Citation:

Lin, X., M. Simon, AND N. Niu. Scientific Software Testing Goes Serverless: Creating and Invoking Metamorphic Functions. IEEE Software. IEEE Computer Society, Washington, DC, 38(1):61-67, (2021). https://doi.org/10.1109/MS.2020.3029468

Impact/Purpose:

The purpose of this manuscript is to introduce the tools of metamorphic testing and Function as a Service (FaaS) developers of scientific software.

Description:

We developed a Function-as-a-Service (FaaS) framework to transform end users’ questions into automatic tests for scientific software. Our case study of applying the FaaS framework to the U.S. EPA’s Storm Water Management Model (SWMM) demonstrated the value of detecting real-world defects in the software and revealed performance-cost tradeoffs. Our work illustrates FaaSification of scientific software testing and the importance of value-based evaluations.

URLs/Downloads:

Record Details: