SRBench: A Streaming RDF/SPARQL Benchmark

We introduce {\it SRBench}, the first general-purpose \underline{bench}mark primarily designed for \underline{s}treaming \underline{R}DF/SPARQL engines, completely based on real-world datasets from the Linked Open Data cloud. With the increasing problem of too much streaming data but not enough knowledge, researchers have set out for solutions in which Semantic Web technologies are adapted and extended for the publishing, sharing, analysing and understanding of streaming data. To help researchers and users to compare streaming RDF/SPARQL (StrRS) engines in a standardised application scenario, we have designed SRBench, with which one can assess the abilities of a StrRS engine to cope with a broad range of use cases typically encountered in real-world scenarios. The datasets used in the benchmark have been carefully chosen, such that they represent a realistic and relevant usage of streaming data. The benchmark defines a consice set of queries that cover the major aspects of StrRS engines, ranging from simple pattern matching queries on a single streaming dataset to queries with complex reasoning tasks on multiple interlinked datasets. Since StrRS processing is a fresh topic and so the most systems in this area are still experimental, in this paper, we complement our benchmarking work with a functional evaluation on three currently leading StrRS engines, \sst, C-SPARQL and CQELS. The presented results are meant to give a first baseline and illustrate the state-of-the-art.