Network Packet Processing Mode-Aware Power Management for Data Center Servers

In data center servers, power management (PM) exploiting Dynamic Voltage and Frequency Scaling (DVFS) for processors can play a crucial role to improve energy efficiency. However, we observe that current PM policies (i.e., governors) not only considerably increase tail response time (i.e., violate a given Service Level Objective (SLO)) but also hurt energy efficiency. Tackling limitations of current PM governors, we propose <monospace>NMAP</monospace>, <bold>N</bold>etwork packet processing <bold>M</bold>ode-<bold>A</bold>ware <bold>P</bold>ower management. <monospace>NMAP</monospace> improves energy efficiency while satisfying given SLOs, considering packet processing status on a core for PM by monitoring transitions between network packet processing modes – interrupt and polling. Tracking the transitions, <monospace>NMAP</monospace> detects moments that a core cannot process packets fast enough and forces the core to immediately raise the voltage and frequency (V/F) state. As a result, <monospace>NMAP</monospace> can provide not only low response time but also low energy consumption. Our experiment shows that <monospace>NMAP</monospace> improves tail response time by up to 4.1× compared with the <monospace>ondemand</monospace> governor, reducing energy by up to 44.6 percent compared with the <monospace>performance</monospace> governor.