CPU Consolidation versus Dynamic Voltage and Frequency Scaling in a Virtualized Multi-Core Server : Which is More Effective and When

— Companies operating large data centers are focusing on how to reduce the electrical energy costs of operating data centers 24-7. A common way of reducing this cost is to perform dynamic voltage and frequency scaling (DVFS), thereby matching the CPU's performance and power level to the incoming workload. Another power saving technique is CPU consolidation, which uses the minimum number of CPUs necessary to meet the service request demands and turns off the remaining idle CPUs. A key question that must be answered is which of these two techniques is more effective and under what conditions. This is the question that is addressed in this report. After analyzing the power consumption in a modern server system and developing appropriate power and performance models for the same, this report provides an extensive set of hardware-based experimental results and makes suggestions about how to maximize energy savings through CPU consolidation and/or DVFS. In addition, the report also presents new online CPU consolidation algorithms. The proposed algorithms reduce the energy delay product more than the Linux default DVFS algorithm (up to 13 %).