Combinatorial algorithms test sets (CATS): The ACM/EATCS platform for experimental research We describe CATS, a recently started project, endorsed by the ACM and the EATCS, to develop a Web-based collection of test sets. These test sets, each specialized for a specific problem (such as network flow or satisfiability), are maintained by volunteers (associate editors), using contributions from researchers (authors). The purpose is to facilitate experimental research by standardizing common benchmarks, providing a mechanism for their evolution, and making them easily accessible and usable; and to identify significant open questions in the design of good test sets and the evaluation of performance of existing algorithms. The test sets should also facilitate algorithm selection for applications by characterizing subproblems and the behavior of competitive algorithms on these subproblems and encourage the development of high-quality implementations of advanced algorithms and data structures.

This software is also peer reviewed by journal TOMS.