Ontologies are receiving an increasing attention in database management, since they often provide richer and more accurate descriptions of the data wrt the original database schemas. These descriptions are especially useful in mapping applications, where one needs to translate data across different representation. We present GROM a tool conceived to handle schema-mappings among ontologies over relational databases. It allows data architects to define high-level mappings between a source and a target ontology, defined as a set of views over the original databases. GROM rewrites these mappings between the virtual descriptions in terms of executable mappings between the two physical databases.