Test coverage monitoring techniques are an integral part of modern methodologies for testing computer software. For instance, tools such as automated fault localizers, test adequacy calculators, and debuggers all use a coverage report for various purposes. Recently developed monitoring methods track the coverage of the nodes and edges in a program’s control flow graph, definition-use associations involving program variables, or interaction with the state and structure of a database. However, a coverage report often balloons in size as the monitor includes additional details about the behavior of the program, test suite, and other software components such as the operating system and database. The marked increase in coverage report size is particularly problematic when testing occurs in a resource constrained embedded environment or on a build/test server that collects coverage results for many programs over a long time period. Large coverage reports may also limit the efficiency and effectiveness of defect isolation methods that monitor a remote program and thus transmit coverage data across a network.
Ostrofsky, E., & Kapfhammer, G. M. (2009). An empirical comparison of methods for compressing test coverage reports. Poster Compendium of the 3rd International Symposium on Empirical Software Engineering and Measurement.
Want to cite this paper? Look in the BiBTeX file of gkapfham/research-bibliography for the key "Ostrofsky2009".