Schema-Independent Querying and Manipulation for Heterogeneous Collections in NoSQL Document Stores

摘要

NoSQL document stores offer native support to efficiently store documents with different schema within a same collection. However, this flexibility made it difficult and complex to formulate queries or to manipulate collections with multiple schemas. Hence, the user has to build complex queries or to reformulate existing ones whenever new schemas appear in the collection. In this paper, we propose a novel approach, grounded on formal foundations, for enabling schema-independent queries for querying and maintaining multi-structured documents. We introduce a query reformulation mechanism which consults a pre-constructed dictionary. This dictionary binds each possible path in the documents to all its corresponding absolute paths in all the documents. We automate the process of query reformulation via a set of rules that reformulate most document store operators, such as select, project and aggregate. In addition, we automate the process of reformulating the classical manipulation operators (insert, delete and update queries) in order to update the dictionary according to the different structural changes made in the collection. These two processes produce queries which are compatible with the native query engine of the underlying document store. To evaluate our approach, we conduct experiments on synthetic datasets. Our results show that the induced overhead when querying or updating can be acceptable when compared to the efforts made to restructure the data and the time required to execute several queries corresponding to the different schemas inside the collection.