Fast Power and Energy Management for Future Many-Core Systems

Future servers will incorporate many active low-power modes for each core and for the main memory subsystem. Though these modes provide flexibility for power and/or energy management via Dynamic Voltage and Frequency Scaling (DVFS), prior work has shown that they must be managed in a coordinated manner. This requirement creates a combinatorial space of possible power mode configurations. As a result, it becomes increasingly challenging to quickly select the configuration that optimizes for both performance and power/energy efficiency. In this article, we propose a novel queuing model for working with the abundant active low-power modes in many-core systems. Based on the queuing model, we derive two fast algorithms that optimize for performance and efficiency using both CPU and memory DVFS. Our first algorithm, called FastCap, maximizes the performance of applications under a full-system power cap, while promoting fairness across applications. Our second algorithm, called FastEnergy, maximizes the full-system energy savings under predefined application performance loss bounds. Both FastCap and FastEnergy operate online and efficiently, using a small set of performance counters as input. To evaluate them, we simulate both algorithms for a many-core server running different types of workloads. Our results show that FastCap achieves better application performance and fairness than prior power capping techniques for the same power budget, whereas FastEnergy conserves more energy than prior energy management techniques for the same performance constraint. FastCap and FastEnergy together demonstrate the applicability of the queuing model for managing the abundant active low-power modes in many-core systems.

[1]  Meeta Sharma Gupta,et al.  System level analysis of fast, per-core DVFS using on-chip switching regulators , 2008, 2008 IEEE 14th International Symposium on High Performance Computer Architecture.

[2]  Qingyuan Deng,et al.  MemScale: active low-power modes for main memory , 2011, ASPLOS XVI.

[3]  Mahmut T. Kandemir,et al.  DRPM: dynamic speed control for power management in server class disks , 2003, 30th Annual International Symposium on Computer Architecture, 2003. Proceedings..

[4]  Mor Harchol-Balter,et al.  AutoScale: Dynamic, Robust Capacity Management for Multi-Tier Data Centers , 2012, TOCS.

[5]  Stefanos Kaxiras,et al.  Green governors: A framework for Continuously Adaptive DVFS , 2011, 2011 International Green Computing Conference and Workshops.

[6]  Carla Schlatter Ellis,et al.  The Synergy Between Power-Aware Memory Systems and Processor Voltage Scaling , 2003, PACS.

[7]  Alan J. Weger,et al.  Power management of multi-core chips: Challenges and pitfalls , 2012, 2012 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[8]  Kirk Pruhs,et al.  Speed scaling to manage energy and temperature , 2007, JACM.

[9]  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).

[10]  Yuguang Fang.  Analysis of queuing networks with blocking [Book Review] , 2002, IEEE Wireless Communications.

[11]  Bin Li,et al.  Dynamo: Facebook's Data Center-Wide Power Management System , 2016, 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA).

[12]  Margaret Martonosi,et al.  An Analysis of Efficient Multi-Core Global Power Management Policies: Maximizing Performance for a Given Power Budget , 2006, 2006 39th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO'06).

[13]  Thomas F. Wenisch,et al.  CoScale: Coordinating CPU and Memory System DVFS in Server Systems , 2012, 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture.

[14]  Lachlan L. H. Andrew,et al.  Power-Aware Speed Scaling in Processor Sharing Systems , 2009, IEEE INFOCOM 2009.

[15]  Pradip Bose,et al.  Mitigating Power Contention: A Scheduling Based Approach , 2017, IEEE Computer Architecture Letters.

[16]  Pradip Bose,et al.  Evaluating design tradeoffs in on-chip power management for CMPs , 2007, Proceedings of the 2007 international symposium on Low power electronics and design (ISLPED '07).

[17]  Yonggang Wen,et al.  Data Center Energy Consumption Modeling: A Survey , 2016, IEEE Communications Surveys & Tutorials.

[18]  Mor Harchol-Balter,et al.  Performance Modeling and Design of Computer Systems: Queueing Theory in Action , 2013 .

[19]  Nam Sung Kim,et al.  Queuing theoretic analysis of power-performance tradeoff in power-efficient computing , 2013, 2013 47th Annual Conference on Information Sciences and Systems (CISS).

[20]  Hong Liu,et al.  Energy proportional datacenter networks , 2010, ISCA.

[21]  Li Shang,et al.  Multi-Optimization power management for chip multiprocessors , 2008, 2008 International Conference on Parallel Architectures and Compilation Techniques (PACT).

[22]  Simonetta Balsamo,et al.  Analysis of Queueing Networks with Blocking , 2010 .

[23]  Josep Torrellas,et al.  Variation-Aware Application Scheduling and Power Management for Chip Multiprocessors , 2008, 2008 International Symposium on Computer Architecture.

[24]  Chris Fallin,et al.  Memory power management via dynamic voltage/frequency scaling , 2011, ICAC '11.

[25]  Lachlan L. H. Andrew,et al.  Power-aware speed scaling in processor sharing systems: Optimality and robustness , 2012, Perform. Evaluation.

[26]  Xue Li,et al.  Coordinating processor and main memory for efficientserver power control , 2011, ICS '11.

[27]  Mor Harchol-Balter,et al.  How data center size impacts the effectiveness of dynamic power management , 2011, 2011 49th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[28]  Kai Ma,et al.  Scalable power control for many-core architectures running multi-threaded applications , 2011, 2011 38th Annual International Symposium on Computer Architecture (ISCA).

[29]  Xiaodong Li,et al.  Performance directed energy management for main memory and disks , 2004, ASPLOS XI.

[30]  Ricardo Bianchini,et al.  Conserving disk energy in network servers , 2003, ICS '03.

[31]  Zhao Zhang,et al.  Decoupled DIMM: building high-bandwidth memory system using low-speed DRAM devices , 2009, ISCA '09.

[32]  Nam Sung Kim,et al.  SleepScale: Runtime joint speed scaling and sleep states management for power efficient data centers , 2014, 2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA).

[33]  Xiaodong Li,et al.  Cross-component energy management: Joint adaptation of processor and memory , 2007, TACO.

[34]  Stefanos Kaxiras,et al.  Power Token Balancing: Adapting CMPs to Power Constraints for Parallel Multithreaded Workloads , 2011, 2011 IEEE International Parallel & Distributed Processing Symposium.

[35]  Rizwana Begum,et al.  Algorithms for CPU and DRAM DVFS under inefficiency constraints , 2016, 2016 IEEE 34th International Conference on Computer Design (ICCD).

[36]  Mahmut T. Kandemir,et al.  CPM in CMPs: Coordinated Power Management in Chip-Multiprocessors , 2010, 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis.

[37]  Minyi Guo,et al.  AgileRegulator: A hybrid voltage regulator scheme redeeming dark silicon for power efficiency in a multicore architecture , 2012, IEEE International Symposium on High-Performance Comp Architecture.

[38]  Thomas F. Wenisch,et al.  Power management of online data-intensive services , 2011, 2011 38th Annual International Symposium on Computer Architecture (ISCA).

[39]  Gero Dittmann,et al.  Exploring power management in multi-core systems , 2008, 2008 Asia and South Pacific Design Automation Conference.

[40]  Ricardo Bianchini,et al.  FastCap: An efficient and fair algorithm for power capping in many-core systems , 2016, 2016 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).