Benchmark Suite for Distributed Publish/Subscribe Systems

摘要

Building a distributed publish/subscribe infrastructure amounts to defining a service model (or interface) and providing an implementation for it. A typical distributed implementation is designed as a network of dispatcher components, each one implementing appropriate protocols and algorithms, that collectively realize the chosen service model. The service model should provide a value-added service for a wide variety of applications, while the implementation should gracefully scale up to handle an intense traffic of publications and subscriptions. The authors believe that the design of such service models and implementations must be guided by a systematic evaluation method, which in turn must be based on a carefully chosen benchmark suite. In this paper, they lay out a set of requirements for a benchmark suite for distributed publish/subscribe services, and outline its primary components. The ideas proposed in this paper are based on their experience in building and studying publish/subscribe infrastructures, and on existing evaluation frameworks.