Abstract. From a theoretical point of view, the Semantic Web is understood in terms of a stack with RDF being one of its layers. A Semantic Web application operates on the common data model expressed in RDF. Reality is a bit different, though. As legacy data has to be processed in order to realise the Semantic Web, a number of questions arise when one is after processing RDF graphs on the Semantic Web. This work addresses performance and scalability issues (PSI), viz. proposing a metric for virtual RDF graphs on the Semantic Web—in contrast to a local RDF repository, or distributed, but native RDF stores. 1 Motivation The Semantic Web is—slowly—starting to take-off; as it seems, this is mainly due to a certain pressure stemming from the Web 2.0 success stories. From a theoretical point of view the Semantic Web is understood in terms of a stack. The Resource Description Framework (RDF) [1] is one of the layers in this stack, representing the common data model of the Semantic Web...