Linear Road : benchmarking stream-based data management systems

Abstract

This thesis describes the design, implementation, and execution of the Linear Road benchmark for stream-based data management systems. The motivation for benchmarking and the selection of the benchmark application are described. Test harness implementation is discussed, as are experiences using the benchmark to evaluate the Aurora engine. Effects of this work on the evolution of the Aurora engine are also discussed. Streams consist of continuous feeds of data from external data sources such as sensor networks or other monitoring systems. Stream data management systems execute continuous and historical queries over these streams, producing query results in real-time. This benchmark provides a means of comparing the functionality and performance of stream-based data management systems relative to each other and to relational systems. The benchmark presented is motivated by the increasing prevalence of "variable tolling" on highway systems throughout the world. Variable tolling uses dynamically determined factors such as congestion levels and accident proximity to calculate tolls. Linear Road specifies a variable tolling system for a fictional urban area, including such features as accident detection and alerts, traffic congestion measurements, toll calculations, and ad hoc requests for travel time predictions and account balances. This benchmark has already been adopted in the Aurora [ACC⁺03] and STREAM [MWA⁺03] streaming data management systems.