Abstract
In an FPGA system-on-chip design, it is often insufficient to merely assess the power consumption of the entire circuit by compile-time estimation or runtime power measurement. Instead, to make better decisions, one must understand the power consumed by each module in the system. In this work, we combine measurements of register-level switching activity and system-level power to build an adaptive online model that produces live breakdowns of power consumption within the design. Online model refinement avoids time-consuming characterization while also allowing the model to track long-term operating condition changes. Central to our method is an automated flow that selects signals predicted to be indicative of high power consumption, instrumenting them for monitoring. We named this technique KAPow, for ‘K’ounting Activity for Power estimation, which we show to be accurate and to have low overheads across a range of representative benchmarks. We also propose a strategy allowing for the identification and subsequent elimination of counters found to be of low significance at runtime, reducing algorithmic complexity without sacrificing significant accuracy. Finally, we demonstrate an application example in which a module-level power breakdown can be used to determine an efficient mapping of tasks to modules and reduce system-wide power consumption by up to 7%.
- Altera. 2015. Cyclone V SoC Development Board -- Reference Manual. Retrieved from https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/manual/rm_cv_soc_dev_board.pdf.Google Scholar
- Altera. 2016. Stratix: High-Performance ALM and Interconnect. Retrieved from https://www.altera.com/products/fpga/features/stx-architecture.html.Google Scholar
- J. H. Anderson and F. N. Najm. 2004. Power estimation techniques for FPGAs. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 12, 10 (2004), 1015--1027.Google Scholar
Digital Library
- H. Esmaeilzadeh, E. Blem, R. S. Amant, K. Sankaralingam, and D. Burger. 2011. Dark silicon and the end of multicore scaling. In Proceedings of the International Symposium on Computer Architecture (ISCA). 365--376. Google Scholar
Digital Library
- C. F. Gauss. 1821. Theoria Combinationis Observationum Erroribus Minimis Obnoxiae. H. Dieterich, 1--71.Google Scholar
- E. Hung, J. J. Davis, J. M. Levine, E. A. Stott, P. Y. K. Cheung, and G. A. Constantinides. 2016. KAPow: A system identification approach to online per-module power estimation in FPGA designs. In Proceedings of the International Symposium on Field-Programmable Custom Computing Machines (FCCM). 56--63. Google Scholar
Cross Ref
- E. Hung, T. Todman, and W. Luk. 2014. Transparent insertion of latency-oblivious logic onto FPGAs. In Proceedings of the International Conference on Field-Programmable Logic and Applications (FPL). 1--8. Google Scholar
Cross Ref
- E. Hung and S. J. E. Wilton. 2013. Scalable signal selection for post-silicon debug. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 21, 6 (2013), 1103--1115.Google Scholar
Digital Library
- E. Hung and S. J. E. Wilton. 2015. Zero-overhead FPGA debugging. Reconfigurable Logic: Architecture, Tools, and Applications 48 (2015), 71--96.Google Scholar
- A. Lakshminarayana, S. Ahuja, and S. Shukla. 2011. High-level power estimation models for FPGAs. In Proceedings of the IEEE Computer Society Annual Symposium on VLSI (ISVLSI). 7--12. Google Scholar
Digital Library
- J. Lamoureux and S. J. E. Wilton. 2006. Activity estimation for field-programmable gate arrays. In Proceedings of the International Conference on Field-Programmable Logic and Applications (FPL). 1--8. Google Scholar
Cross Ref
- D. Lee, T. Kim, K. Han, Y. Hoskote, L. K. John, and A. Gerstlauer. 2015. Learning-based power modeling of system-level black-box IPs. In Proceedings of the International Conference on Computer-Aided Design (ICCAD). 847--853. Google Scholar
Digital Library
- J. M. Levine, E. Stott, G. A. Constantinides, and P. Y. K. Cheung. 2012. Online measurement of timing in circuits: For health monitoring and dynamic voltage 8 frequency scaling. In Proceedings of the IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM). 109--116. Google Scholar
Digital Library
- Linear Technology. 2009. LTC2978: Octal Digital Power Supply Manager with EEPROM. Retrieved from http://cds.linear.com/docs/en/datasheet/2978fd.pdf.Google Scholar
- L. Ljung. 1998. System Identification. Birkhäuser, 163--173.Google Scholar
- K. E. Murray, S. Whitty, S. Liu, J. Luu, and V. Betz. 2013. Titan: Enabling large and complex benchmarks in academic CAD. In Proceedings of the International Conference on Field-Programmable Logic and Applications (FPL). 1--8. Google Scholar
Cross Ref
- M. Najem, P. Benoit, F. Bruguier, G. Sassatelli, and L. Torres. 2014. Method for dynamic power monitoring on FPGAs. In Proceedings of the International Conference on Field-Programmable Logic and Applications (FPL). 1--6. Google Scholar
Cross Ref
- F. N. Najm. 1994. A survey of power estimation techniques in VLSI circuits. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 2, 4 (1994), 446--455.Google Scholar
Digital Library
- R. L. Plackett. 1950. Some theorems in least squares. Biometrika 37 (1950), 149--157. Google Scholar
Cross Ref
- E. Stott, Z. Guan, J. M. Levine, J. S. J. Wong, and P. Y. K. Cheung. 2013. Variation and reliability in FPGAs. IEEE Design 8 Test 30, 6 (2013), 50--59.Google Scholar
Cross Ref
- E. Stott, J. S. J. Wong, and P. Y. K. Cheung. 2010. Degradation analysis and mitigation in FPGAs. In Proceedings of the International Conference on Field-Programmable Logic and Applications (FPL). 428--433. Google Scholar
Digital Library
- Xilinx. 1996. Efficient Shift Registers, LFSR Counters, and Long Pseudo-Random Sequence Generators. Retrieved from http://www.xilinx.com/support/documentation/application_notes/xapp052.pdf.Google Scholar
- K. M. Zick and J. P. Hayes. 2010. On-line sensing for healthier FPGA systems. In Proceedings of the ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA). 239--248. Google Scholar
Digital Library
Index Terms
KAPow: High-Accuracy, Low-Overhead Online Per-Module Power Estimation for FPGA Designs
Recommendations
A Message-Scheduling Scheme for Energy Conservation in Multimedia Wireless Systems
Reducing power consumption of wireless networks has become a major goal in designing modern multimedia wireless systems. In an effort to reduce power consumption, this paper addresses the issue of scheduling real-time messages in multimedia wireless ...
The Power of Isolation
CSE '12: Proceedings of the 2012 IEEE 15th International Conference on Computational Science and EngineeringNon-functional system properties such as CPU and memory utilization as well as power consumption are usually non-compositional. However, such properties can be made compositional by isolating individual system components through over-provisioning. The ...
Power-Aware Management in Cloud Data Centers
CloudCom '09: Proceedings of the 1st International Conference on Cloud ComputingPower efficiency is a major concern in operating cloud data centers. It affects operational costs and return on investment, with a profound impact on the environment. Current data center operating environments, such as management consoles and cloud ...






Comments