A statistical data set comprises a collection of observations made at some points across a logical space and is often organized as what is called a data cube. The proper definition of the data cubes, especially of their dimensions, helps to process the observations and, more importantly, helps to combine observations from different data cubes. In this context, the Linked Data Principles can be profitably applied to the definition of data cubes, in the sense that the principles offer a strategy to provide the missing semantics of the dimensions, including their values. This work introduces a mediation architecture to help consume linked data cubes, exposed as RDF triples, but stored in relational databases. The data cubes are described in a catalogue using standardized vocabularies and are accessed by HTTP methods using REST principles. Therefore, this work aims at taking advantage of both Linked Data and REST principles in order to describe and consume linked data cubes in a simple but efficient way.