Log-Based Reliability Analysis of Software as a Service (SaaS)

Abstract

Software as a Service (SaaS) has gained momentum in the past few years and businesses have been increasingly moving to SaaS model for their IT solutions. SaaS is a newer and transformed model where software is delivered to customers as a service over the web. With the SaaS model, there is a need for service providers to ensure that the services are available and reliable for end users at all times, which introduces significant pressure on the service provider to ensure right test processes and methodologies to minimize any impact to the provisions in Service Level Agreements (SLA). There is lack of research on the unique approaches to reliability analysis of SaaS suites. In this paper, we expand traditional approaches to reliability analysis of traditional web servers and propose methods tailored towards assessing the workload and reliability of SaaS applications. In addition we show the importance of data filtration when assessing SaaS reliability from log files. Finally, we discuss the suitability of reliability measures with respect to their relevance in the context of SLAs.