Abstract | ||
---|---|---|
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 within an error of four percent(4%) as verified using a set of common processor benchmarks. |
Year | Venue | Keywords |
---|---|---|
2008 | HotPower | subsystem model,hypertransport bus model,real-time energy prediction model,energy consumption,statistical model,run-time energy consumption estimation,linear regression model,server energy consumption,power consumption,server system,run-time power consumption,system-wide energy consumption model,prediction model,system modeling,real time,ambient temperature |
Field | DocType | Citations |
Workload,Simulation,Computer science,Server,Parallel computing,Electrical measurements,Real-time computing,Statistical model,Small set,Energy consumption,HyperTransport,Linear regression | Conference | 43 |
PageRank | References | Authors |
1.93 | 7 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Adam Lewis | 1 | 67 | 6.19 |
Soumik Ghosh | 2 | 63 | 5.17 |
N.-F. Tzeng | 3 | 173 | 16.95 |