Safe Overprovisioning: Using Power Limits to Increase Aggregate Throughput

Management of power in data centers is driven by the need to not exceed circuit capacity. The methods employed in the oversight of these power circuits are typically static and ad-hoc. New power-scalable system components allow for dynamically controlling power consumption with an accompanying effect on performance. Because the incremental performance gain from operating in a higher performance state is less than the increase in power, it is possible to overprovision the hardware infrastructure to increase throughput and yet still remain below an aggregate power limit. In overprovisioning, if each component operates at maximum power the limit would be exceeded with disastrous results. However, safe overprovisioning regulates power consumption locally to meet the global power budget. Host-based and network-centric models are proposed to monitor and coordinate the distribution of power with the fundamental goal of increasing throughput. This research work presents the advantages of overprovisioning and describes a general framework and an initial prototype. Initial results with a synthetic benchmark indicate throughput increases of nearly 6% from a staticly assigned, power managed environment and over 30% from an unmanaged environment.

[1]  Paul Horton,et al.  A Quantitative Analysis of Disk Drive Power Management in Portable Computers , 1994, USENIX Winter.

[2]  Fred Douglis,et al.  Adaptive Disk Spin-Down Policies for Mobile Computers , 1995, Comput. Syst..

[3]  Darrell D. E. Long,et al.  A dynamic disk spin-down technique for mobile computing , 1996, MobiCom '96.

[4]  David Mosberger,et al.  httperf—a tool for measuring web server performance , 1998, PERV.

[5]  Thomas D. Burd,et al.  The simulation and evaluation of dynamic voltage scaling algorithms , 1998, Proceedings. 1998 International Symposium on Low Power Electronics and Design (IEEE Cat. No.98TH8379).

[6]  Carla Schlatter Ellis,et al.  The case for higher-level power management , 1999, Proceedings of the Seventh Workshop on Hot Topics in Operating Systems.

[7]  Mahadev Satyanarayanan,et al.  PowerScope: a tool for profiling the energy usage of mobile applications , 1999, Proceedings WMCSA'99. Second IEEE Workshop on Mobile Computing Systems and Applications.

[8]  Amin Vahdat,et al.  Every joule is precious: the case for revisiting operating system design for energy efficiency , 2000, ACM SIGOPS European Workshop.

[9]  J. Flinn,et al.  Energy-aware adaptation for mobile applications , 1999, SOSP.

[10]  Frank Bellosa,et al.  The benefits of event: driven energy accounting in power-sensitive systems , 2000, ACM SIGOPS European Workshop.

[11]  Soonhoi Ha,et al.  Dynamic voltage scheduling technique for low-power multimedia applications using buffers , 2001, ISLPED '01.

[12]  Flavius Gruian Hard real-time scheduling for low-energy using stochastic data and DVS processors , 2001, ISLPED'01: Proceedings of the 2001 International Symposium on Low Power Electronics and Design (IEEE Cat. No.01TH8581).

[13]  Enrique V. Carrera,et al.  Load balancing and unbalancing for power and performance in cluster-based systems , 2001 .

[14]  Margaret Martonosi,et al.  Run-time power estimation in high performance microprocessors , 2001, ISLPED '01.

[15]  K. Langendoen,et al.  Energy priority scheduling for variable voltage processors , 2001, ISLPED'01: Proceedings of the 2001 International Symposium on Low Power Electronics and Design (IEEE Cat. No.01TH8581).

[16]  Amin Vahdat,et al.  Managing energy and server resources in hosting centers , 2001, SOSP.

[17]  Trevor N. Mudge,et al.  Power: A First-Class Architectural Design Constraint , 2001, Computer.

[18]  David F. Heidel,et al.  An Overview of the BlueGene/L Supercomputer , 2002, ACM/IEEE SC 2002 Conference (SC'02).

[19]  Michael Kistler,et al.  The case for power management in web servers , 2002 .

[20]  Ragunathan Rajkumar,et al.  Critical power slope: understanding the runtime effects of frequency scaling , 2002, ICS '02.

[21]  E. N. Elnozahy,et al.  Energy-Efficient Server Clusters , 2002, PACS.

[22]  Vincent W. Freeh,et al.  Dynamic Power Management using Feedback , 2002 .

[23]  Amin Vahdat,et al.  Currentcy: Unifying Policies for Resource Management , 2002 .

[24]  Vincent K. N. Lau,et al.  Automatic Performance Setting for Dynamic Voltage Scaling , 2002, Wirel. Networks.

[25]  Ricardo Bianchini,et al.  Dynamic cluster reconfiguration for power and performance , 2003 .

[26]  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..

[27]  Amin Vahdat,et al.  Currentcy: A Unifying Abstraction for Expressing Energy Management Policies , 2003, USENIX Annual Technical Conference, General Track.

[28]  Mahmut T. Kandemir,et al.  Reducing Disk Power Consumption in Servers with DRPM , 2003, Computer.

[29]  Karthick Rajamani,et al.  Energy Management for Commercial Servers , 2003, Computer.

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

[31]  Richard E. Harper,et al.  Workload-based power management for parallel computer systems , 2003, IBM J. Res. Dev..

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

[33]  Kevin Skadron,et al.  Power-aware QoS management in Web servers , 2003, RTSS 2003. 24th IEEE Real-Time Systems Symposium, 2003.

[34]  Y. Charlie Hu,et al.  Program counter based techniques for dynamic power management , 2004, 10th International Symposium on High Performance Computer Architecture (HPCA'04).

[35]  David Robinson,et al.  The Common Gateway Interface (CGI) Version 1.1 , 2004, RFC.

[36]  Jason Flinn,et al.  Self-Tuning Wireless Network Power Management , 2005, Wirel. Networks.