Migrating software testing to the cloud

Monday, September 13 image

scott
Scott Tilley,
Florida Institute of Technology,
USA

tauhida
Tauhida Parveen,
Florida Institute of Technology,
USA

Regression testing is often performed as part of the software maintenance process. The amount of tests cases for a large-scale system can range from several hundred to many thousands, requiring significant computing resources and lengthy execution times, often precluding their use in an interactive setting. Traditional approaches to reduce the execution time for regression testing typically focus on excluding selected tests from the suite that need to be run after a change is made to the system. Cloud computing offers an alternate solution to this problem: the use of virtualized hardware, effectively unlimited storage, and software services that can aid in reducing the execution time of large test suites in a cost-effective manner. However, migrating software testing to the cloud is not without cost, nor is it necessarily the best solution to all testing problems. This tutorial presents a decision framework called SMART-T for migrating software testing to the cloud. The framework is based on the “SOA Migration, Adoption, and Reuse Technique” (SMART) from Carnegie Mellon University’s Software Engineering Institute, which has been successfully used to migrate legacy components to a service-oriented environment. SMART-T helps organizations identify their current testing process, describe the requirements of the target cloud computing environment for performing software testing, and through a gap analysis of these two states map out the issues, effort, and potential benefits of migrating their software testing to the cloud. Several case studies are used to illustrate the use of the SMART-T framework.