RDF Data Query Optimization Based On Semantic Cache

RDF, which is short for “Resource Description Framework”, is the core of semanticWeb framework. It is used for describing resources and integrating the different tuples.With RDF, tuples could keep semantic information during transformation and the internetresources are processed automatically. With the increasingly application of semantic web,it becomes more and more important to manage the large-scale RDF data.At present, there are three management schemes with their own strengths andweaknesses: relative database based, triple based and graph based. Faced with thelarge-scale data, this management system has the low query efficiency and bad scalability.By analyzing the current technology of RDF data management, a query optimizationmethod, which is based on relation-algebra semantic cache and predicate-partitionsemantic cache, is proposed.In this method, two level semantic caches are used. The first is based onrelation-algebra semantic cache. In this level, the query efficiency is improved by reusingthe result of cached sub-tree. The second is based on predicate-partition semantic cache.In this level, the query efficiency is improved by fetch the triples of cached predicates.The part query which does not hit the cache is sent to the MonetDB for integrity of queryresults.With the experimental confirmation and result analysis, the method proposed iseffective on improving the query efficiency of RDF data.