Run-time Energy Consumption Estimation Based on Workload in Server Systems

This paper proposes to develop a system-wide energy consumption model for servers by making use of hardware performance counters and experimental measurements. We develop a real-time energy prediction model that relates server energy consumption to its overall thermal envelope. While previous studies have attempted system-wide modeling of server power consumption through subsystem models, our approach is different in that it uses a small set of tightly correlated parameters to create a model relating system energy input to subsystem energy consumption. We develop a linear regression model that relates processor power, bus activity, and system ambient temperatures into real-time predictions of the power consumption of long jobs and as result controlling their thermal impact. Using the HyperTransport bus model as a case study and through electrical measurements on example server subsystems, we develop a statistical model for estimating run-time power consumption. Our model is accurate w...

• Type: Conference
• Year: 2008
• Where: OSDI
• Authors: Adam Lewis, Soumik Ghosh, Nian-Feng Tzeng
• Keywords: Energy Consumption Model, Operating System, OSDI 2008, Server Energy Consumption, Subsystem Energy Consumption