XQuery on SQL Hosts

This paper extends this relational XML processing stack and shows that an RDBMS can also serve as a highly efficient XQuery runtime environment and also sees that the XQuery compiler can make good use of SQL's OLAP functionality.

Abstract

This chapter describes an XQuery compiler that targets SQL-based relational database backends and thus extends the relational XML processing stack, which was already known to be capable of providing XML mass storage as well as efficient XPath support. The compilation procedure is largely based on a specific encoding of sequences that allows for the set-oriented evaluation of nested for-loops. Operations on this encoding receive excellent support from widely available online analytical processing (OLAP) extensions to the SQL: 1999 standard. The XQuery to SQL compiler offers a variety of interesting hooks for extension and optimization. The core of the compilation procedure trades XQuery's notions of variable scopes and nested iteration (FLWOR blocks) for equijoins. The resulting relational XQuery processor closely adheres to the language semantics. XQuery compiler can also make significant use of SQL's OLAP functionality.