Evaluating Transport Protocols for Real-time Event Stream Processing Middleware and Applications

Real-time event stream processing (RT-ESP) applications must synchronize continuous data streams, e.g., infrared scans and video monitoring for survivor detection after a natural disaster, despite fluctuations in resource availability. Satisfying these needs of RT-ESP applications requires predictable QoS from the underlying publish/subscribe (pub/sub) middleware. If a transport protocol is not capable of meeting the QoS requirements within a dynamic environment, the middleware must be able to adapt the transport protocol at runtime to react to changing operating conditions. Realizing such adaptive RT-ESP pub/sub middleware requires a thorough understanding of how different transport protocols behave under different operating conditions. This paper makes three contributions to work on achieving that understanding. First, we define a composite evaluation metric that combines packet latency and reliability to evaluate transport protocol performance. Second, we use this metric to quantify the performance of various transport protocols integrated with the OMG’s Data Distribution Service (DDS) QoS-enabled pub/sub middleware standard. Third, we use our metric to pinpoint configurations involving sending rate, network loss, and number of receivers that exhibit the pros and cons of the protocols. Our results show that a NAK-based reliable multicast protocol and a lateral error correction protocol provide the best performance. Moreover, the reliable multicast protocol provides lower average latency with small number of receivers and small network loss, whereas the lateral error correction protocol provides greater consistency in overall performance and the flexibility to trade-off lower average latency with network bandwidth usage.