Mobile Networks and Applications

, Volume 23, Issue 4, pp 840–853 | Cite as

A Queue Model for Reliable Forecasting of Future CPU Consumption

  • Hugo Lewi HammerEmail author
  • Anis Yazidi
  • Alfred Bratterud
  • Hårek Haugerud
  • Boning Feng


Statistical queuing models are popular to analyze a computer systems ability to process different types requests. A common strategy is to run stress tests by sending artificial requests to the system. The rate and sizes of the requests are varied to investigate the impact on the computer system. A challenge with such an approach is that we do not know if the artificial requests processes are realistic when the system is applied in a real setting. Motivated by this challenge, we develop a method to estimate the properties of the underlying request processes to the computer system when the system is used in a real setting. In particular we look at the problem of recovering the request patterns to a CPU processor. It turns out that this is a challenging statistical estimation problem since we do not observe the request process (rate and size of the requests) to the CPU directly, but only the average CPU usage in disjoint time intervals. In this paper we demonstrate that, quite astonishingly, we are able to recover the properties of the underlying request process (rate and sizes of the requests) by using specially constructed statistics of the observed CPU data and apply a recently developed statistical framework called Approximate Bayesian Computing. Further we apply the model to forecast future CPU consumption. Our results show that the model forecast future CPU consumption with less error than both the hidden Markov model (HMM) in (Hammer et al. 2016) and an ARIMA model. Another good property of the queue model is that we can forecast the instantaneous CPU consumption at any time point in the future, while the HMM in (Hammer et al. 2016) and time series models are limited to only forecasting the average CPU consumption in disjoint time intervals.


Approximate Bayesian computing CPU consumption Forecasting Queue process and time series 


