| Abstract | Software components are increasingly assembled from othercomponents. Each component may further depend on oth-
ers, and each may have multiple active versions. The to-
tal number of configurations—combinations of components
and their versions—deployed by end users can be very large.
Component developers, therefore, spend considerable time
and effort doing compatibility testing – determining whether
their components can be built correctly for all deployed con-
figurations. In previous work we developed Rachet to sup-
port large-scale compatibility testing of components.
In this paper, we describe and evaluate methods to enable
Rachet to perform incremental compatibility testing. We de-
scribe algorithms to compute differences in component com-
patibilities between current and previous component builds,
a formal test adequacy criterion based on covering the differ-
ences, and cache-aware configuration sampling and testing
methods that attempt to reuse effort from previous testing
sessions. We evaluate our approach using the 5-year evo-
lution history of a scientific middleware component. Our
results show significant performance improvements over Ra-
chet’s previous retest-all approach, making the process of
compatibility testing practical for evolving components.
|
