Mathematical Service Discovery: Architecture, Implementation and Performance

Abstract

Service discovery and matchmaking in a distributed environment has
been an active research issue since at least the mid 1990s.
Previous work on matchmaking has typically presented the problem
and service descriptions as free or structured (marked-up) text,
so that keyword searches, tree-matching or simple constraint
solving are sufficient to identify matches. In this paper, we
discuss the problem of matchmaking for mathematical services,
where the semantics play a critical role in determining the
applicability or otherwise of a service. A matchmaking
architecture supporting the use of match plug-ins is first
described, followed by the types of plug-ins that can be
supported. The matched services are ranked based on the score
obtained from each plug-in, with the user being able to decide
which plug-in is most significant in the context of their
particular application. We consider the effect of pre- and
post-conditions of mathematical service descriptions on matching,
and how and why to reduce queries into DNF and CNF before
matching. Application examples demonstrate in detail how the
matching process works for all four algorithms. Additionally, an
evaluation of the ontological mode is provided, regarding
performance of loading ontologies, query response time and the
overall scalability is conducted. The performance results are used
to demonstrate scalability issues in supporting ontology-based
discovery within a Web Services environment.