Search Mailing List Archives
[protege-owl] Database and Owl Ontology
tredmond at stanford.edu
Sat Aug 15 23:59:45 PDT 2009
There are several reasons why people would want to put a very large
ontology into a database. For example,
1. a database can be loaded on a machine with limited memory. This
allows people to work with ontologies on machines with limited
capabilities when access to more capable machines is not possible.
2. an ontology in a database can be loaded very quickly allowing
people to access an ontology without having to (re)parse the entire
3. a server may need to open many ontologies open at the same time and
to be able to open large ontologies quickly based on demand from users
or clients. This is important for an ontology repository for example.
The main reason for not wanting to use a database backend is its
complexity and its much slower performance.
But it is also important to point out that the Manchester OWL api
(used by Protege 4) can hold very large ontologies in limited memory.
So for example, my 2GB old laptop is almost capable of loading
snomed. This may suggest that the most important reason for having a
database backend for the Manchester OWL API is the third case.
On Aug 15, 2009, at 7:48 PM, Sarah G wrote:
> Hi All,
> Could someone please explain why you should have/use a relational
> database as backend for an ontology? I am new to using owl and was
> wondering why or if I will need to have a database.
> Get free photo software from Windows Live Click here.
> protege-owl mailing list
> protege-owl at lists.stanford.edu
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the protege-owl