Abstract: With the growing popularity and application of XML related standards, large repositories of XML documents have emerged on the Web. It is necessary to store these documents into a database to make them manageable. Storage schemes have become an important research topic in the XML data management field. Mapping XML documents to relational tables and storing them into a traditional RDBMS will break down the tree structure of XML data and cause a decline in query efficiency. This paper presents a novel storage scheme, called XN-Store, for native XML databases. Based on the index structure, this scheme directly stores XML nodes as records into a paged file to build up the persistent document object model, thus retaining the original tree structure of XML data. XN-Store not only reduces the storage space overhead of XML documents, but also implements the fast serialization and access of XML nodes. As a general purpose native XML storage scheme, XN-Store supports the creation of various secondary indexes to improve the efficiency of XML query processing. Extensive experiments are conducted on XN-Store and several previous XML storage schemes using a range of real and synthetic datasets, comparing the storage space, storage time, serialization time and node access time. The experimental results show that XN-Store is a high performance storage scheme for native XML databases.