Managing and forecasting hardware resource consumption

Computer systems managers make decisions about hardware/software selection, performance evaluation, capacity planning, and other resource variables on the basis of actual data, accounting data, judgment, and assumptions about the resource consumption of the jobs being run. This paper presents a method for forecasting hardware resource consumption to assist the manager in making hardware policy decisions by giving early warning of possible bottlenecks; forecasting growth of hardware usage; etc. This method is based on a model for analyzing and evaluating computer resource consumption developed in a previous paper. This paper proposes a method that can be applied to the management of hardware resources, particularly addressing the issues of forecasting and measuring the demand sensitivity of hardware resources.