More Web Proxy on the site http://driver.im/

article

Optimality analysis of energy-performance trade-off for server farm management

Authors:

Mor Harchol-Balter,

Michael A. KozuchAuthors Info & Claims

Performance Evaluation, Volume 67, Issue 11

Pages 1155 - 1171

https://doi.org/10.1016/j.peva.2010.08.009

Published: 01 November 2010 Publication History

Abstract

A central question in designing server farms today is how to efficiently provision the number of servers to extract the best performance under unpredictable demand patterns while not wasting energy. While one would like to turn servers off when they become idle to save energy, the large setup cost (both, in terms of setup time and energy penalty) needed to switch the server back on can adversely affect performance. The problem is made more complex by the fact that today's servers provide multiple sleep or standby states which trade off the setup cost with the power consumed while the server is 'sleeping'. With so many controls, finding the optimal server farm management policy is an almost intractable problem - How many servers should be on at any given time, how many should be off, and how many should be in some sleep state? In this paper, we employ the popular metric of Energy-Response time Product (ERP) to capture the energy-performance trade-off, and present the first theoretical results on the optimality of server farm management policies. For a stationary demand pattern, we prove that there exists a very small, natural class of policies that always contains the optimal policy for a single server, and conjecture it to contain a near-optimal policy for multi-server systems. For time-varying demand patterns, we propose a simple, traffic-oblivious policy and provide analytical and empirical evidence for its near-optimality.

References

[1]

}}US Environmental Protection Agency, EPA Report on Server and Data Center Energy Efficiency, 2007.

[2]

}}Barroso, L.A. and Hölzle, U., The case for energy-proportional computing. Computer. v40 i12. 33-37.

[3]

}}Kang, C.W., Abbaspour, S. and Pedram, M., Buffer sizing for minimum energy-delay product by using an approximating polynomial. In: GLSVLSI'03: Proceedings of the 13th ACM Great Lakes Symposium on VLSI, ACM, New York, NY, USA. pp. 112-115.

[4]

}}Stan, M.R. and Skadron, K., Power-aware computing: guest editorial. IEEE Comput. v36 i12. 35-38.

[5]

}}Gonzalez, R. and Horowitz, M., Energy dissipation in general purpose microprocessors. IEEE J. Solid-State Circuits. v31 i9. 1277-1284.

[6]

}}J. Kin, M. Gupta, W. Mangione-Smith, The filter cache: an energy efficient memory structure, in: Microarchitecture, in: IEEE/ACM International Symposium on, 0, 1997, p. 184.

[7]

}}Juang, P., Wu, Q., Peh, L.-S., Martonosi, M. and Clark, D.W., Coordinated, distributed, formal energy management of chip multiprocessors. In: ISLPED'05: Proceedings of the 2005 International Symposium on Low Power Electronics and Design, ACM, New York, NY, USA. pp. 127-130.

[8]

}}Albers, S. and Fujiwara, H., Energy-efficient algorithms for flow time minimization. ACM Trans. Algorithms. v3 i4. 49

[9]

}}Bansal, N., Chan, H.-L. and Pruhs, K., Speed scaling with an arbitrary power function. In: SODA'09: Proceedings of the Nineteenth Annual ACM-SIAM Symposium on Discrete Algorithms, Society for Industrial and Applied Mathematics, Philadelphia, PA, USA. pp. 693-701.

[10]

}}A. Wierman, L.L.H. Andrew, A. Tang, Power-aware speed scaling in processor sharing systems. in: INFOCOM, 2009.

[11]

}}Irani, S., Shukla, S. and Gupta, R., Algorithms for power savings. ACM Trans. Algorithms. v3 i4. 41

[12]

}}Eggert, L. and Touch, J.D., Idletime scheduling with preemption intervals. SIGOPS Oper. Syst. Rev. v39 i5. 249-262.

[13]

}}Riska, A., Mi, N., Smirni, E. and Casale, G., Feasibility regions: exploiting trade-offs between power and performance in disk drives. SIGMETRICS Perform. Eval. Rev. v37 i3. 43-48.

