ABSTRACT
Minimising energy use is an important factor in the operation of many classes of embedded systems - in particular, battery-powered devices. Dynamic voltage and frequency scaling (DVFS) provides some control over a processor's performance and energy consumption. In order to employ DVFS for managing a system's energy use, it is necessary to predict the effect this scaling has on the system's total energy consumption. Simple (yet widely-used) energy models lead to dramatically incorrect results for important classes of application programs.
Predicting the energy used under scaling requires (i) a prediction of the dependency of program performance (and hence duration of execution) on the frequencies and (ii) a prediction of the power drawn by the execution as a function of the frequencies and voltages.
As both of these characteristics are workload-specific our approach builds a model that, given a workload execution at one frequency setpoint, will predict the run-time and power at any other frequency setpoint. We assume temporal locality (which is valid for the vast majority of applications) so predicting the characteristics of one time slice, frame, or other instance of a task, will imply the characteristics of subsequent time slices, frames or instances (e.g. MPEG video decoding).
We present a systematic approach to building these models for a hardware platform, determining the best performance counters and weights. This characterisation, done once for a particular platform, produces platform-specific but workload-independent performance and power models.
We implemented the model on a real system and evaluated it under a comprehensive benchmark suite against measurements of the actual energy consumption. The results show that the model can accurately predict the energy use of a wide class of applications and is highly responsive to changes in the application behaviour.
- A. Acquaviva, L. Benini, and B. Ricco. Software-controlled processor speed setting for low-power streaming multimedia. IEEE Transactions on CAD ICAS, 20(11):1283--1292, Nov. 2001. Google ScholarDigital Library
- F. Bellosa. The benefits of event-driven energy accounting in power-sensitive systems. In Proceedings of the 9th SIGOPS European Workshop, Kolding, Denmark, Sept. 17-20 2000. Google ScholarDigital Library
- L. Bircher, M. Valluri, J. Law, and L. John. Runtime identification of microprocessor energy saving opportunities. In Proceedings of the 2005 International Symposium on Low Power Electronics and Design, 2005. Google ScholarDigital Library
- G. Contreras and M. Martonosi. Power prediction for Intel XScale processors using performance monitoring unit events. In Proceedings of the 2005 International Symposium on Low Power Electronics and Design, San Diego, CA, USA, Aug. 2005. Google ScholarDigital Library
- E. Duesterwald, C. Cascaval, and S. Dwarkadas. Characterizing and predicting program behavior and its variability. In Proceedings of the 12th International Conferenece on Parallel Architectures and Compilation Techniques, Sept. 2003. Google ScholarDigital Library
- J. Flinn and M. Satyanarayanan. Powerscope: A tool for profiling the energy usage of mobile applications. In Proceedings of the Second IEEE, Workshop on Mobile Computing Systems and Applications, 1999. Google ScholarDigital Library
- F. Fruth. Run-time energy characterization of the Intel PXA. Study thesis, Operating System Group, University of Erlangen, Germany, Apr. 2005.Google Scholar
- D. Grunwald, P. Levis, K. I. Farkas, C. B. Morrey III, and M. Neufeld. Policies for dynamic clock scheduling. In Proceedings of the 4th USENIX Symposium on Operating Systems Design and Implementation, pages 73--86, San Diego, CA, USA, Oct. 2000. Google ScholarDigital Library
- M. R. Guthaus, J. S. Reingenberg, D. Ernst, T. M. Austing, T. Mudge, and R. B. Brown. Mibench: A free, commercially representative embedded benchmark suite. In Proceedings of the 4th IEEE Annual Workshop on Workload Characterization, Dec. 2001. Google ScholarDigital Library
- Intel Corporation. Intel PXA250 and PXA210 applications processors developers manual. http://www.intel.com/design/pca/products/pxa255/techdocs.htm, 2005.Google Scholar
- T. L. Martin and D. P. Siewiorek. Nonideal battery and main memory effects on cpu speed-setting for low power. IEEE Transactions on Very Large Scale Integration Systems, 9(1):29--34, Feb. 2001. Google ScholarDigital Library
- Microchip Technology Incorporated. MCP3909: Energy metering IC with SPI interface and active power pulse output. http://www.microchip.com/MCP3909, 2006.Google Scholar
- A. Miyoshi, C. Lefurgy, E. V. Hensbergen, R. Rajamony, and R. Rajkumar. Critical power slope: understanding theruntime effects of frequency scaling. In Proceedings of the 16th International Conference on Supercomputing, pages 35--44, New York, NY, USA, 2002. ACM Press. Google ScholarDigital Library
