Discovering Linear Models of Grid Workload

Despite extensive research focused on enabling QoS for grid users through economic and intelligent resource provisioning, no consensus has emerged on the most promising strategies. On top of intrinsically challenging problems, the complexity and size of data has so far drastically limited the number of comparative experiments. An alternative to experimenting on real, large, and complex data, is to look for well-founded and parsimonious representations. The goal of this paper is to answer a set of preliminary questions, which may help steering the design of those along feasible paths: is it possible to exhibit consistent models of the grid workload? If such models do exist, which classes of models are more appropriate, considering both simplicity and descriptive power? How can we actually discover such models? And finally, how can we assess the quality of these models on a statistically rigorous basis? Our main contributions are twofold. First we found that grid workload models can consistently be discovered from the real data, and that limiting the range of models to piecewise linear time series models is sufficiently powerful. Second, we presents a bootstrapping strategy for building more robust models from the limited samples at hand. This study is based on exhaustive information representative of a significant fraction of e-science computing activity in Europe