Performance Evaluation for Dynamic Voltage and Frequency Scaling Using Runtime Performance Counters

Abstract:

Dynamic voltage and frequency scaling (DVFS) is an effective technique for reducing power consumption. The system performance is not easy to evaluate through Dynamic Voltage and Frequency Scaling. Most of studies use the execution time as an indicator while measuring the performance. However, DVFS adjusted processor speed during a fixed-length period so it cannot rely on the execution time to evaluate the system performance. This study proposes a novel and simple performance evaluation method to evaluate the system performance when DVFS is activated. Based on the performance evaluation method, this study also proposes a DVFS algorithm (P-DVFS) for a general-purpose operating system. The algorithm has been implemented on the Linux operating system and used a PXA270 development board. The results show that P-DVFS could accurately predict the suitable frequency, given runtime statistics information of a running program. In this way, the user can easily control the energy consumption by specifying allowable performance loss factor.

• Periodical: Applied Mechanics and Materials (Volumes 284-287)
• Pages: 2575-2579
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.284-287.2575
• Online since: January 2013
• Authors: Wen Yew Liang, Ming Feng Chang, Yen Lin Chen, Jenq Haur Wang
• Keywords: DVFS, Linux, Low Power Software Design, Performance Evaluation, Pxa270

References

[1] K. Choi, R. Soma, and M. Pedram: Fine-Grained Dynamic Voltage and Frequency Scaling for Precise Energy and Performance Trade-Off Based on the Ratio of Off-Chip Access to On-Chip Computation Times. Proceedings of the Design, Automation and Test in Europe, Vol. 1 (2004).

DOI: 10.1109/date.2004.1268819

[2] W. -Y. Liang, S. -C. Chen, Y. -L. Chang, J. -P. Fang: Memory-Aware Dynamic Voltage and Frequency Prediction for Portable Devices. Proceedings of the 14th IEEE Int. Conf. Embedded and Real-Time Computing Systems and Applications, (2008).

DOI: 10.1109/rtcsa.2008.19

[3] C. Poellabauer, L. Singleton, and K. Schwan: Feedback-Based Dynamic Voltage and Frequency Scaling for Memory-Bound Real-Time Applications. Proceedings of the IEEE Real-Time and Embedded Technology and Applications Symposium, Mar. (2005).

DOI: 10.1109/rtas.2005.23

[4] S. Eyerman, L. Eeckhout. A Counter Architecture for Online DVFS Profitability Estimation. Proceedings of the 2010 IEEE Transactions on Computers, Nov. (2010), pp.1576-1583.

DOI: 10.1109/tc.2010.65

[5] C. Poellabauer, K. Schwan: Energy-aware traffic shaping for wireless real-time applications. Proceedings of RTAS' 04, May (2004), p.48–55.

DOI: 10.1109/rttas.2004.1317248

[6] B. Mochocki, D. Rajan, X. Sharon Hu, C. Poellabauer, K. Otten, T. Chantem: Network-aware dynamic voltage and frequency scaling. Proceedings of the 13th IEEE Real-Time and Embedded Technology and Applications Symposium, (2007).

DOI: 10.1109/rtas.2007.20

[7] M. Anand, E.B. Nightingale, J. Flinn: Self-tuning network power management. Proceedings Ninth Annual International Conference on Mobile Computing and Networking, September (2003).

DOI: 10.1145/938985.939004

[8] N. Abou Ghazaleh, R.N. Mayo, P. Ranganathan: Idle Time Power Management for Personal Wireless Devices, Technical Report, HP Laboratories, September (2003).

[9] X. Liu, P. Shenoy, and W. Gong: A time series-based approach for power management in mobile processors and disks. Proceedings of the 14th international workshop on Network and operating systems support for digital audio and video, (2004).

DOI: 10.1145/1005847.1005864

[10] A. Weissel and F. Bellosa: Self-learning hard disk power management for mobile devices. In Proceedings of the Second International Workshop on Software Support for Portable Storage, Oct. (2006), pp.33-40.

[11] V. Delaluz, A. Sivasubramaniam, M. Kandemir, N. Vijaykrishran, M.J. Irwin: Scheduler-based DRAM energy management. Proceedings of DAC 2002, June (2002), pp.697-702.

DOI: 10.1145/513918.514095

[12] D. Rajan, R. Zuck, and C. Poellabauer: A Dual Speed Approach to Workload-Aware Voltage Scaling. Technical Report TR-2006-05, University of Notre Dame. (2006).

DOI: 10.1109/rtcsa.2006.64

[13] D. Rajan, R. Zuck, and C. Poellabauer: Workload-Aware Dual-Speed Dynamic Voltage Scaling. Proceedings. 12th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, (2006), pp.251-256.

DOI: 10.1109/rtcsa.2006.64

[14] V. Pallipadi and A. Starikovskiy: The on-demand governor-past, present, and future. In Proc. Linux Symp., (2006), p.223.

[15] Intel XScale® Technology, available on http: /www. intel. com/design/intelxscale.