Google Research

User inactivity modeling and simulation for power management of PC fleets

Proceedings of the International Symposium on Performance Evaluation of Computer and Telecommunication Systems, Society for Computer Simulation International, San Diego, CA, USA (2015), pp. 1-6

Abstract

Energy consumption of the computer fleet can be one of the main costs for corporations in maintaining local sites or offices. In order to adjust the power management policies used to put computers in a low-consumption state, the utilization pattern of the users must be known. In particular, the inactivity periods of the users must be characterized. This paper presents a method to extract the parameters that make it possible to model the inactivity of in order to simulate the behavior of the whole fleet. A simulation tool developed in Python with SimPy is also presented. The results of the simulation tool can be used as input in an informed decision about what power management policies to apply to the computer fleet.

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work