Empirical Study of Session-Based Workload and Reliability for Web Servers

Goševa-Popstojanova, Katerina; Mazimdar, Sunil; Singh, Ajay Deep

doi:10.1109/issre.2004.19

Cited by 27 publications

(22 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…17. The request-based reliability for NASA IV&V Web servers is in the range of 0.9500-0.9971, which is higher than the request-based reliability for the data sets considered in Goševa-Popstojanova et al (2004) which was in the range of 0.8648-0.9857. The improvements are due to fixing errors based on our earlier analysis and regular maintenance done by Web administrators.…”

Section: Request-based Reliabilitymentioning

confidence: 65%

“…Our analysis is based on data extracted from Web logs of eight public and three private Web servers. This paper extends our work presented in Goševa-Popstojanova et al (2004); it includes analysis of significantly larger data sets for six out of ten servers considered in Goševa-Popstojanova et al (2004), as well as a new data set from an additional Web server. Even more, we present more in-depth analysis for the metrics introduced in Goševa- Popstojanova et al (2004) which allows us to further explore the Web workload and reliability.…”

Section: Related Work and Our Contributionsmentioning

confidence: 69%

“…Note that the number of sessions per user depends on the intensity of the workload and tends to be larger for data sets that span longer time period. For example, the maximum number of sessions per user for NASA IV&V Web sites during 20 weeks considered in this paper is one order of magnitude higher than the maximum number of sessions per user for 3 weeks (Goševa-Popstojanova et al, 2004).…”

Section: Inter-session Characteristicsmentioning

confidence: 74%

“…Table 1 summarizes the raw data from the eleven access logs which span different time durations, from two weeks to seven months. It should be emphasized that the NASA IV&V data sets analyzed in this paper are for duration of 20 weeks, which is significantly longer period than the 3 weeks considered in Goševa-Popstojanova et al (2004). The eleven servers have different workloads which vary by four orders of magnitude.…”

Section: Data Extraction and Analysis Processmentioning

confidence: 99%

See 3 more Smart Citations

Empirical Characterization of Session–Based Workload and Reliability for Web Servers

et al. 2006

View full text Add to dashboard Cite

The growing availability of Internet access has led to significant increase in the use of World Wide Web. If we are to design dependable Web-based systems that deal effectively with the increasing number of clients and highly variable workload, it is important to be able to describe the Web workload and errors accurately. In this paper we focus on the detailed empirical analysis of the sessionbased workload and reliability based on the data extracted from actual Web logs of eleven Web servers. First, we introduce and rigourously analyze several intrasession and inter-session metrics that collectively describe Web workload in terms of user sessions. Then, we analyze Web error characteristics and estimate the requestbased and session-based reliability of Web servers. Finally, we identify the invariants of the Web workload and reliability that apply through all data sets considered. The results presented in this paper show that session-based workload and reliability are better indicators of the users perception of the Web quality than the requestbased metrics.

show abstract

Section: Request-based Reliabilitymentioning

confidence: 65%

Section: Related Work and Our Contributionsmentioning

confidence: 69%

Section: Inter-session Characteristicsmentioning

confidence: 74%

Section: Data Extraction and Analysis Processmentioning

confidence: 99%

See 2 more Smart Citations

Empirical Characterization of Session–Based Workload and Reliability for Web Servers

et al. 2006

View full text Add to dashboard Cite

show abstract

“…Our previous work was focused on detailed characterization of the Web workload in terms of user sessions based on data extracted from actual Web logs of ten Web servers [8]. In this paper we take a different approach -we analyze the performance of e-commerce applications using synthetically generated workload which allows us to run controlled experiments.…”

Section: Workload Descriptionmentioning

confidence: 99%