Wes Felter
ABSTRACT
The combination of increasing component power consumption, a desire for denser systems, and the required performance growth in the face of technology-scaling issues are posing enormous challenges for powering and cooling of server systems. The challenges are directly linked to the peak power consumption of servers.Our solution, Power Shifting, reduces the peak power consumption of servers minimizing the impact on performance. We reduce peak power consumption by using workload-guided dynamic allocation of power among components incorporating real-time performance feedback, activity-related power estimation techniques, and performance-sensitive activity-regulation mechanisms to enforce power budgets.We apply our techniques to a computer system with a single processor and memory. Power shifting adds a system power manager with a dynamic, global view of the system's power consumption to continuously re-budget the available power amongst the two components. Our contributions include:• Demonstration of the greater effectiveness of dynamic power allocation over static budgeting,• Evaluation of different power shifting policies,• Analysis of system and workload factors critical to successful power shifting, and• Proposal of performance-sensitive power budget enforcement mechanisms that ensure system reliability.
- Heng Zeng, Xiaobo Fan, Carla Ellis, Alvin Lebeck, and Amin Vahdat. ECOSystem: Managing energy as a first class operating system resource. In Proceedings of 10th Intl. Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-X), October, 2002. Google Scholar
Digital Library
- Heng Zeng, Carla S. Ellis, Alvin R. Lebeck, and Amin Vahdat. Currentcy: A Unifying Abstraction for Expressing Energy Management Policies. In Proceedings of the USENIX 2003 Annual Technical Conference. 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 IEEE Workshop on Hot Topics in Operating Systems (HotOS-VIII), May, 2001. Google ScholarDigital Library
- Murali Annavaram, Ed Grochowski, John Shen. Mitigating Amdahl's Law Through EPI Throttling. In Proceedings of the 32nd Annual International Symposium on Computer Architecture (ISCA-2005), June, 2005. Google ScholarDigital Library
- Charles Lefurgy, Karthick Rajamani, Freeman Rawson, Wes Felter, Michael Kister, and Tom W. Keller. Energy Management for Commercial Servers. In IEEE Computer, Volume 36 (12), December, 2003. Google ScholarDigital Library
- Alper Buyuktosunoglu, Tejas Karkhanis, David H. Albonesi, and Pradip Bose. Energy Efficient Co-Adaptive Instruction Fetch and Issue. In Proceedings of the 30th Annual International Symposium on Computer Architecture (ISCA-2003), June, 2003. Google ScholarDigital Library
- H. Sanchez, B. Kuttanna, T. Olson, M. Alexander, G. Gerosa, R. Philip, J. Alvarez. Thermal Management System for High Performance PowerPC Microprocessors. IEEE, 1997. Google ScholarDigital Library
- IBM PowerPC 750FX RISC Microprocessor User's Manual version 1.01. February, 2003.Google Scholar
- Thermal Monitoring and Protection, IA-32 Intel Architecture Software Developer's Manual, System Programming Guide, Volume 3, 2004.Google Scholar
- Thermal Specifications and Design Considerations, Intel Pentium M Processor on 90-nm Process with 2-MB L2 Cache, Datasheet, October, 2004.Google Scholar
- David Brooks and Margaret Martonosi. Dynamic Thermal Management for High-Performance Microprocessors. In Proceedings of the Seventh International Symposium on High-Performance Computer Architecture (HPCA-7), February, 2001. Google ScholarDigital Library
- Jaime H. Moreno, Mayan Moudgill. Turandot Users's Guide. IBM Research Report RC 21968, February 2001.Google Scholar
- David Brooks, Pradip Bose, Viji Srinivasan, Michael Gschwind, Philip G. Emma, Michael G. Rosenfield. Microarchitectural-Level Power-Performance Analysis: The PowerTimer Approach. In IBM Journal of Research and Development, Volume 47, No. 5/6, Nov. 2003 Google ScholarDigital Library
- Zhigang Hu, David Brooks, Viktor Zyuban and Pradip Bose. Microarchitecture-level power-performance simulators: Modeling, validation, and impact on design. Tutorial at the 36th Annual IEEE/ACM International Symposium on Microarchitecture, December, 2003.Google Scholar
- MEMSIM Users' Guide, IBM Research Report RC23431, October, 2004.Google Scholar
- Mesquite Software, Inc. User's Guide - CSIM19 Simulation Engine.Google Scholar
- Peter Sandon, PowerPC 970: First in a new family of 64-bit high performance PowerPC processors, presented at Microprocessor Forum, March 2002 (available at H/http:/www-06.ibm.com/Hchips/techlib/techlib.nsf/products/PowerPC_970_and_970FX_Microprocessors).Google Scholar
- Standard Performance Evaluation Corporation (SPEC). SPEC CPU2000. Hhttp://www.specbench.org/cpu2000H, October, 2004.Google Scholar
- Standard Performance Evaluation Corporation (SPEC). SPEC JBB2000 (Java Business Benchmark). Hhttp://www.specbench.org/jbb2000H, October 2004.Google Scholar
- Patrick Bhorer, Elmootazbellah Elnozahy, Ahmed Gheith, Charles Lefurgy, Tarun Nakra, Jim Peterson, Ram Rajamony, Ron Rockhold, Hazim Shafi, Rick Simpson, William Speight, Kartik Sudeep, Eric Van Hensbergen, and Lixin Zhang. Mambo - A Full System Simulator for the PowerPC Architecture. ACM SIGMETRICS Performance Evaluation Review, Volume 31, Number 4, March 2004. Google ScholarDigital Library
- John McCalpin. STREAM: Sustainable Memory Bandwidth in High Performance Computers. Hhttp://www.cs.virginia.edu/stream/H, October, 2004.Google Scholar
Index Terms
- A performance-conserving approach for reducing peak power consumption in server systems
Recommendations
Maximizing Performance Under a Power Cap: A Comparison of Hardware, Software, and Hybrid Techniques
ASPLOS '16Power and thermal dissipation constrain multicore performance scaling. Modern processors are built such that they could sustain damaging levels of power dissipation, creating a need for systems that can implement processor power caps. A particular ...
The challenges of implementing fine-grained power gating
GLSVLSI '10: Proceedings of the 20th symposium on Great lakes symposium on VLSIPower consumption in digital systems, especially in portable devices, is a crucial design factor. Due to downscaling of technology, dynamic switching power is not the only relevant source of power consumption anymore as power dissipation caused by ...
Maximizing Performance Under a Power Cap: A Comparison of Hardware, Software, and Hybrid Techniques
ASPLOS '16: Proceedings of the Twenty-First International Conference on Architectural Support for Programming Languages and Operating SystemsPower and thermal dissipation constrain multicore performance scaling. Modern processors are built such that they could sustain damaging levels of power dissipation, creating a need for systems that can implement processor power caps. A particular ...
Comments