  1. 1.
    Acharya SK (1999) On normal approximation for maximum likelihood estimation from single server queues. Queueing systems 31(3-4):207–216MathSciNetCrossRefzbMATHGoogle Scholar
  2. 2.
    Aggarwal V, Gautam N, Kumara SRT, Greaves M (2005) Stochastic fluid flow models for determining optimal switching thresholds. Perform Eval 59(1):19–46CrossRefGoogle Scholar
  3. 3.
    Benhammadi F, Gessoum Z, Mokhtari A et al. (2011) Cpu load prediction using neuro-fuzzy and bayesian inferences. Neurocomputing 74(10):1606–1616CrossRefGoogle Scholar
  4. 4.
    Blum MGB, François O (2010) Non-linear regression models for approximate bayesian computation. Stat Comput 20(1):63–73MathSciNetCrossRefGoogle Scholar
  5. 5.
    Blum MGB, Nunes MA, Prangle D, Sisson SA (2013) A comparative review of dimension reduction methods in approximate bayesian computation. Stat Sci 28(2):189–208MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Bouterse B, Perros H (2012) Scheduling cloud capacity for time-varying customer demand IEEE 1st International Conference on Cloud Networking (CLOUDNET), p 2012Google Scholar
  7. 7.
    Christodoulopoulos K, Gkamas V, Varvarigos EA (2008) Statistical analysis and modeling of jobs in a grid environment. Journal of Grid Computing 6(1):77–101CrossRefGoogle Scholar
  8. 8.
    Cox DR (1955) The statistical analysis of congestion. J R Stat Soc Ser A (General) 118(3):324–335MathSciNetCrossRefGoogle Scholar
  9. 9.
    Dinda PA (2002) Online prediction of the running time of tasks. Clust Comput 5(3):225–236CrossRefGoogle Scholar
  10. 10.
    Dinda PA (2000) Resource Signal Prediction and Its Application to Real-time Scheduling Advisors. PhD thesis, Pittsburgh, PA, USA. AAI9986588Google Scholar
  11. 11.
    Drovandi CC, Pettitt AN, Lee A, et al. (2015) Bayesian indirect inference using a parametric auxiliary model. Stat Sci 30(1):72– 95MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Fearnhead P (2004) Filtering recursions for calculating likelihoods for queues based on inter-departure time data. Stat Comput 14(3):261–266MathSciNetCrossRefGoogle Scholar
  13. 13.
    Feller W (1950) An introduction to probability theory and its applications. Vol. IGoogle Scholar
  14. 14.
    Forbes C, Evans M, Hastings N, Peacock B (2011) Statistical distributions. WileyGoogle Scholar
  15. 15.
    Guo L, Yan T, Zhao S, Jiang C (2014) Dynamic performance optimization for cloud computing using m/m/m queueing system. J Appl Math, 2014Google Scholar
  16. 16.
    Hammer HL, Yazidi A, Begnum K (2016) An Inhomogeneous Hidden Markov Model for Efficient Virtual Machine Placement in Cloud Computing Environments Journal of ForecastingGoogle Scholar
  17. 17.
    Hammer HL, Yazidi A, Bratterud A, Haugerud H, Feng B (2017) Recovering Request Patterns to a CPU Processor from Observed CPU Consumption Data Industrial Networks and Intelligent Systems. Springer, pp 14–28Google Scholar
  18. 18.
    Heggland Knut, Frigessi Arnoldo (2004) Estimating functions in indirect inference. J R Stat Soc Ser B (Statistical Methodology) 66(2):447–462MathSciNetCrossRefzbMATHGoogle Scholar
  19. 19.
    Hyndman RJ, Khandakar Y (2008) Automatic time series forecasting: the forecast package for r. J Stat Softw 26(3):1–22Google Scholar
  20. 20.
    Hyndman RJ (2010) Forecasting with long seasonal periods.
  21. 21.
    Sudha J (1992) Relative efficiency of a parameter for a M/G/1 queueing system based on reduced and full likelihood functions. Commun Stat Simul Comput 21(2):597–606CrossRefzbMATHGoogle Scholar
  22. 22.
    Kim J, Shin KG (1996) Execution time analysis of communicating tasks in distributed systems. IEEE Trans Comput 45(5):572–579CrossRefzbMATHGoogle Scholar
  23. 23.
    McFadden D (1989) A method of simulated moments for estimation of discrete response models without numerical integration. Econometrica: Journal of the Econometric Society, 995–1026Google Scholar
  24. 24.
    Mei J, Li K, Ouyang A, Li K (2015) A profit maximization scheme with guaranteed quality of service in cloud computing. IEEE Trans Comput 64(11):3064–3078MathSciNetCrossRefzbMATHGoogle Scholar
  25. 25.
    Muniz G, Golam Kibria BM (2009) On some ridge regression estimators: An empirical comparisons. Commun Stat Simul Comput 38(3):621–630MathSciNetCrossRefzbMATHGoogle Scholar
  26. 26.
    Core Team R (2015) R: a language and environment for statistical computing. r foundation for statistical computing. Vienna, AustriaGoogle Scholar
  27. 27.
    Seneviratne S, Levy DC (2011) Task profiling model for load profile prediction. Futur Gener Comput Syst 27(3):245–255CrossRefGoogle Scholar
  28. 28.
    Shen Z, Subbiah S, Xiaohui G, Wilkes J (2011) Cloudscale: elastic resource scaling for multi-tenant cloud systems Proceedings of the 2nd ACM Symposium on Cloud Computing. ACM, p 5Google Scholar
  29. 29.
    Sjöström M, Wold S, Lindberg W, Persson J, Martens H (1983) A multivariate calibration problem in analytical chemistry solved by partial least-squares models in latent variables. Anal Chim Acta 150:61–70CrossRefGoogle Scholar
  30. 30.
    Wolski R, Spring NT, Hayes J (1999) The network weather service: a distributed resource performance forecasting service for metacomputing. Futur Gener Comput Syst 15(5):757–768CrossRefGoogle Scholar
  31. 31.
    Xiong W, Altiok T (2008) Queueing analysis of a server node in transaction processing middleware systems. Comput Oper Res 35(8):2561–2578CrossRefzbMATHGoogle Scholar
  32. 32.
    Xu J, Fortes J (2011) A multi-objective approach to virtual machine management in datacenters Proceedings of the 8th ACM international conference on Autonomic computing. ACM, pp 225–234Google Scholar
  33. 33.
    Yang D, Cao J, Jiwen F, Wang J, Guo J (2013) A pattern fusion model for multi-step-ahead cpu load prediction. J Syst Softw 86(5):1257–1266CrossRefGoogle Scholar
  34. 34.
    Yang L, Foster Ian, Schopf JM (2003) Homeostatic and tendency-based cpu load predictions 2003 IEEE International Parallel and Distributed Processing Symposium. IEEE, p 9Google Scholar
  35. 35.
    Yuzukirmizi Mustafa, MacGregor Smith J (2008) Optimal buffer allocation in finite closed networks with multiple servers. Comput Oper Res 35(8):2579–2598CrossRefzbMATHGoogle Scholar
  36. 36.
    Zhang Y, Wei S, Inoguchi Y (2007) Cpu load predictions on the computational grid. IEICE TRANSACTIONS on Information and Systems 90(1):40–47CrossRefGoogle Scholar
  37. 37.
    Zhang Z, Weiguo F (2008) Web server load balancing: A queueing analysis. Eur J Oper Res 186(2):681–693MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2017

Authors and Affiliations

  1. 1.Department of Computer ScienceOslo and Akershus University College of Applied SciencesOsloNorway

Personalised recommendations