Predicting Cpu Availability on the Computational Grid Using the Network Weather Service

Abstract

In this paper, we focus on the problem of predicting CPU availability for Computational Grid settings in which individual machines may be either time-shared or batch-controlled. We use the Network Weather Service -a distributed system that monitors and forecasts resource performance in Computational Grid environments -to measure and predict CPU availability. We examine the accuracy with which CPU availability can be predicted in an interactive cluster computing environment under production load conditions, and compare these results with a similar study of a production batch system. Our work shows that in the environments we have studied, the availability of clustered interactive resources is significantly more predictable that that of the batch system.

NWSLite

“…For example, if an exponential smoothing predictor is the most accurate predictor at one point in time and conditions change so that another predictor becomes the most accurate, the system will choose the latter predictor as winner, only if the change is persistent enough to cause the aggregated error of the latter to be smaller than the former one. A detailed discussion of NWS forecaster selection is given in [Wolski 1999[Wolski , 2003Wolski et al 1999].…”