Rapid Development of Application-Specific Network Performance Tests

Analyzing the performance of networks and messaging layers is important for diagnosing anomalous performance in parallel applications. However, general-purpose benchmarks rarely provide sufficient insight into any particular application’s behavior. What is needed is a facility for rapidly developing customized network performance tests that mimic an application’s use of the network but allow for easier experimentation to help determine performance bottlenecks. In this paper, we contrast four approaches to developing customized network performance tests: straight C, C with a helper library, Python with a helper library, and a domain-specific language. We show that while a special-purpose library can result in significant improvements in functionality without sacrificing language familiarity, the key to facilitating rapid development of network performances tests is to use a domain-specific language designed expressly for that purpose.