Abstract
At a high level, data centres are large IT facilities hosting physical machines (servers) that often run a large number of virtual machines (VMs)—but at a lower level, data centres are an intricate collection of interconnected and virtualised computers, connected services, complex service-level agreements. While data centre managers know that reassigning VMs to the servers that would best serve them and also minimise some cost for the company can potentially save a lot of money—the search space is large and constrained, and the decision complicated as they involve different dimensions. This paper consists of a comparative study of heuristics and exact algorithms for the multi-objective machine reassignment problem. Given the common intuition that the problem is too complicated for exact resolutions, all previous works have focused on various (meta)heuristics such as First-Fit, GRASP, NSGA-II or PLS. In this paper, we show that the state-of-art solution to the single objective formulation of the problem (CBLNS) and the classical multi-objective solutions fail to bridge the gap between the number, quality and variety of solutions. Hybrid metaheuristics, on the other hand, have proven to be more effective and efficient to address the problem—but as there has never been any study of an exact resolution, it was difficult to qualify their results. In this paper, we present the most relevant techniques used to address the problem, and we compare them to an exact resolution (\(\epsilon \)-Constraints). We show that the problem is indeed large and constrained (we ran our algorithm for 30 days on a powerful node of a supercomputer and did not get the final solution for most instances of our problem) but that a metaheuristic (GeNePi) obtains acceptable results: more (+ 188%) solutions than the exact resolution and a little more than half (52%) the hypervolume (measure of quality of the solution set).
Similar content being viewed by others
Notes
Source: http://www.ovh.com/fr/backstage/—accessed on 16/05/2018.
A Service-Level Agreement (SLA) is a contract agreed between a data centre provider and a customer which describes the service provided (e.g., allocated resources, time to recover after an outage).
The concept of safety capacity is introduced in the Google/ROADEF/EURO challenge (2012): if one or several resources of a machine are over-loaded then the machine may not be able to satisfy its SLAs.
Pareto set: a set of non-dominated solutions (i.e., better than all other solutions in one or more objectives).
References
Alsheddy, A., Tsang, E.E.P.K.: Guided pareto local search based frameworks for biobjective optimization. In: CEC, pp. 1–8 (2010)
Angel, E., Bampis, E., Gourves, L.: A dynasearch neighborhood for the bicriteria traveling salesman problem. In: Metaheuristics for Multiobjective Optimisation, pp. 153–176 (2004)
Arcuri, A., Briand, L.: A practical guide for using statistical tests to assess randomized algorithms in software engineering. In: ICSE, pp. 1–10 (2011)
Basseur, M.: Design of cooperative algorithms for multi-objective optimization: application to the flow-shop scheduling problem. 4OR, pp. 255–258 (2006)
Beck, J., Siewiorek, D.: Modeling multicomputer task allocation as a vector packing problem. In: ISSS, p. 115 (1996)
Beloglazov, A., Abawajy, J., Buyya, R.: Energy-aware resource allocation heuristics for efficient management of data centers for cloud computing. In: FGCS, pp. 755–768 (2012)
Bin, E., Biran, O., Boni, O., Hadad, E., Kolodner, E.K., Moatti, Y., Lorenz, D.H.: Guaranteeing high availability goals for virtual machine placement. In: ICDCS, pp. 700–709 (2011)
Brandt, F., Speck, J., Völker, M.: Constraint-based large neighborhood search for machine reassignment. Ann. Oper. Res. 242, 63–91 (2016)
Butelle, F., Alfandari, L., Coti, C., Finta, L., Létocart, L., Plateau, G., Roupin, F., Rozenknop, A., Calvo, R.W.: Fast machine reassignment. Ann. Oper. Res. (2016). https://doi.org/10.1007/s10479-015-2082-3
Caprara, A., Toth, P.: Lower bounds and algorithms for the 2-dimensional vector packing problem. Discrete Appl. Math. 111, 231–262 (2001)
Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.A.M.T.: A fast and elitist multiobjective genetic algorithm: Nsga-ii. TEVC 6, 182–197 (2002)
Doddavula, S.K., Kaushik, M., Jain, A.: Implementation of a fast vector packing algorithm and its application for server consolidation. In: CloudCom, pp. 332–339 (2011)
