A static power model for architects
Proceedings of the 33rd annual ACM/IEEE international symposium on Microarchitecture
Machine Learning
Profile-based dynamic voltage and frequency scaling for a multiple clock domain microprocessor
Proceedings of the 30th annual international symposium on Computer architecture
Exploring the Energy-Time Tradeoff in MPI Programs on a Power-Scalable Cluster
IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Papers - Volume 01
Analysis of dynamic voltage/frequency scaling in chip-multiprocessors
ISLPED '07 Proceedings of the 2007 international symposium on Low power electronics and design
Dynamic voltage frequency scaling for multi-tasking systems using online learning
ISLPED '07 Proceedings of the 2007 international symposium on Low power electronics and design
Seeking information in realistic books: a user study
Proceedings of the 8th ACM/IEEE-CS joint conference on Digital libraries
Variation-Aware Application Scheduling and Power Management for Chip Multiprocessors
ISCA '08 Proceedings of the 35th Annual International Symposium on Computer Architecture
The PARSEC benchmark suite: characterization and architectural implications
Proceedings of the 17th international conference on Parallel architectures and compilation techniques
Integrated CPU cache power management in multiple clock domain processors
HiPEAC'08 Proceedings of the 3rd international conference on High performance embedded architectures and compilers
Compiler control power saving scheme for multi core processors
LCPC'05 Proceedings of the 18th international conference on Languages and Compilers for Parallel Computing
Identifying the optimal energy-efficient operating points of parallel workloads
Proceedings of the International Conference on Computer-Aided Design
Predicting Performance Impact of DVFS for Realistic Memory Systems
MICRO-45 Proceedings of the 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture
Hi-index | 0.00 |
Dynamic Voltage/Frequency Scaling (DVFS) is a useful tool for improving system energy efficiency, especially in multi-core chips where energy is more of a limiting factor. Per-core DVFS, where cores can independently scale their voltages and frequencies, is particularly effective. We present a DVFS policy using machine learning, which learns the best frequency choices for a machine as a decision tree. Machine learning is used to predict the frequency which will minimize the expected energy per user-instruction (epui) or energy per (user-instruction)2 (epui2). While each core independently sets its frequency and voltage, a core is sensitive to other cores' frequency settings. Also, we examine the viability of using only partial training to train our policy, rather than full profiling for each program. We evaluate our policy on a 16-core machine running multiprogrammed, multithreaded benchmarks from the PARSEC benchmark suite against a baseline fixed frequency as well as a recently-proposed greedy policy. For 1ms DVFS intervals, our technique improves system epui2 by 14.4% over the baseline no-DVFS policy and 11.3% on average over the greedy policy.