Characterizing the impact of the workload on the value of dynamic resizing in data centers

Energy consumption imposes a significant cost for data centers; yet much of that energy is used to maintain excess service capacity during periods of predictably low load. Resultantly, there has recently been interest in developing designs that allow the service capacity to be dynamically resized to match the current workload. However, there is still much debate about the value of such approaches in real settings. In this paper, we show that the value of dynamic resizing is highly dependent on statistics of the workload process. In particular, both slow time-scale non-stationarities of the workload (e.g., the peak-to-mean ratio) and the fast time-scale stochasticity (e.g., the burstiness of arrivals) play key roles. To illustrate the impact of these factors, we combine optimization-based modeling of the slow time-scale with stochastic modeling of the fast time-scale. Within this framework, we provide both analytic and numerical results characterizing when dynamic resizing does (and does not) provide benefits.

[1]  Jerome A. Rolia,et al.  Workload Analysis and Demand Prediction of Enterprise Data Center Applications , 2007, 2007 IEEE 10th International Symposium on Workload Characterization.

[2]  Nick G. Duffield,et al.  Exponential bounds for queues with Markovian arrivals , 1994, Queueing Syst. Theory Appl..

[3]  Cheng-Shang Chang,et al.  Performance guarantees in communication networks , 2000, Eur. Trans. Telecommun..

[4]  Vladimir Vinogradov.  Refined Large Deviation Limit Theorems , 1994 .

[5]  Albert G. Greenberg,et al.  The cost of a cloud: research problems in data center networks , 2008, CCRV.

[6]  Adam Wierman,et al.  Renewable and cooling aware workload management for sustainable data centers , 2012, SIGMETRICS '12.

[7]  Florin Ciucu,et al.  Delay Bounds in Communication Networks With Heavy-Tailed and Self-Similar Traffic , 2009, IEEE Transactions on Information Theory.

[8]  Yuan Chen,et al.  Integrated management of application performance, power and cooling in data centers , 2010, 2010 IEEE Network Operations and Management Symposium - NOMS 2010.

[9]  Donald F. Towsley,et al.  Exponential bounds with applications to call admission , 1997, JACM.

[10]  M. Yor,et al.  Continuous martingales and Brownian motion , 1990 .

[11]  V. B. Iversen,et al.  Waiting time distribution in M/D/1 queueing systems , 1999 .

[12]  J. Kingman.  A martingale inequality in the theory of queues , 1964 .

[13]  L H AndrewLachlan,et al.  Dynamic right-sizing for power-proportional data centers , 2013 .

[14]  Mor Harchol-Balter,et al.  The case for sleep states in servers , 2011, HotPower '11.

[15]  Minghua Chen,et al.  Simple and effective dynamic provisioning for power-proportional data centers , 2011, 2012 46th Annual Conference on Information Sciences and Systems (CISS).

[16]  Frank Kelly,et al.  Notes on effective bandwidths , 1994 .

[17]  Florin Ciucu,et al.  Network Calculus Delay Bounds in Queueing Networks with Exact Solutions , 2007, ITC.

[18]  Nagarajan Kandasamy,et al.  Power and performance management of virtualized computing environments via lookahead control , 2008, 2008 International Conference on Autonomic Computing.

[19]  A. Pakes.  ON THE TAILS OF WAITING-TIME DISTRIBUTIONS , 1975 .

[20]  Thomas F. Wenisch,et al.  PowerNap: eliminating server idle power , 2009, ASPLOS.

[21]  Erik D. Demaine,et al.  Energy-Efficient Algorithms , 2016, ITCS.

[22]  Xue Liu,et al.  OptiTuner: On Performance Composition and Server Farm Energy Minimization Application , 2011, IEEE Transactions on Parallel and Distributed Systems.

[23]  Luiz André Barroso,et al.  The Case for Energy-Proportional Computing , 2007, Computer.

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

[25]  Ricardo Bianchini,et al.  Limiting the power consumption of main memory , 2007, ISCA '07.

[26]  T. N. Vijaykumar,et al.  Joint optimization of idle and cooling power in data centers while maintaining response time , 2010, ASPLOS XV.

[27]  Ward Whitt,et al.  Waiting-time tail probabilities in queues with long-tail service-time distributions , 1994, Queueing Syst. Theory Appl..

[28]  David A. Patterson,et al.  A Case For Adaptive Datacenters To Conserve Energy and Improve Reliability , 2008 .

[29]  Lachlan L. H. Andrew,et al.  Online algorithms for geographical load balancing , 2012, 2012 International Green Computing Conference (IGCC).

[30]  Karsten Schwan,et al.  Robust and flexible power-proportional storage , 2010, SoCC '10.

[31]  Gargi Dasgupta,et al.  Server Workload Analysis for Power Minimization using Consolidation , 2009, USENIX Annual Technical Conference.

[32]  Mor Harchol-Balter,et al.  Optimality analysis of energy-performance trade-off for server farm management , 2010, Perform. Evaluation.

[33]  Austin Donnelly,et al.  Sierra: a power-proportional, distributed storage system , 2009 .

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

[35]  Quanyan Zhu,et al.  Dynamic energy-aware capacity provisioning for cloud computing environments , 2012, ICAC '12.

[36]  Suman Nath,et al.  Energy-Aware Server Provisioning and Load Dispatching for Connection-Intensive Internet Services , 2008, NSDI.

[37]  Andrew Warfield,et al.  Live migration of virtual machines , 2005, NSDI.

[38]  Lachlan L. H. Andrew,et al.  Online dynamic capacity provisioning in data centers , 2011, 2011 49th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[39]  Anand Sivasubramaniam,et al.  Statistical profiling-based techniques for effective power provisioning in data centers , 2009, EuroSys '09.

[40]  Lachlan L. H. Andrew,et al.  Dynamic Right-Sizing for Power-Proportional Data Centers , 2011, IEEE/ACM Transactions on Networking.

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

[42]  Ravi Mazumdar,et al.  Performance Modeling, Loss Networks, and Statistical Multiplexing , 2010, Performance Modeling, Loss Networks, and Statistical Multiplexing.