize the overall job mean response time and maximize the system utilizationand throughput at the steady state. To evaluate the performance of the proposed load balancing policy,an analytical model is developed. The results obtained analytically are validated by simulating the modelusing Arena simulation package. The results show that the overall mean job response time obtained bysimulation is very close to that obtained analytically. Also, the simulation results show that theperformance of the proposed load balancing policy outperforms that of the Random and Uniformdistribution load balancing policies in terms of mean job response time. The maximum improvement ratioobtained when the system workload is low. It decreases slowly as the system workload graduallyincreases and it decreases rapidly when the system arrival rate λ approaches the system processing rateμ because the system gradually approaches its saturation point.