TAC – Template Aware Coverage

TAC is a utility over coverage data. It combines efficient storage of coverage data per test template and data analytics techniques to extract useful information from this data.

The basis of TAC is the coverage matrix that maintains first order statistics about the coverage of each event by each test template. The statistics include the probability of hitting the event with a test case generated from the test template and the average number of times the event is hit by such test case.

With this coverage matrix, the descriptive information on how well a test template covers its target verification goal is obtained. The predications on the coverage rate of a feature come from temporal changes to the coverage events related to the feature. The optimized scheduling policy is a prescriptive analysis, providing the user with the best policy per coverage goals.


Anatoly Koyfman, Verification & Quality Analytics, IBM Research - Haifa