PAXQuery: Parallel Analytical XML Processing

Jesús Camacho-Rodríguez (1) , Dario Colazzo (2) , Ioana Manolescu (3, 4) , Juan A. M. Naranjo (3, 4)

1 Hortonworks Inc.
2 LAMSADE - Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision
3 OAK - Database optimizations and architectures for complex large data
4 LRI - Laboratoire de Recherche en Informatique

XQuery is a general-purpose programming language for processing semi-structured data, and as such, it is very expressive. As a consequence , optimizing and parallelizing complex analytics XQuery queries is still an open, challenging problem. We demonstrate PAXQuery, a novel system that parallelizes the execution of XQuery queries over large collections of XML documents. PAXQuery compiles a rich subset of XQuery into plans expressed in the PArallelization ConTracts (PACT) programming model. Thanks to this translation, the resulting plans are optimized and executed in a massively parallel fashion by the Apache Flink system. The result is a scalable system capable of querying massive amounts of XML data very efficiently, as proved by the experimental results we outline.

Domaines: Base de données [cs.DB]

Jesús Camacho-Rodríguez, Dario Colazzo, Ioana Manolescu, Juan A. M. Naranjo. PAXQuery: Parallel Analytical XML Processing. ACM SIGMOD International Conference on Management of Data 2015, May 2015, Melbourne, Victoria, Australia. pp.1117-1122, ⟨10.1145/2723372.2735374⟩. ⟨hal-01178490⟩