US-RSE Software Testing Talk Series
- –
The Testing working group is happy to announce a new talk on Software Testing on Wednesday December 17, 1-2 PM EDT.
Title: “Testing the (almost) untestable: some ideas for automated testing of scientific software that is big, slow or non-reproducible”
Presenter: Gaurav Vaidya
Abstract
Testing libraries and tools have made it easier to ensure that scientific software work correctly, but some scientific software is too big, slow or non-reproducible for simple testing. In this presentation, I will describe the testing and validation strategies and tools that I have developed to test and validate one such piece of software I currently maintain: the Babel pipelines that combine identifier cross-references from dozens of biomedical data sources to generate over 490M sets of equivalent identifiers for the NCATS Translator project. Please bring your own examples of testing difficult-to-test software and the strategies you’ve used to test them to share and discuss!
Biography
Gaurav Vaidya, PhD has been a scientific software developer since his undergraduate days at the National University of Singapore in the mid-2000s (and was introduced to automated software testing soon afterwards). He completed his PhD in Ecology and Evolutionary Biology at the University of Colorado Boulder in 2017, and has worked as a Semantic Web Technologist at the Renaissance Computing Institute (RENCI) at the University of North Carolina at Chapel Hill since 2019. He creates scientific software tools for the NCATS Translator project and the HEAL Data Stewardship Group, and tries his best to incorporate software testing into his work wherever he can.
Registration details
To register follow this link: Testing Talk Series Registration