Ontology-based tools in the service of hardware verification

By using an ontology-based approach as the underlying scheme of hardware verification tools, knowledge about the specific design under test (DUT) is kept separate from the tool's generic service engine. This separation enables tools to easily adapt to new DUTs. In this paper, we discuss how we successfully adopted an ontology-based approach to the development of hardware verification tools. We also present ClassMate, an ontology modeling platform developed and used in IBM that is especially suited to support hardware verification ontologies. The ontology-based approach has boosted the impact of IBM's hardware verification tools by facilitating reuse - enabling adaptation to a large number of complex DUTs including several generations of the same design, lowering maintenance costs per DUT, and driving the accumulation of deep domain knowledge.