Evaluating Library Configurations

Abstract

A wide range of libraries are available for a developer to choose from when building a software system, but once the library is chosen, the developer must determine which version of the library to use. Is there some characteristic that can identify the correct version of a library to use? Even if a library compiles correctly, there could be a better version of that library that will provide superior performance. In particular, the developer would prefer to avoid poor con figurations: that is sets of libraries that perform poorly, or not at all. This paper describes a methodology by which the sub-performing version of a library can be identifi ed from the behavior observed from diff erent con figuration of libraries. These are measured by time, static and dynamic analysis of the results of executing the test suite in a project. During the course of these runs, different configurations of the libraries are substituted in and the results collected to be analyzed. The results of this analysis shows that there is no quick way to identify a sub performing library. However this library can be determined through concentrated eff orts to collect and analyze time-based data.