Donyagard Vahed, N., Ghobaei-Arani, M., Souri, A.: Multiobjective virtual machine placement mechanisms using nature-inspired metaheuristic algorithms in cloud environments: a comprehensive review. Int. J. Commun. Syst. 32, e4068 (2019)
European Commission. Datacentre energy efficiency (2007). http://re.jrc.ec.europa.eu
Falkenauer, E.: Genetic Algorithms and Grouping Problems. Wiley, New York (1998)
Feo, T.A., Resende, M.G.C.: Greedy randomized adaptive search procedures. JGO 6, 109–133 (1995)
Filani, D., He, J., Gao, S., Rajappa, M., Kumar, A., Shah, P., Nagappan, R.: Dynamic data center power management: trends, issues, and solutions. Intel Technol. J. 12(1), 59–67 (2008)
Fleischer, M.: The measure of pareto optima applications to multi-objective metaheuristics. In: EMO, pp. 519–533 (2003)
Gabay, M., Zaourar, S.: A GRASP approach for the machine reassignment problem. In: EURO (2012)
Gandibleux, X.: Peek–shape–grab: a methodology in three stages for approximating the non-dominated points of multiobjective discrete/combinatorial optimization problems with a multiobjective metaheuristic. In: EMO, pp. 221–235 (2017)
Gavranović, H., Buljubašić, M., Demirović, E.: Variable neighborhood search for google machine reassignment problem. ENDM 39, 209–216 (2012)
Gendreau, M., Laporte, G., Semet, F.: Heuristics and lower bounds for the bin packing problem with conflicts. Comput. Oper. Res. 31, 347–358 (2004)
Google/ROADEF/EURO CHALLENGE 2012. Google/roadef/euro challenge (2012). http://challenge.roadef.org/2012/en/
Graham, R.L: Bounds on multiprocessing anomalies and related packing algorithms. In: SJCC, pp. 205–217 (1972)
Hoffmann, R., Riff, M.C., Montero, E., Rojas, N.: Google challenge: a hyperheuristic for the machine reassignment problem. In: CEC, pp. 846–853 (2015)
Jansen, K., Öhring, S.: Approximation algorithms for time constrained scheduling. Inform. Comput. 132, 85–108 (1997)
Jaśkowski, W., Szubert, M., Gawron, P.: A hybrid mip-based large neighborhood search heuristic for solving the machine reassignment problem. Ann. Oper. Res. 242, 1–30 (2015)
Jung, G., Hiltunen, M., Joshi, K.R., Schlichting, R.D., Pu, C., et al.: Mistral: dynamically managing power, performance, and adaptation cost in cloud infrastructures. In: ICDCS, pp. 62–73 (2010)
Kellerer, H., Kotov, V.: An approximation algorithm with absolute worst-case performance ratio 2 for two-dimensional vector packing. Oper. Res. Lett. 31, 35–41 (2003)
Leinberger, W., Karypis, G., Kumar, V.: Multi-capacity bin packing algorithms with applications to job scheduling under multiple constraints. In: ICPP, pp. 404–412 (1999)
Li, X., Ventresque, A., Stokes, N., Thorburn, J., Murphy, J.: ivmp: an interactive VM placement algorithm for agile capital allocation. In: CLOUD, pp. 950–951 (2013)
Li, X., Ventresque, A., Murphy, J., Thorburn, J.: A fair comparison of VM placement heuristics and a more effective solution. In: IEEE 13th International Symposium on Parallel and Distributed Computing, ISPDC, pp. 35–42 (2014)
Lien, C.-H., Bai, Y.-W., Lin, M.-B.: Estimation by software for the power consumption of streaming-media servers. TIM 56, 1859–1870 (2007)
Lopes, R., Morais, V.W.C., Noronha, T.F., Souza, V.A.A.: Heuristics and matheuristics for a real-life machine reassignment problem. ITOR 22, 77–95 (2015)
Malitsky, Y., Mehta, D., O’Sullivan, B., Simonis, H.: Tuning parameters of large neighborhood search for the machine reassignment problem. In: CPAIOR, pp. 176–192 (2013)
Mann, Z.A.: Allocation of virtual machines in cloud data centers-a survey of problem models and optimization algorithms. CSUR 48, 11 (2015)
Marler, R.T., Arora, J.S.: Survey of multi-objective optimization methods for engineering. Struct. Multidisc. Optim. 26, 369–395 (2004)
Masson, R., Vidal, T., Michallet, J., Penna, P.H.V., Petrucci, V., Subramanian, A., Dubedout, H.: An iterated local search heuristic for multi-capacity bin packing and machine reassignment problems. Expert Syst. Appl. 40, 5266–5275 (2013)
Mavrotas, G.: Effective implementation of the \(\varepsilon \)-constraint method in multi-objective mathematical programming problems. Appl. Math. Comput. 213, 455–465 (2009)
Mehta, D., O’Sullivan, B., Simonis, H.: Comparing solution methods for the machine reassignment problem. In: CP, pp. 782–797 (2011)
Mills, K., Filliben, J., Dabrowski, C.: Comparing VM-placement algorithms for on-demand clouds. In: CloudCom, pp. 91–98 (2011)
Panigrahy, R., Talwar, K., Uyeda, L., Wieder, U.: Heuristics for vector bin packing. research. microsoft. com (2011)
