The chapter proves that the web user’s dynamic behavior is a crucial issue that must be addressed in web performance studies in order to accurately estimate system performance indexes. To this end, in a first step we analyze and measure the effect of considering different levels of dynamic workload on web performance evaluation, instead of traditional (static) workloads. The more realistic workloads show that processor utilization is not uniformly balanced along time, but overloaded peaks rise when considering user’s dynamic behavior. As a consequence, the probability of a long response time is higher, and the number of user abandonments can increase (up to 40% according to technical studies). Then, in a second step we analyze and measure the effect of considering the User-Browser Interaction (UBI) as a part of user’s dynamic behavior on web workload characterization in performance studies. To this end, we evaluate a typical e-commerce scenario and compare the obtained results for different behaviors that take the user’s interaction with browser interface facilities into account, such as the use of the back button and parallel browsing originated by browser tabs or opening new windows when surfing a website. Experimental results show that these interaction patterns allow users to achieve their navigation objectives sooner than when browsing in a sequential way, so increasing their productivity up to 200% when surfing the website. In addition, results prove that when these types of behavior are taken into account, performance indexes can widely differ and relax the stress borderline of the server. For instance, the server utilization drops as much as 45% due to parallel browsing behavior, permitting the system either to devote more resources to other applications or to serve more users.
[1]
Ana Pont,et al.
Dweb model: Representing Web 2.0 dynamism
,
2009,
Comput. Commun..
[2]
Ana Pont,et al.
The impact of user-browser interaction on web performance
,
2013,
SAC '13.
[3]
Rolph E. Anderson,et al.
Customer loyalty in e-commerce: an exploration of its antecedents and consequences
,
2002
.
[4]
Saul Greenberg,et al.
Issues of Page Representation and Organisation in Web Browser's Revisitation Tools
,
2000,
Australas. J. Inf. Syst..
[5]
Ana Pont,et al.
The Impact of User's Dynamic Behavior on Web Performance
,
2012,
2012 IEEE 11th International Symposium on Network Computing and Applications.
[6]
F. Reichheld,et al.
E-LOYALTY: YOUR SECRET WEAPON ON THE WEB
,
2003
.
[7]
Tao Wang,et al.
Workload-aware anomaly detection for Web applications
,
2014,
J. Syst. Softw..
[8]
Ana Pont,et al.
Analyzing web server performance under dynamic user workloads
,
2013,
Comput. Commun..
[9]
Chang Liu,et al.
Web sites of the Fortune 500 companies: Facing customers through home pages
,
1997,
Inf. Manag..
[10]
Ana Pont,et al.
Surfing the Web Using Browser Interface Facilities: A Performance Evaluation Approach
,
2015,
J. Web Eng..
[11]
Andy Cockburn,et al.
Pushing back: evaluating a new behaviour for the back and forward buttons in web browsers
,
2002
.