188 Large Scale and Big Data
modications to Hadoop. Another approach to achieve this is to use algebraic optimi-
zation—rewriting queries using a new set of operators that can achieve the same goal.
In [36], such an approach has been presented. However, this approach has additional
advantages over the approach in HadoopRDF in terms of management of intermediate
results because it uses a nested data model as opposed to the classical relational model.
Also, HadoopRDF, like most systems, requires a preprocessing phase.
This book chapter will discuss state-of-art techniques that have been proposed for
RDF data processing, particularly graph pattern matching queries, on MapReduce. It
will also cover related techniques that have not yet been applied ...