[14]

}}Borst, S.C., Mandelbaum, A., Reiman, M.I. and Centrum, M., Dimensioning large call centers. Oper. Res. v52. 17-34.

[15]

}}Jennings, O.B., Mandelbaum, A., Massey, W.A. and Whitt, W., Server staffing to meet time-varying demand. Manage. Sci. v42. 1383-1394.

[16]

}}Adan, I. and van der Wal, J., Combining make to order and make to stock. OR Spektrum. v20. 73-81.

[17]

}}Irani, S. and Pruhs, K.R., Algorithmic problems in power management. SIGACT News. v36 i2. 63-76.

[18]

}}F. Yao, A. Demers, S. Shenker, A scheduling model for reduced CPU energy, Annual IEEE Symposium on Foundations of Computer Science, 0: 374 1995.

[19]

}}Bansal, N., Kimbrel, T. and Pruhs, K., Dynamic speed scaling to manage energy and temperature. In: FOCS'04: Proceedings of the 45th Annual IEEE Symposium on Foundations of Computer Science, IEEE Computer Society, Washington, DC, USA. pp. 520-529.

Digital Library

[20]

}}Bansal, N., Kimbrel, T. and Pruhs, K., Speed scaling to manage energy and temperature. J. ACM. v54 i1. 1-39.

[21]

}}Pruhs, K., Uthaisombut, P. and Woeginger, G., Getting the best response for your ERG. ACM Trans. Algorithms. v4 i3. 1-17.

[22]

}}Intel Corp. Intel Math Kernel Library 10.0-LINPACK, 2007. http://www.intel.com/cd/software/products/asmo-na/eng/266857.htm.

[23]

}}A. Gandhi, V. Gupta, M. Harchol-Balter, M. Kozuch, Energy-efficient dynamic capacity provisioning in server farms, Technical Report CMU-CS-10-108, School of Computer Science, Carnegie Mellon University, 2010.

[24]

}}Halfin, S. and Whitt, W., Heavy-traffic limits for queues with many exponential servers. Oper. Res. v29 i3. 567-588.

[25]

}}The internet traffic archives: WorldCup98. Available at: http://ita.ee.lbl.gov/html/contrib/WorldCup.html.

Cited By

Bhuyan NMukherjee DWierman ASalakhutdinov RKolter ZHeller KWeller AOliver NScarlett JBerkenkamp F(2024)Best of both worlds guarantees for smoothed online quadratic optimizationProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3692225(3850-3888)Online publication date: 21-Jul-2024
https://dl.acm.org/doi/10.5555/3692070.3692225
Rutten DMukherjee D(2024)A New Approach to Capacity Scaling Augmented with Unreliable Machine Learning PredictionsMathematics of Operations Research10.1287/moor.2023.136449:1(476-508)Online publication date: 1-Feb-2024
https://dl.acm.org/doi/10.1287/moor.2023.1364
Fé INguyen TMauro MPostiglione FRamos ASoares AChoi EMin DLee JSilva F(2024)Energy-aware dynamic response and efficient consolidation strategies for disaster survivability of cloud microservices architectureComputing10.1007/s00607-024-01305-x106:8(2737-2783)Online publication date: 1-Aug-2024
https://dl.acm.org/doi/10.1007/s00607-024-01305-x
Show More Cited By

Recommendations

Server farms with setup costs

In this paper we consider server farms with a setup cost. This model is common in manufacturing systems and data centers, where there is a cost to turn servers on. Setup costs always take the form of a time delay, and sometimes there is additionally a ...
Managing Server Clusters on Renewable Energy Mix

As climate change has become a global concern and server energy demand continues to soar, many IT companies have started to explore server clusters running on various renewable energy sources. Existing green data center designs often yield suboptimal ...
Prediction free energy neutral power management for energy harvesting wireless sensor nodes

Current power management mechanisms for energy harvesting wireless sensors typically rely on predicted information about the amount of energy that can be harvested in the future. However, such mechanisms suffer from inevitable prediction errors, which ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image Performance Evaluation

Performance Evaluation Volume 67, Issue 11

November, 2010

313 pages

ISSN:0166-5316

Issue’s Table of Contents

Copyright © Elsevier B.V. © 2010.

