Optimizing Egalitarian Performance in the Side-Effects Model of Colocation for Data Center Resource Management

  • Fanny Pascual
  • Krzysztof Rzadca
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10417)


In data centers, up to dozens of tasks are colocated on a single physical machine. Machines are used more efficiently, but tasks’ performance deteriorates, as colocated tasks compete for shared resources. As tasks are heterogeneous, the resulting performance dependencies are complex. In our previous work [18] we proposed a new combinatorial optimization model that uses two parameters of a task—its size and its type—to characterize how a task influences the performance of other tasks allocated to the same machine.

In this paper, we study the egalitarian optimization goal: maximizing the worst-off performance. This problem generalizes the classic makespan minimization on multiple processors (\(P||C_{\max }\)). We prove that polynomially-solvable variants of \(P||C_{\max }\) are NP-hard and hard to approximate when the number of types is not constant. For a constant number of types, we propose a PTAS, a fast approximation algorithm, and a series of heuristics. We simulate the algorithms on instances derived from a trace of one of Google clusters. Algorithms aware of jobs’ types lead to better performance compared to algorithms solving \(P||C_{\max }\).

The notion of type enables us to model degeneration of performance caused by colocation using standard combinatorial optimization methods. Types add a layer of additional complexity. However, our results—approximation algorithms and good average-case performance—show that types can be handled efficiently.


Cloud computing Scheduling Heterogeneity Co-tenancy Complexity 



We thank Paweł Janus for his help in processing the Google cluster data. This research has been partly supported by a Polish National Science Center grant Sonata (UMO-2012/07/D/ST6/02440), and a Polonium grant (joint programme of the French Ministry of Foreign Affairs, the Ministry of Science and Higher Education and the Polish Ministry of Science and Higher Education).


