A RDF data compress model based on octree structure

With the advent of the big data, large size of RDF datasets was flooded the Web. These data supported and managed through SPARQL endpoints, and it plays a role as a semantic net nodes based on the Linked Data Projects methods. The size of these RDF datasets was so huge that when systems need to process these datasets, their indexes cannot be fully loaded in main memory. For the reason given above, when system manages these huge datasets, the SPARQL queries can only perform through the slow disk accesses, which causes serious query performance problems. This paper proposed a method to solve this problem, offered a compact indexed RDF structure called oct-triples, which based on the octree structure, to store the indexes saved as three-dimensional matrix, which mapped by the dictionary and RDF triples. And this method has clearly an advantage on the compressibility compared with traditional compression solutions.