Portal, Gabriel M., Ritt, Marcus, Borba, Leonardo M., Buriol, Luciana S.: Simulated annealing for the machine reassignment problem. Ann. Oper. Res. 242, 1–22 (2012)
Purshouse, R.C., Fleming, P.J.: On the evolutionary optimization of many conflicting objectives. TEVC 11, 770–784 (2007)
Saber, T.: Multi-objective Virtual Machine Reassignment for Large Data Centres. PhD thesis, University College Dublin (2017)
Saber, T., Ventresque, A., Gandibleux, X., Murphy, L.: Genepi: A multi-objective machine reassignment algorithm for data centres. In: HM, pp. 115–129 (2014a)
Saber, T., Ventresque, A., Murphy, L., Talbi, E.: Multi-objective VM reassignment for the enterprise. In: Meta (2014b)
Saber, T., Ventresque, A., Brandic, I., Thorburn, J., Murphy, L.: Towards a multi-objective VM reassignment for large decentralised data centres. In: UCC (2015a)
Saber, T., Ventresque, A., Marques-Silva, J., Thorburn, J., Murphy, L.: Milp for the multi-objective VM reassignment problem. In: ICTAI, pp. 41–48 (2015b)
Saber, T., Marques-Silva, J., Thorburn, J., Ventresque, A.: Exact and hybrid solutions for the multi-objective VM reassignment problem. IJAIT 26, 1760004 (2017)
Saber, T., Florian, D., Mike, P., Michael, O., Ventresque, A.: A hybrid algorithm for multi-objective test case selection. In: CEC (2018a)
Saber, T., Thorburn, J., Murphy, L., Ventresque, A.: VM reassignment in hybrid clouds for large decentralised companies: a multi-objective challenge. Future Gener. Comput. Syst. 79, 751–764 (2018b)
Schroeder, B., Gibson, G.A.: A large-scale study of failures in high-performance computing systems. TDSC 7, 337–351 (2010)
Shachnai, H., Tamir, T.: Approximation schemes for generalized two-dimensional vector packing with application to data placement. JDA 10, 35–48 (2012)
Stillwell, M., Vivien, F., Casanova, H.: Virtual machine resource allocation for service hosting on heterogeneous distributed platforms. In: IPDPS, pp. 786–797 (2012)
Turky, A., Sabar, N.R., Sattar, A., Song, A.: Parallel late acceptance hill-climbing algorithm for the google machine reassignment problem. In: Australasian Joint Conference on Artificial Intelligence, pp. 163–174 (2016)
Turky, A., Sabar, N.R., Sattar, A., Song, A.: Evolutionary learning based iterated local search for google machine reassignment problems. In: Asia-Pacific Conference on Simulated Evolution and Learning, pp. 409–421 (2017a)
Turky, A., Sabar, N.R., Song, A.: Neighbourhood analysis: a case study on google machine reassignment problem. In: Australasian Conference on Artificial Life and Computational Intelligence, pp. 228–237 (2017b)
Turky, A., Sabar, N.R., Song, A.: Cooperative evolutionary heterogeneous simulated annealing algorithm for google machine reassignment problem. GPEM 19, 183–210 (2018)
Vargha, A., Delaney, H.D.: A critique and improvement of the cl common language effect size statistics of mcgraw and wong. J. Educ. Behav. Stat. 25, 101–132 (2000)
Visée, M., Teghem, J., Pirlot, M., Ulungu, E.L.: Two-phases method and branch and bound procedures to solve the bi-objective knapsack problem. J. Global Optim. 12, 139–155 (1998)
Voorsluys, W., Broberg, J., Venugopal, S., Buyya, R.: Cost of virtual machine live migration in clouds: a performance evaluation. In: CloudCom, pp. 254–265 (2009)
Xu, J., Fortes, J.: A multi-objective approach to virtual machine management in datacenters. In: ICAC, pp. 225–234 (2011)
Xu, J., Fortes, J.A.B.: Multi-objective virtual machine placement in virtualized data center environments. In: GreenCom, pp. 179–188 (2010)
Zitzler, E., Thiele, L.: Multiobjective optimization using evolutionary algorithms: a comparative case study. In: PPSN, pp. 292–301 (1998)
Zitzler, E., Laumanns, M., Thiele, L., Fonseca, C.M., da Fonseca, V.G.: Why quality assessment of multiobjective optimizers is difficult. In: GECCO, pp. 666–673 (2002)
Zitzler, E., Thiele, L., Laumanns, M., Fonseca, C.M., Da Fonseca, V.G.: Performance assessment of multiobjective optimizers: an analysis and review. TEVC 7, 117–132 (2003)
Acknowledgements
This work was supported, in part, by Science Foundation Ireland (SFI) grant 13/IA/1850 and grants 10/CE/I1855 and 13/RC/2094 to Lero-the Irish Software Research Centre (www.lero.ie).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Saber, T., Gandibleux, X., O’Neill, M. et al. A comparative study of multi-objective machine reassignment algorithms for data centres. J Heuristics 26, 119–150 (2020). https://doi.org/10.1007/s10732-019-09427-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10732-019-09427-8