HaPPy: hyperthread-aware power profiling dynamically

Abstract

Quantifying the power consumption of individual appli-cations co-running on a single server is a critical compo-nent for software-based power capping, scheduling, and provisioning techniques in modern datacenters. How-ever, with the proliferation of hyperthreading in the last few generations of server-grade processor designs, the challenge of accurately and dynamically performing this power attribution to individual threads has been signifi-cantly exacerbated. Due to the sharing of core-level re-sources such as functional units, prior techniques are not suitable to attribute the power consumption between hy-perthreads sharing a physical core. In this paper, we present a runtime mechanism that quantifies and attributes power consumption to individ-ual jobs at fine granularity. Specifically, we introduce a hyperthread-aware power model that differentiates be-tween the states when both hardware threads of a core are in use, and when only one thread is in use. By capturing these two different states, we are able to accurately at-tribute power to each logical CPU in modern servers. We conducted experiments with several Google production workloads on an Intel Sandy Bridge server. Compared to prior hyperthread-oblivious model, HaPPy is substan-tially more accurate, reducing the prediction error from 20.5 % to 7.5 % on average and from 31.5 % to 9.4 % in the worst case. 1