Energy-Efficient Heuristics for Insensitive Job Assignment in Processor-Sharing Server Farms

Energy efficiency of server farms is an important design consideration of the green datacenter initiative. One effective approach is to optimize power consumption of server farms by controlling the carried load on the networked servers. In this paper, we propose a robust heuristic policy called E* for stochastic job assignment in a server farm, aiming to improve the energy efficiency by maximizing the ratio of job throughput to power consumption. Our model of the server farm considers a parallel system of finite-buffer processor-sharing queues with heterogeneous server speeds and energy consumption rates. We devise E* as an insensitive policy so that the stationary distribution of the number of jobs in the system depends on the job size distribution only through its mean. We provide a rigorous analysis of E* and compare it with a baseline approach, known as most energy-efficient server first (MEESF), that greedily chooses the most energy-efficient servers for job assignment. We show that E* has always a higher job throughput than that of MEESF, and derive realistic conditions under which E* is guaranteed to outperform MEESF in energy efficiency. Extensive numerical results are presented and demonstrate that E* can improve the energy efficiency by up to 100%.

[1]  A. Wierman,et al.  Optimality, fairness, and robustness in speed scaling designs , 2010, SIGMETRICS '10.

[2]  Yiqiang Zhao,et al.  Queueing Analysis of a Jockeying Model , 1995, Oper. Res..

[3]  Ward Whitt,et al.  Insensitivity for PS server farms with JSQ routing , 2007, PERV.

[4]  Philip S. Yu,et al.  The state of the art in locally distributed Web-server systems , 2002, CSUR.

[5]  Lisandro Zambenedetti Granville,et al.  Data Center Network Virtualization: A Survey , 2013, IEEE Communications Surveys & Tutorials.

[6]  Spyridon Antonakopoulos,et al.  Energy-aware scheduling algorithms for network stability , 2011, 2011 Proceedings IEEE INFOCOM.

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

[8]  Parag Agrawal,et al.  The case for RAMClouds: scalable high-performance storage entirely in DRAM , 2010, OPSR.

[9]  Moshe Zukerman,et al.  Insensitive Job Assignment With Throughput and Energy Criteria for Processor-Sharing Server Farms , 2014, IEEE/ACM Transactions on Networking.

[10]  Ivo J. B. F. Adan,et al.  Matrix-geometric analysis of the shortest queue problem with threshold jockeying , 1993, Oper. Res. Lett..

[11]  N. Bansal,et al.  Speed scaling with an arbitrary power function , 2009, SODA 2009.

[12]  Moshe Zukerman,et al.  Energy-efficient heuristics for job assignment in processor-sharing server farms , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[13]  F. Kelly,et al.  Networks of queues , 1976, Advances in Applied Probability.

[14]  Shaler Stidham,et al.  Forward Recursion for Markov Decision Processes with Skip-Free-to-the-Right Transitions, Part I: Theory and Algorithm , 1986, Math. Oper. Res..

[15]  Esa Hyytiä,et al.  Energy-aware job assignment in server farms with setup delays under LCFS and PS , 2014, 2014 26th International Teletraffic Congress (ITC).

[16]  W. Whitt,et al.  Analysis of join-the-shortest-queue routing for web server farms , 2007, Perform. Evaluation.

[17]  Maurizio Portolani,et al.  Data Center Fundamentals , 2003 .

[18]  Steven A. Lippman,et al.  Applying a New Device in the Optimization of Exponential Queuing Systems , 1975, Oper. Res..

[19]  Mor Harchol-Balter,et al.  Stochastic Models and Analysis for Resource Management in Server Farms , 2011 .

[20]  J. Michael Harrison,et al.  Dynamic Control of a Queue with Adjustable Service Rate , 2001, Oper. Res..

[21]  Yutaka Sakuma,et al.  Asymptotic behavior for MAP/PH/c queue with shortest queue discipline and jockeying , 2010, Oper. Res. Lett..

[22]  F. Frances Yao,et al.  A scheduling model for reduced CPU energy , 1995, Proceedings of IEEE 36th Annual Foundations of Computer Science.

[23]  Flavio Bonomi,et al.  On Job Assignment for a Parallel System of Processor Sharing Queues , 1990, IEEE Trans. Computers.

[24]  Azer Bestavros,et al.  Self-similarity in World Wide Web traffic: evidence and possible causes , 1996, SIGMETRICS '96.

[25]  Dimitri P. Bertsekas,et al.  Dynamic Programming and Optimal Control, Two Volume Set , 1995 .

[26]  Kenli Li,et al.  Energy-Efficient Stochastic Task Scheduling on Heterogeneous Computing Systems , 2014, IEEE Transactions on Parallel and Distributed Systems.

[27]  Wei Xiang,et al.  Design and Performance Analysis of An Energy-Efficient Uplink Carrier Aggregation Scheme , 2014, IEEE Journal on Selected Areas in Communications.

[28]  Urtzi Ayesta,et al.  Load balancing in processor sharing systems , 2011, Telecommun. Syst..

[29]  Ward Whitt,et al.  Deciding Which Queue to Join: Some Counterexamples , 1986, Oper. Res..

[30]  Frank A. Haight,et al.  TWO QUEUES IN PARALLEL , 1958 .

[31]  Albert Y. Zomaya,et al.  A Taxonomy and Survey of Energy-Efficient Data Centers and Cloud Computing Systems , 2010, Adv. Comput..

[32]  R. Weber.  On the optimal assignment of customers to parallel servers , 1978, Journal of Applied Probability.

[33]  Shaler Stidham,et al.  Monotonic and Insensitive Optimal Policies for Control of Queues with Undiscounted Costs , 1989, Oper. Res..

[34]  Mor Harchol-Balter,et al.  Optimal power allocation in server farms , 2009, SIGMETRICS '09.

[35]  Anthony Ephremides,et al.  A simple dynamic routing problem , 1980 .

[36]  Kirk Pruhs,et al.  Speed scaling for weighted flow time , 2007, SODA '07.

[37]  A. Barbour.  Networks of queues and the method of stages , 1976, Advances in Applied Probability.

[38]  P. Taylor.  Insensitivity in Stochastic Models , 2011 .

[39]  Steven Swanson,et al.  Gordon: using flash memory to build fast, power-efficient clusters for data-intensive applications , 2009, ASPLOS.

[40]  Susanne Albers,et al.  Energy-efficient algorithms for flow time minimization , 2006, STACS.

[41]  Subodha Gunawardena,et al.  Service Response Time of Elastic Data Traffic in Cognitive Radio Networks , 2013, IEEE J. Sel. Areas Commun..

[42]  Tapani Lehtonen,et al.  On the optimality of the shortest line discipline , 1984 .

[43]  Isi Mitrani.  Managing performance and power consumption in a server farm , 2013, Ann. Oper. Res..

[44]  Esa Hyytiä,et al.  Task assignment in a heterogeneous server farm with switching delays and general energy-aware cost structure , 2014, Perform. Evaluation.