This paper describes the performance evaluation of a storage-independent model used for SPARQL-to-SQL translation algorithms. Recently, several web ontology languages have been developed for realizing Semantic Web. In particular, query language and web ontology storages based on RDB have been developed for searching and managing web ontology data. Among the proposed query languages, SPARQL is the most representative. However, SPARQL cannot be used to search data in an RDB. To solve this problem, an SPARQL query must be translated to an SQL query (SQL is the RDB query language). Recently, several translation algorithms have been proposed for SPARQL-to-SQL translation. However, the disadvantage of these algorithms is that they are dependent on the storage structure. We have previously developed a model for storage-independent translation algorithms. In this study, we evaluate the performance of a storage-dependent model and a storage-independent model and discuss the obtained results.