Able to start multiple reasoners: RacerManager can start multiple RACERs on the local machine.
Management of multiple reasoners: RacerManager can initialise(configure each RACER and preload required KBs) and manage multiple RACER instances.
Query dispatching: Queries are dispatched to different RACERs if necessary (e.g. if several queries arrive at the same time).
Iterative Query Answering: It allows clients to limit the number of answers they are interested to get for their query.
Support for queries under premise: The system can answer queries that include a premise, which are also called if-then queries.
Load Balancing: The load is distributed among the RACER instances.
Caching: Each query sent by clients and each answer to that query gained through reasoning can be cached by the system. In this case no communication with a RACER instance is necessary, if the question is already answered by the system. i.e. clients benefit from reasoned knowledge gained through queries posed by other clients.
If a query references a KB that is not loaded by any RACER instance managed, RacerManager can instruct an idle RACER instance to load the KB (referenced by the URL in the query) before answering the query. I.e. clients can reference any KB in their query.
RacerManager responds with a termination token if the client received all known results to his query.
