TY - UNPB
T1 - AjaxTracker: A Tool for High Fidelity Characterization of Ajax Applications
AU - Lee, Myungjin
AU - Singh, Sumeet
AU - Kompella, Ramana Rao
PY - 2008/7
Y1 - 2008/7
N2 - Interactive web applications powered by new technologies
such as Asynchronous Javascript and XML (AJAX) have
undeniably altered the course of the Web. Despite their popularity,
the associated machinery for characterization, measurement,
and monitoring of these AJAX-based applications
has lagged behind significantly. Part of the problem is that
there exists no systematic way to generate AJAX workloads
and observe their network behavior today. This paper focuses
on addressing this issue by developing a tool, called
AjaxTracker, that automatically mimics a human interaction
with an AJAX-powered web site and collects associated
network traces. These traces can further be post-processed
to understand various characteristics of AJAX applications.
Given the completely automated nature of our tool, we can
study several popular AJAX Web services, such as mail and
maps applications from popular providers, both across time
as well as space (different vantage points). Our results using
a month-long study revealed that AJAX application signatures
exhibit remarkable similarity along several features
across days. Somewhat surprisingly, there seems to be a lot
of similarity across applications that belong to the same category
(such as maps, mail) but different providers (such as
Google, Yahoo).
AB - Interactive web applications powered by new technologies
such as Asynchronous Javascript and XML (AJAX) have
undeniably altered the course of the Web. Despite their popularity,
the associated machinery for characterization, measurement,
and monitoring of these AJAX-based applications
has lagged behind significantly. Part of the problem is that
there exists no systematic way to generate AJAX workloads
and observe their network behavior today. This paper focuses
on addressing this issue by developing a tool, called
AjaxTracker, that automatically mimics a human interaction
with an AJAX-powered web site and collects associated
network traces. These traces can further be post-processed
to understand various characteristics of AJAX applications.
Given the completely automated nature of our tool, we can
study several popular AJAX Web services, such as mail and
maps applications from popular providers, both across time
as well as space (different vantage points). Our results using
a month-long study revealed that AJAX application signatures
exhibit remarkable similarity along several features
across days. Somewhat surprisingly, there seems to be a lot
of similarity across applications that belong to the same category
(such as maps, mail) but different providers (such as
Google, Yahoo).
M3 - Working paper
T3 - Computer Science Technical Report
SP - 1
EP - 6
BT - AjaxTracker: A Tool for High Fidelity Characterization of Ajax Applications
PB - Purdue University
ER -