Federated Path Queries For Heterogeneous Databases

In recent years,with the era of big data,the amount of data has become larger.At the same time,the types and sources of data are also varied.In this paper,we propose a method to process heterogeneous database queries consisting of both relational databases and graph data at the same time,named federated path queries(FPQ).The main contributions of this article are two points.Firstly,we design and present the federated path queries.On the basis of the existing nested regular path queries,federated path queries extend the expressive ability by adding more operators.Secondly,by theory and proof,we conclude that the federated path queries are more expressive than regular path queries and SPARQL.In addition,we evaluate the complexity of the federated path queries by data complexity and combined complexity.On the basis of the regular query language,we implement the federation path queries in combination with the actual scene of traffic.We use conjunctive operation to integrate the query results of relational databases and the RDF graph.Moreover,we test the performance of the federated path queries based on the existing DBpedia data set and the LUBM dataset.We achieve the federated path queries using of conjunctive operation.Federated path queries can handle heterogeneous database queries and provide rich query services.Besides,federated path queries can also provide theoretical support for heterogeneous data queries in various fields,and provide a solution for the design of heterogeneous query processing languages.