Abstract

More and more RDF data is exposed on the Web via SPARQL endpoints. With the recent SPARQL 1.1 standard, these datasets can be queried in novel and more powerful ways, e.g., complex analysis tasks involving grouping and aggregation, and even data from multiple SPARQL endpoints, can now be formulated in a single query. This enables Business Intelligence applications that access data from federated web sources and can combine it with local data. However, as both aggregate and federated queries have become available only recently, state-of-the-art systems lack sophisticated optimization techniques that facilitate efficient execution of such queries over large datasets. To overcome these shortcomings, we propose a set of query processing strategies and the associated Cost-based Optimizer for Distributed Aggregate queries (CoDA) for executing aggregate SPARQL queries over federations of SPARQL endpoints. Our comprehensive experiments show that CoDA significantly improves performance over current state-of-the-art systems.

Authors: Dilshod Ibragimov, Katja Hose, Torben Bach Pedersen, and Esteban Zimanyi

Queries

SSB defines 13 queries. They represent 4 "prototypical" queries with different selectivity factors. A brief description of the queries is given in a table below. We converted all 13 queries defined into SPARQL and used the SERVICE keyword to query endpoints for all triple patterns.

• SPARQL Queries with one SERVICE Endpoint: one SERVICE endpoint
• SPARQL Queries with two SERVICE Endpoint: two SERVICE endpoint
• SPARQL Queries with three SERVICE Endpoint: three SERVICE endpoint

Datasets

The data in SSB is generated as relational data. We used different scale factors (1 to 5 - 6M to 30M observations) to generated multiple datasets of different sizes. We translated the datasets into RDF using a vocabulary that strongly resembles the SSB tabular structure. For example, a lineorder tuple is represented as a starshaped set of triples where the subject (URI) is linked via a property (e.g., rdfh:lo_orderdate) to a an object (e.g., rdfh:lo_orderdate_19931201) which in turn can be subject of another star-shaped graph. Values such as quantity and discount are connected to lineorder entities as literals. A simplified schema of the RDF structure is illustrated in the figure below. Converted datasets contain 110,5M (scale factor 1) to 547,5M (scale factor 5) triples

• SSB Scale Factor 1: SSB RDF SF1 (300 MB)
• SSB Scale Factor 2: SSB RDF SF2 (600 MB)
• SSB Scale Factor 3: SSB RDF SF3 (900 MB)
• SSB Scale Factor 4: SSB RDF SF4 (1.2 GB)
• SSB Scale Factor 5: SSB RDF SF5 (1.5 GB)