Performance comparison for multi-dimensional index structures of rdf data

The development of semantic web poses new challenges for RDF (resource description framework) data retrieval performance. How to efficiently manage RDF data has become an important research direction in the past decade. Multi-dimensional indexing structures are gaining more acceptance as a major solution. This paper proposes a solution to improve retrieval performance by utilizing M-tree indexing structure. Meanwhile, performance comparison and analysis are performed in initialization and retrieval with mainstream RDF index structures. Compared with other index structures, M-tree shows stable performance both in initialization and retrieval. Through experimental data, we confirmed the possibility of improving the performance by minimizing the index space overlap and minimizing the node radius.