International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 77 - Number 10 |
Year of Publication: 2013 |
Authors: Ankita Agrawal, Rudesh Shah |
10.5120/13427-1120 |
Ankita Agrawal, Rudesh Shah . Host Load Prediction in Computational Grid Environment. International Journal of Computer Applications. 77, 10 ( September 2013), 1-6. DOI=10.5120/13427-1120
When sudden load arises in a grid then load should be transferred to some idle node hence due to load sharing, server down condition not occur hence we predict the load on node and shared it to idle node this is termed as load forecasting. In this paper, we did simulation of grid CPU load in distributed manner which provide the monitoring on each host in network and load of grid is predicted using bpn (back propagation neural network) algorithm. Which provides the effective results in prediction, in addition of that a new predictive algorithm is proposed implemented and compared to the bpn algorithm. In proposed algorithm we calculate the accuracy and compared with bpn algorithm accuracy, after implementation of both methods and simulation of Host load we find the proposed method is much effective then the previously proposed method of BPN algorithm.