Maximizing Performance Under a Power Cap

Abstract

Power and thermal dissipation constrain multicore performance scaling. Modern processors are built such that they could sustain damaging levels of power dissipation, creating a need for systems that can implement processor power caps. A particular challenge is developing systems that can maximize performance within a power cap, and approaches have been proposed in both software and hardware. Software approaches are flexible, allowing multiple hardware resources to be coordinated for maximum performance, but software is slow, requiring a long time to converge to the power target. In contrast, hardware power capping quickly converges to the the power cap, but only manages voltage and frequency, limiting its potential performance. In this work we propose PUPiL, a hybrid software/hardware power capping system. Unlike previous approaches, PUPiL combines hardware's fast reaction time with software's flexibility. We implement PUPiL on real Linux/x86 platform and compare it to Intel's commercial hardware power capping system for both single and multi-application workloads. We find PUPiL provides the same reaction time as Intel's hardware with significantly higher performance. On average, PUPiL outperforms hardware by from 1:18-2:4 depending on workload and power target. Thus, PUPiL provides a promising way to enforce power caps with greater performance than current state-of-the-art hardware-only approaches.

Deep Software Variability: Towards Handling Cross-Layer Configuration

“…Numerous research papers study the effects of variability layers on software's performances or on configuration options: hardware [10,35,43], workloads [19,20,24,30,43], variants [22,37], versions [13,29,37,41], compilation options [15,27] and input data [3,8]. Such studies provide evidence that some layers have a noticeable impact on the software (configuration) layer.…”

“…Various research papers [15,19,20,24,27,43] configure their variability layer to optimize non-functional properties of software, while others, as [8,30], adapt configuration options of software to variability layers. Such works are in line with the vision exposed in Section 3.5 though only one layer at a time is usually considered as part of the specialization process.…”