Context-Aware and Reinforcement Learning-Based Load Balancing System for Green Clouds

  • Ionut AnghelEmail author
  • Tudor Cioara
  • Ioan Salomie
Part of the Computer Communications and Networks book series (CCN)


This chapter describes a context-aware adaptive load balancing system capable of dynamically taking adaptation decisions to scale up/down Data Centre (DC) resources aiming at decreasing its energy consumption. The decision process is based on a reinforcement leaning technique which starts from the current DC state (Cloud Snapshot) and builds a learning tree by simulating the execution of load balancing actions with the goal of reducing the load fragmentation on the DC servers. The Cloud Snapshots are constructed by collecting DC context data related to workload distribution on the servers, computing resources usage and associated energy consumption values. Context and energy awareness is enacted by detecting those snapshots that are inefficient in terms of workload distribution and power/energy usage by using state of the art metrics and indicators. As a proof of concept implementation we present the Green Cloud Scheduler plug-in which augments OpenNebula Middleware with energy-awareness features.


Reinforcement Learning Physical Server Context Data Reinforcement Learning Algorithm Workload Distribution 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Abowd, G.D., Dey, A.K., Brown, P.J., et al.: Towards a Better Understanding of Context and Context-Awareness. In: Proceedings of the 1st International Symposium on Handheld and Ubiquitous Computing, pp. 304–307 (2000).
  2. 2.
  3. 3.
    Industry Outlook: Data Center Energy Efficiency (2014).
  4. 4.
  5. 5.
    Ardito, L.: Green IT—Available data and guidelines for reducing energy consumption in IT systems. Sustainable Comput. 4(1), 24–32 (2013)Google Scholar
  6. 6.
    OpenNebula middleware.
  7. 7.
    Green Cloud Scheduller OpenNebula component.
  8. 8.
    Murphy, A.: Virtualization defined: eight different ways. White paper (2007).
  9. 9.
    Brasol, S.M.: Analysis of Advantages and Disadvantages to Server Virtualization. Master Thesis (2009)Google Scholar
  10. 10.
    Sharif, M.I., Lee, W., Cui, W., Lanzi, A.: Secure In-VM monitoring using hardware virtualization. In: Proceedings of the 16th ACM Conference on Computer and Communications Security, pp. 477–487 (2009).
  11. 11.
    Reducing Data Center Energy Consumption, Intel Whitepaper (2010).
  12. 12.
    Talaber, R., Brey, T., Lamers, L.: Using Virtualization to Improve Data Center Efficiency, Green Grid White paper (2009).
  13. 13.
    Niles, S., Donovan, P.: Virtualization and Cloud Computing: Optimized Power, Cooling, and Management Maximizes Benefits, White Paper Published by APC Schneider Electric (2012).
  14. 14.
    Srikantaiah, S., Kansal, A., Zhao, F.: Energy Aware Consolidation for Cloud Computing, Microsoft Research (2009)Google Scholar
  15. 15.
    Wolf, C.: The myths of virtual machine consolidation (2006).
  16. 16.
    Kansal, A., Zhao, F., Liu, J., Kothari, N., Bhattacharya, A.: Virtual Machine Power Metering and Provisioning, SOCC (2010)Google Scholar
  17. 17.
    Stoess, J., Lang, C., Bellosa, F.: Energy management for hypervisor-based virtual machines. In: USENIX Annual Technical Conference (2007)Google Scholar
  18. 18.
    Nathuji, R., England, P., Sharma, P., Singh, A.: Feedback Driven QoS-Aware Power Budgeting for Virtualized Servers, Microsoft Research (2010)Google Scholar
  19. 19.
    Verma, A., Dasgupta, G., Kumar Nayak, T., et al.: Server workload analysis for power mini-mization using consolidation. In: USENIX Annual Technical Conference (2009)Google Scholar
  20. 20.
    Zhu, Q., Zhu, J., Agrawal G.: Power-aware consolidation of scientific workflows in virtualized environments. High Performance Comput. Networking, Storage Anal. 1–12 (2010).
  21. 21.
    Uddin, M., Rahman, A.A.: Server consolidation: an approach to make data centers energy efficient. Green Int. J. Sci. Eng. Res. 1(1), (2010).
  22. 22.
    Borgetto, D., Stolf, P., Da Costa, G., Pierson, J.M.: Energy aware autonomic manager. In: 1st International Conference on Energy-Efficient Computing and Net-working (2010)Google Scholar
  23. 23.
    Torres, J., Carrera, D., Beltran, V.: Tailoring resources: the energy efficient consolidation strategy goes beyond virtualization. In: International Conference on Autonomic Computing, pp. 197–198 (2008).
  24. 24.
    Jerger, N.E., Vantrease, D., Lipasti, M.: An Evaluation of server consolidation workloads for multi-core designs. In: Proceedings of the 2007 IEEE 10th International Symposium on Workload Characterization, pp. 47–56 (2007).
  25. 25.
    Patel, C., Sharma, R., Bash, C., Graupner, S.: Energy aware grid: global workload placement based on energy efficiency. In: International Mechanical Engineering Congress and Exposition (2003).
  26. 26.
    Kusic, D., Kephart, J.O., Hanson, J.E., Kandasamy, N.: Power and per-formance management of virtualized computing environments via lookahead control. In: Proceedings of the 2008 International Conference on Autonomic Computing (2008).
  27. 27.
    Feller, E., Rillingy, L., Morin, C.: Energy-aware ant colony based workload placement in clouds. In: Proceedings of the IEEE/ACM 12th International Conference on Grid Computing, pp. 26–33 (2011).
  28. 28.
    Sharifi, M., Salimi, H., Najafzadeh, M.: Power-efficient distributed scheduling of virtual machines using workload-aware consolidation techniques. J. Supercomputing (2011).
  29. 29.
    Nagios, the Industry Standard in IT Infrastructure Monitoring.
  30. 30.
    The Green Grid Data Center Power Efficiency Metrics: PUE and DCiE, Green Grid White Paper (2007).
  31. 31.
    Stanley, J.R., Brill, K.G., Koomey, J.G.: Four Metrics Define Data Center Greenness, Uptime Institute Whitepaper (2007).
  32. 32.
    Dayan, P., Watkins, C.: Reinforcement learning. Encycl. Cogn. Sci. (1999).
  33. 33.
  34. 34.
    CentOS Overview.
  35. 35.
    Kernel Based Virtual Machine.
  36. 36.
  37. 37.

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  1. 1.Technical University of Cluj-NapocaCluj-NapocaRomania

Personalised recommendations