An evaluation of approaches to federated query processing over linked data

The Web has evolved from a global information space of linked documents to a web of linked data. The Web of Data enables answering complex, structured queries that could not be answered by a single data source alone. While the current procedure to work with multiple, distributed linked data sources is to load the desired data into a single RDF store and process queries in a centralized way against the merged data set, such an approach may not always be practically feasible or desired. In this paper, we analyze alternative approaches to federated query processing over linked data and how different design alternatives affect the performance and practicality of query processing. To this end, we define a benchmark for federated query processing, comprising a selection of data sources in various domains and representative queries. Using the benchmark, we perform experiments with different federation alternatives and provide insights about their advantages and disadvantages.