How will Your Workload Look Like in 6 Years? Analyzing Wikimedia´s Workload

Author

Eldin, Ahmed Ali ; Rezaie, Ali ; Mehta, A. ; Razroev, Stanislav ; SjoÌˆstedt-de Luna, Sara ; Seleznjev, Oleg ; Tordsson, Johan ; Elmroth, Erik

Author_Institution

Dept. of Comput. Sci., Umea Univ., Umeå, Sweden

fYear

2014

fDate

11-14 March 2014

Firstpage

349

Lastpage

354

Abstract

Accurate understanding of workloads is key to efficient cloud resource management as well as to the design of large-scale applications. We analyze and model the workload of Wikipedia, one of the world´s largest web sites. With descriptive statistics, time-series analysis, and polynomial splines, we study the trend and seasonality of the workload, its evolution over the years, and also investigate patterns in page popularity. Our results indicate that the workload is highly predictable with a strong seasonality. Our short term prediction algorithm is able to predict the workload with a Mean Absolute Percentage Error of around 2%.

Keywords

Web sites; cloud computing; mean square error methods; resource allocation; splines (mathematics); time series; Web sites; Wikimedia workload analysis; cloud resource management; large-scale applications; mean absolute percentage error; page popularity; polynomial splines; short term prediction algorithm; time-series analysis; Aggregates; Electronic publishing; Encyclopedias; Internet; Market research; Servers;

fLanguage

English

Publisher

ieee

Conference_Titel

Cloud Engineering (IC2E), 2014 IEEE International Conference on

Conference_Location

Boston, MA

Type

conf

DOI

10.1109/IC2E.2014.50

Filename

6903494