  1. 1.
    Beaumont, O., Eyraud-Dubois, L., Thraves Caro, C., Rejeb, H.: Heterogeneous resource allocation under degree constraints. IEEE TPDS 24(5), 926–937 (2013)Google Scholar
  2. 2.
    Bobroff, N., Kochut, A., Beaty, K.: Dynamic placement of virtual machines for managing SLA violations. In: Proceedings of IM. IEEE (2007)Google Scholar
  3. 3.
    Bu, X., Rao, J., Xu, C.: Interference and locality-aware task scheduling for mapreduce applications in virtual clusters. In: Proceedings of HPDC. ACM (2013)Google Scholar
  4. 4.
    Chiang, R.C., Huang, H.H.: TRACON: interference-aware scheduling for data-intensive applications in virtualized environments. In: Proceedings of SC. ACM (2011)Google Scholar
  5. 5.
    Coffman Jr., E.G., Garey, M.R., Johnson, D.S.: Approximation algorithms for bin packing: a survey. In: Approximation Algorithms for NP-Hard Problems. PWS (1996)Google Scholar
  6. 6.
    Di, S., Kondo, D., Wang, C.: Optimization of composite cloud service processing with virtual machines. IEEE Trans. Comput. 64(6), 1755–1768 (2015)MathSciNetzbMATHGoogle Scholar
  7. 7.
    Garey, M.R., Johnson, D.S.: Computers and Intractability: A Guide to the Theory of NP-Completeness. W.H. Freeman & Co., New York (1979)zbMATHGoogle Scholar
  8. 8.
    Graham, R.L.: Bounds on multiprocessing timing anomalies. SIAP 17(2), 416–429 (1969)MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Hochbaum, D.S., Shmoys, D.B.: Using dual approximation algorithms for scheduling problems theoretical and practical results. JACM 34(1), 144–162 (1987)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Jersak, L.C., Ferreto, T.: Performance-aware server consolidation with adjustable interference levels. In: Proceedings of SAC (2016)Google Scholar
  11. 11.
    Jin, X., Zhang, F., Wang, L., Hu, S., Zhou, B., Liu, Z.: Joint optimization of operational cost and performance interference in cloud data centers. IEEE Trans. Cloud Comput. (2015). doi: 10.1109/TCC.2015.2449839
  12. 12.
    Kambadur, M., Moseley, T., Hank, R., Kim, M.A.: Measuring interference between live datacenter applications. In: Proceedings of SC. IEEE (2012)Google Scholar
  13. 13.
    Kim, S., Hwang, E., Yoo, T.K., Kim, J.S., Hwang, S., Choi, Y.R.: Platform and co-runner affinities for many-task applications in distributed computing platforms. In: Proceedings of CCGrid. IEEE CS (2015)Google Scholar
  14. 14.
    Koh, Y., Knauerhase, R., Brett, P., Bowman, M., Wen, Z., Pu, C.: An analysis of performance interference effects in virtual environments. In: Proceedings of ISPASS. IEEE (2007)Google Scholar
  15. 15.
    Koutsoupias, E., Papadimitriou, C.: Worst-case equilibria. In: Meinel, C., Tison, S. (eds.) STACS 1999. LNCS, vol. 1563, pp. 404–413. Springer, Heidelberg (1999). doi: 10.1007/3-540-49116-3_38 CrossRefGoogle Scholar
  16. 16.
    Kundu, S., Rangaswami, R., Dutta, K., Zhao, M.: Application performance modeling in a virtualized environment. In: HPCA. IEEE (2010)Google Scholar
  17. 17.
    Kundu, S., Rangaswami, R., Gulati, A., Zhao, M., Dutta, K.: Modeling virtualized applications using machine learning techniques. In: SIGPLAN Notices, vol. 47. ACM (2012)Google Scholar
  18. 18.
    Pascual, F., Rzadca, K.: Partition with side effects. In: Proceedings of HiPC 2015 (2015)Google Scholar
  19. 19.
    Pascual, F., Rzadca, K.: Optimizing egalitarian performance in the side-effects model of colocation for data center resource management. CoRR abs/1610.07339v3 (2017).
  20. 20.
    Pietri, I., Sakellariou, R.: Mapping virtual machines onto physical machines in cloud computing: a survey. CSUR 49(3), Article No. 49 (2016)Google Scholar
  21. 21.
    Podzimek, A., Bulej, L., Chen, L.Y., Binder, W., Tuma, P.: Analyzing the impact of CPU pinning and partial CPU loads on performance and energy efficiency. In: Proceedings of CCGrid (2015)Google Scholar
  22. 22.
    Reiss, C., Tumanov, A., Ganger, G.R., Katz, R.H., Kozuch, M.A.: Heterogeneity and dynamicity of clouds at scale: Google trace analysis. In: Proceedings of SoCC. ACM (2012)Google Scholar
  23. 23.
    Song, W., Xiao, Z., Chen, Q., Luo, H.: Adaptive resource provisioning for the cloud using online bin packing. IEEE ToC 63(11), 2647–2660 (2014)MathSciNetzbMATHGoogle Scholar
  24. 24.
    Stillwell, M., Vivien, F., Casanova, H.: Virtual machine resource allocation for service hosting on heterogeneous distributed platforms. In: Proceedings of IPDPS. IEEE (2012)Google Scholar
  25. 25.
    Tang, X., Li, Y., Ren, R., Cai, W.: On first fit bin packing for online cloud server allocation. In: Proceedings of IPDPS (2016)Google Scholar
  26. 26.
    Verma, A., Pedrosa, L., Korupolu, M., Oppenheimer, D., Tune, E., Wilkes, J.: Large-scale cluster management at Google with Borg. In: Proceedings of EuroSys. ACM (2015)Google Scholar
  27. 27.
    Xu, Y., Musgrave, Z., Noble, B., Bailey, M.: Bobtail: avoiding long tails in the cloud. In: Proceedings of NSDI (2013)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Sorbonne Universités, UPMC, LIP6, CNRS, UMR 7606ParisFrance
  2. 2.Institute of InformaticsUniversity of WarsawWarsawPoland

Personalised recommendations