Tuesday, October 18, 2016

Automated Combinatorial Testing for Software (Week3)

Based on NIST website evidence suggests that nearly all software errors are triggered by the interaction of one to six parameters.This article posted by National Institute of Standards and Technology is saying that if faults in  systems can be caused by a combination of n or fewer parameters, then testing all n-way combinations of parameters can provide high confidence that nearly all faults have been discovered. NIST is creating methods and tools to generate tests for n-ways combinations of parameter values. They are using combinatorial testing algorithms for designing arrays and automated model checking. This work will have applications in high assurance software, security and safety. This research project is interesting because NIST is creating new ways of testing. In order to apply combinatorial testing it is required to find set of test inputs that covers all t-way combinations of parameter values and to match up each set of inputs with the expected output, this way of testing is traditional and difficult problems but the cool thing is that these problems can be solved with the new algorithms on currently available hardware. Some of the elements for this project/program are fault localization,  integration into the development process, improved covering array algorithms and distribution of interaction faults. I think Automated Combinatorial Testing is a very interesting process to solve problems especially when we use new algorithms to solve problems.

Citation 

https://www.nist.gov/programs-projects/automated-combinatorial-testing-software-acts

No comments:

Post a Comment