RDF-TDAA: Optimizing RDF indexing and querying with a trie based on Directly Addressable Arrays and a path-based strategy

Abstract

The rapid expansion of RDF knowledge graphs in scale and complexity poses significant challenges for optimizing storage efficiency and query performance, with existing solutions often limited by high storage costs or slow retrieval speeds. This study introduces RDF-TDAA, a novel RDF data management engine built on an advanced trie-based index that integrates Directly Addressable Arrays, Characteristic Sets, and integer sequence compression to achieve exceptional data compactness while maintaining high-speed query processing. RDF-TDAA also employs a unique path-based query planning approach, which constructs efficient execution plans based on the paths in query graphs, and integrates a worst-case optimal join algorithm to further streamline query processing. To validate our approach, we conducted extensive experiments using both synthetic and real-world datasets. The results demonstrate that RDF-TDAA surpasses leading RDF management systems in both storage efficiency and query speed. These findings underscore RDF-TDAA’s scalability and effectiveness as a robust solution for managing large-scale RDF knowledge graphs, with valuable implications for improving RDF data handling in both academic and practical applications. The code for RDF-TDAA is available at https://github.com/MKMaS-GUET/RDF-TDAA.