Practical performance prediction under Dynamic Voltage Frequency Scaling
Authors:
- 1 July 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Abstract
Predicting performance under Dynamic Voltage Frequency Scaling (DVFS) remains an open problem. Current best practice explores available performance counters to serve as input to linear regression models that predict performance. However, the inaccuracies of these models require that large-scale DVFS runtime algorithms predict performance conservatively in order to avoid significant consequences of mispredictions. Recent theoretical work based on interval analysis advocates a more accurate and reliable solution based on a single new performance counter, Leading Loads. In this paper, we evaluate a processor-independent analytic framework for existing performance counters based on this interval analysis model. We begin with an analysis of the counters used in many published models. We then briefly describe the Leading Loads architectural model and describe how we can use Leading Loads Cycles to predict performance under DVFS. We validate this approach for the NAS Parallel Benchmarks and SPEC CPU 2006 benchmarks, demonstrating an order of magnitude improvement in both error and standard deviation compared to the best existing approaches.Keywords
This publication has 25 references indexed in Scilit:
- Entering the petaflop era: The architecture and performance of Roadrunner2008
- Prediction models for multi-dimensional power-performance optimization on many cores2008
- Just-in-time dynamic voltage scaling: Exploiting inter-node slack to save energy in MPI programsJournal of Parallel and Distributed Computing, 2008
- Bounding energy consumption in large-scale MPI programs2007
- CPU MISER: A Performance-Directed, Run-Time System for Power-Aware Clusters2007 International Conference on Parallel Processing (ICPP 2007), 2007
- Accurate on-line prediction of processor and memoryenergy usage under voltage scaling2007
- Power-Aware Speedup2007
- Online power-performance adaptation of multithreaded programs using hardware event-based prediction2006
- Exploring the Energy-Time Tradeoff in MPI Programs on a Power-Scalable Cluster2005
- Overview of the Blue Gene/L system architectureIBM Journal of Research and Development, 2005