Web Testing for Reliability Improvement - Advances in Computers - page 178

4.3.LimitationsandMotivationforUsage-BasedWebTesting............ 192
5.AHierarchicalApproachforLarge-ScaleWebTesting ................ 193
5.1.OverallFramework:AThree-TieredWebTestingStrategy ........... 193
5.2. Testing High-Level Component Usage with Musa Operational Profiles ..... 196
5.3.UnifiedMarkovModels(UMMs)forUsage-BasedTesting ........... 199
5.4.ConstructingandUsingUMMsforWebTesting ................. 203
5.5.InternalValidationviaCross-ReferenceCharacterization ............ 207
6. Web Workload Characterization and Reliability Analyses ............... 209
6.1. Defining Workload Measures for Reliability Assessment ............ 209
6.2.MeasurementResultsandWorkloadCharacterization .............. 210
6.3. Analyzing Operational Reliability . . ....................... 212
6.4. Evaluating Potential Reliability Improvement ................... 214
6.5. General Observations about Web Reliability and Test Effectiveness ....... 218
7.ConclusionsandPerspectives .............................. 219
Acknowledgements.................................... 221
References......................................... 221

1. Introduction

Web-based applications provide cross-platform universal access to web resources
for the massive user population. With the prevalence of the world wide web (WWW)
and people's reliance on it, testing and quality assurance for the web is becoming
increasingly important. To help us test web-based applications and improve their
reliability, we would like to adapt existing techniques that have been used effectively
to assure and improve quality and reliability for traditional software systems [16,38,
48] . As a prerequisite to successful adaptation, we must have a good understanding of
the differences between the web environment and the traditional software systems,
as well as a good understanding of the existing techniques, their applicability to
different domains, and their effectiveness in dealing with different problems.
Quality in software is generally associated with good user experiences commonly
characterized by the absence of observable problems and satisfaction of user expec-
tations, which can also be intimately related to some internal characteristics of the
software product and its development process [23,38,48] . A quantitative measure of
quality meaningful to both the users and the developers is product reliability , which
is defined as the probability of failure-free operations for a specific time period or
input set under a specific environment [27,34] . Some testing and quality assurance
techniques work directly to assure product reliability by detecting and correcting
problems that are likely to be experienced by target customers and users [30,33] ,
while others work indirectly to ensure some internal integrity for the product under
development or in operation [5,35] .