- R. Nathuji, B. Seshasayee, and K. Schwan. Combining compiler and operating system support for energy efficient i/o on embedded platforms. In Proceedings of the ACM Workshop on Software and Compilers for Embedded Systems, Dallas, TX, USA, 2005. ACM Press. Google ScholarDigital Library
- R. Neugebauer and D. McAuley. Energy is just another resource: Energy accounting and energy pricing in the nemesis os. In Proceedings of the 8th Workshop on Hot Topics in Operating Systems, Schloss Elmau, Germany, 2001. Google ScholarDigital Library
- J. Peddersen and S. Parameswaran. CLIPPER: Counter-based low impact processor power estimation at runtime. In Proceedings of the 12th Asia and South Pacific Design Automation Conference, Yokohama, Japan, Jan. 2007. Google ScholarDigital Library
- Seiko Epson Corporation. S1F81100 technical manual. http://www.epsondevice.com, 2003.Google Scholar
- K. Seth, A. Anantaraman, F. Mueller, and E. Rotenberg. FAST: Frequency-aware static timing analysis. ACM Transactions on Embedded Computing Systems, 5(1):200--224, 2006. Google ScholarDigital Library
- D. C. Snowdon. PLEB 2 web site. http://www.ertos.nicta.com.au/hardware/pleb, 2006.Google Scholar
- D. C. Snowdon, S. M. Petters, and G. Heiser. Power measurement as the basis for power management. In Proceedings of the 2005 Workshop on Operating System Platforms for Embedded Real-Time Applications, Palma, Mallorca, Spain, July 2005.Google Scholar
- D. C. Snowdon, S. Ruocco, and G. Heiser. Power Management and Dynamic Voltage Scaling: Myths and Facts. In Proceedings of the 2005 Workshop on Power Aware Real-time Computing, New Jersey, USA, Sept. 2005. Unpublished proceedings, available from http://ertos.nicta.com.au/publications.Google Scholar
- D. C. Snowdon, G. van der Linden, S. M. Petters, and G. Heiser. Accurate run-time prediction of performance degradation under frequency scaling. In Proceedings of the 2007 Workshop on Operating System Platforms for Embedded Real-Time Applications, Pisa, Italy, July 2007.Google Scholar
- M. Waitz. Accounting and control of power consumption in energy-aware operating systems. Diploma thesis, Operating System Group, University of Erlangen, Germany, Jan. 2003.Google Scholar
- M. Weiser, B. Welch, A. J. Demers, and S. Shenker. Scheduling for reduced CPU energy. In Proceedings of the 1st USENIX Symposium on Operating Systems Design and Implementation, pages 13--23, 1994. Google ScholarDigital Library
- A. Weissel and F. Bellosa. Process cruise control - event-driven clock scaling for dynamic power management. In Proceedings of the International Conference on Compilers, Architecture and Synthesis for Embedded Systems, Grenoble, France, Oct. 8--11 2002. Google ScholarDigital Library
- A. Wei&3223;el, B. Beutel, and F. Bellosa. Cooperative IO - a novel IO semantics for energy-aware applications. In Proceedings of the 5th USENIX Symposium on Operating Systems Design and Implementation, Boston, MA, USA, Dec. 2002. Google ScholarDigital Library
- H. Zeng, C. S. Ellis, A. R. Lebeck, and A. Vahdat. Currentcy: Unifying policies for resource management. In Proceedings of the 2003 Annual USENIX Technical Conference, San Antonio, Texas, June 2003. Google ScholarDigital Library
Index Terms
- Accurate on-line prediction of processor and memoryenergy usage under voltage scaling
Recommendations
Dynamic Voltage Scaling Scheduling on Power-Aware Clusters under Power Constraints
DS-RT '13: Proceedings of the 2013 IEEE/ACM 17th International Symposium on Distributed Simulation and Real Time ApplicationsEnergy consumption of computing systems and especially of large-scale systems has a growing concern nowadays due to economic and ecological reasons. To reduce energy consumption, a variety of mechanisms are proposed in the literature. One of them is ...
Workload-Aware CPU Performance Scaling for Transactional Database Systems
SIGMOD '18: Proceedings of the 2018 International Conference on Management of DataNatural short term fluctuations in the load of transactional data systems present an opportunity for power savings. For example, a system handling 1000 requests per second on average can expect more than 1000 requests in some seconds, fewer in others. ...
Dynamic voltage scaling for systemwide energy minimization in real-time embedded systems
ISLPED '04: Proceedings of the 2004 international symposium on Low power electronics and designTraditionally, dynamic voltage scaling (DVS) techniques have focused on minimizing the processor energy consumption as opposed to the entire system energy consumption. The slowdown resulting from DVS can increase the energy consumption of components ...
Comments