Publisher

Elsevier Science Publishers B. V.

Netherlands

Publication History

Published: 01 November 2010

Author Tags

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

63
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 12 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Bhuyan NMukherjee DWierman ASalakhutdinov RKolter ZHeller KWeller AOliver NScarlett JBerkenkamp F(2024)Best of both worlds guarantees for smoothed online quadratic optimizationProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3692225(3850-3888)Online publication date: 21-Jul-2024
https://dl.acm.org/doi/10.5555/3692070.3692225
Rutten DMukherjee D(2024)A New Approach to Capacity Scaling Augmented with Unreliable Machine Learning PredictionsMathematics of Operations Research10.1287/moor.2023.136449:1(476-508)Online publication date: 1-Feb-2024
https://dl.acm.org/doi/10.1287/moor.2023.1364
Fé INguyen TMauro MPostiglione FRamos ASoares AChoi EMin DLee JSilva F(2024)Energy-aware dynamic response and efficient consolidation strategies for disaster survivability of cloud microservices architectureComputing10.1007/s00607-024-01305-x106:8(2737-2783)Online publication date: 1-Aug-2024
https://dl.acm.org/doi/10.1007/s00607-024-01305-x
Golovin ABasmadjian RAstafiev SRumyantsev A(2023)Little’s Law in a Single-Server System with Inactive State for Demand-Response in Data Centers with Green SLAsCompanion Proceedings of the 14th ACM International Conference on Future Energy Systems10.1145/3599733.3600255(91-97)Online publication date: 20-Jun-2023
https://dl.acm.org/doi/10.1145/3599733.3600255
Hsieh CPhung-Duc TRen YChen J(2023)Design and Analysis of Dynamic Block-Setup Reservation Algorithm for 5G Network SlicingIEEE Transactions on Mobile Computing10.1109/TMC.2022.316903422:9(5140-5154)Online publication date: 1-Sep-2023
https://dl.acm.org/doi/10.1109/TMC.2022.3169034
Williams JHarchol-Balter MWang W(2022)The M/M/k with Deterministic Setup TimesProceedings of the ACM on Measurement and Analysis of Computing Systems10.1145/35706176:3(1-45)Online publication date: 8-Dec-2022
https://dl.acm.org/doi/10.1145/3570617
Ali-Eldin AWang BShenoy Pde Supinski BHall MGamblin T(2021)The hidden cost of the edgeProceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis10.1145/3458817.3476142(1-12)Online publication date: 14-Nov-2021
https://dl.acm.org/doi/10.1145/3458817.3476142
Harchol-Balter M(2021)Open problems in queueing theory inspired by datacenter computingQueueing Systems: Theory and Applications10.1007/s11134-020-09684-697:1-2(3-37)Online publication date: 27-Jan-2021
https://dl.acm.org/doi/10.1007/s11134-020-09684-6
Deiana ELatouche GRemiche M(2021)Fluid Flow Model for Energy-Aware Server Performance EvaluationMethodology and Computing in Applied Probability10.1007/s11009-020-09784-z23:3(801-821)Online publication date: 1-Sep-2021
https://dl.acm.org/doi/10.1007/s11009-020-09784-z
Zhang SLiu YHu NZhou N(2020)A novel cost-aware algorithm for dynamic task placement problem in a heterogeneous Internet-scale data centerThe Journal of Supercomputing10.1007/s11227-019-02892-976:9(6579-6598)Online publication date: 1-Sep-2020
https://dl.acm.org/doi/10.1007/s11227-019-02892-9
Show More Cited By

View Options

View options

Media

Figures

Other

Tables

View Issue’s Table of Contents