Description

The ability to accurately predict future resource capabilities is of great importance for applications in hierarchical grid environment. Our contribution is to predict workload of grid performance in a hierarchical grid environment. The hierarchical grid environment contains set of resources and machines in each resource. The resources provided are in a hierarchical format using some priority. The workload is submitted to the resources based on the hierarchy. The purpose is to smooth the prediction process from being disturbed by load fluctuations. We use ensemble kalman filter to reduce prediction errors. Some of the parameters used for job creation and resource creation are used to reduce the prediction errors using ensemble kalman filter. Confidence window is generated by DA2LPInterval to get the load index range for future workload. Finally Load prediction is done by Autoregressive moving Average. ARMA models can also be used to predict behavior of a time series from past values alone. Such a prediction can be used as a baseline to evaluate possible importance of other variables to the system.