Hierarchical Forecasting of Web Server Workload Using Sequential Monte Carlo Training
- 1 March 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 899-904
- https://doi.org/10.1109/ciss.2006.286594
Abstract
We propose a solution to the Web server load prediction problem based on a hierarchical framework with multiple time scales. This framework leads to adaptive procedures that provide both long-term (in days) and short-term (in minutes) predictions with simultaneous confidence bands which accommodate not only serial correlation but also heavy-tailedness, and non-stationarity of the data. The long-term load is modeled as a dynamic harmonic regression (DHR), the coefficients of which evolve according to a random walk, and are tracked using sequential Monte Carlo (SMC) algorithms; whereas, the short-term load is predicted using an autoregressive model, whose parameters are also estimated using SMC techniques. We evaluate our method using real world web workload data.Keywords
This publication has 9 references indexed in Scilit:
- Monte Carlo Bayesian Signal Processing for Wireless CommunicationsPublished by Springer Nature ,2006
- A Hierarchical Framework for Modeling and Forecasting Web Server WorkloadJournal of the American Statistical Association, 2005
- Wavelet-Based Sequential Monte Carlo Blind Receivers in Fading Channels With Unknown Channel StatisticsIEEE Transactions on Signal Processing, 2004
- Efficient particle filtering for jump markov systems. Application to time-varying autoregressionsIEEE Transactions on Signal Processing, 2003
- Predictive models for proactive network management: application to a production Web serverPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A Filter Bank Approach for Modeling and Forecasting Seasonal PatternsTechnometrics, 2002
- Dynamic resource management considering the real behavior of aggregate trafficIEEE Transactions on Multimedia, 2001
- Dynamic harmonic regressionJournal of Forecasting, 1999
- Internet Web servers: workload characterization and performance implicationsIEEE/ACM Transactions on Networking, 1997