LEAD: Learning-enabled Energy-Aware Dynamic Voltage/frequency scaling in NoCs

Network on Chips (NoCs) are the interconnect fabric of choice for multicore processors due to their superiority over traditional buses and crossbars in terms of scalability. While NoC’s offer several advantages, they still suffer from high static and dynamic power consumption. Dynamic Voltage and Frequency Scaling (DVFS) is a popular technique that allows dynamic energy to be saved, but it can potentially lead to loss in throughput. In this paper, we propose LEAD - Learning-enabled Energy-Aware Dynamic voltage/frequency scaling for NoC architectures wherein we use machine learning techniques to enable energy-performance trade-offs at reduced overhead cost. LEAD enables a proactive energy management strategy that relies on an offline trained regression model and provides a wide variety of voltage/frequency pairs (modes). LEAD groups each router and the router’s outgoing links locally into the same V/F domain, allowing energy management at a finer granularity without additional timing complications and overhead. Our simulation results using PARSEC and Splash-2 benchmarks on a 4 × 4 concentrated mesh architecture show an average dynamic energy savings of 17% with a minimal loss of 4% in throughput and no latency increase.

[1]  Richard G. Hay.  Machine Learning Based DVFS for Energy Efficient Execution of Multithreaded Workloads , 2014 .

[2]  Preeti Ranjan Panda,et al.  Machine Learned Machines: Adaptive co-optimization of caches, cores, and On-chip Network , 2016, 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[3]  Geoff V. Merrett,et al.  Adaptive energy minimization of embedded heterogeneous systems using regression-based learning , 2015, 2015 25th International Workshop on Power and Timing Modeling, Optimization and Simulation (PATMOS).

[4]  Massoud Pedram,et al.  Power punch: Towards non-blocking power-gating of NoC routers , 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

[5]  Diana Marculescu,et al.  Analysis of dynamic voltage/frequency scaling in chip-multiprocessors , 2007, Proceedings of the 2007 international symposium on Low power electronics and design (ISLPED '07).

[6]  Radu Marculescu,et al.  Dynamic power management for multicores: Case study using the intel SCC , 2012, 2012 IEEE/IFIP 20th International Conference on VLSI and System-on-Chip (VLSI-SoC).

[7]  Chita R. Das,et al.  A case for dynamic frequency tuning in on-chip networks , 2009, 2009 42nd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

[8]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[9]  Stijn Eyerman,et al.  Fine-grained DVFS using on-chip regulators , 2011, TACO.

[10]  José González,et al.  Independent Front-end and Back-end Dynamic Voltage Scaling for a GALS Microarchitecture , 2006, ISLPED'06 Proceedings of the 2006 International Symposium on Low Power Electronics and Design.

[11]  Geoff V. Merrett,et al.  Adaptive Energy Minimization of OpenMP Parallel Applications on Many-Core Systems , 2015, PARMA-DITAM '15.

[12]  Mark Horowitz,et al.  1.1 Computing's energy problem (and what we can do about it) , 2014, 2014 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC).

[13]  Radu Marculescu,et al.  An Optimal Control Approach to Power Management for Multi-Voltage and Frequency Islands Multiprocessor Platforms under Highly Variable Workloads , 2012, 2012 IEEE/ACM Sixth International Symposium on Networks-on-Chip.

[14]  Tajana Simunic,et al.  Dynamic voltage frequency scaling for multi-tasking systems using online learning , 2007, Proceedings of the 2007 international symposium on Low power electronics and design (ISLPED '07).

[15]  Christian Bienia,et al.  PARSEC 2.0: A New Benchmark Suite for Chip-Multiprocessors , 2009 .

[16]  Anoop Gupta,et al.  The SPLASH-2 programs: characterization and methodological considerations , 1995, ISCA.

[17]  Xi Chen,et al.  Dynamic voltage and frequency scaling for shared resources in multicore processor designs , 2013, 2013 50th ACM/EDAC/IEEE Design Automation Conference (DAC).

[18]  Li Shang,et al.  Power-efficient Interconnection Networks: Dynamic Voltage Scaling with Links , 2002, IEEE Computer Architecture Letters.

[19]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[20]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Hao Shen,et al.  Learning based DVFS for simultaneous temperature, performance and energy management , 2012, Thirteenth International Symposium on Quality Electronic Design (ISQED).

[22]  Natalie D. Enright Jerger,et al.  Improving DVFS in NoCs with Coherence Prediction , 2015, NOCS.

[23]  Chen Sun,et al.  DSENT - A Tool Connecting Emerging Photonics with Electronics for Opto-Electronic Networks-on-Chip Modeling , 2012, 2012 IEEE/ACM Sixth International Symposium on Networks-on-Chip.

[24]  David R. Kaeli,et al.  Multi2Sim: A simulation framework for CPU-GPU computing , 2012, 2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT).