home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.databases
- Path: sparky!uunet!psgrain!neon!pail!servio!marcs
- From: marcs@slc.com (Marc San Soucie)
- Subject: Re: 500'000 records - who does best?
- Message-ID: <1993Jan5.223945.4180@slc.com>
- Organization: Servio Corp, Beaverton Oregon, US
- References: <18774@mindlink.bc.ca> <1993Jan1.021012.24215@news.arc.nasa.gov> <1993Jan1.182624.2993@uunet!tellab5!odgate>
- Distribution: na
- Date: Tue, 5 Jan 1993 22:39:45 GMT
- Lines: 45
-
- Mike J. Kelly (mike@uunet!tellab5!odgate) writes:
-
- > Hugh LaMaster (lamaster@pioneer.arc.nasa.gov) writes:
-
- > > Mischa Sandberg (Mischa_Sandberg@mindlink.bc.ca) writes:
-
- > > With 30K records, you are dealing
- > >|> with blobs outside the normal record structure; you can run greater
- > >|> risks.
-
- > >Is there a "correct" way to deal with this problem yet? It can show
- > >up in surprisingly simple DBMS applications, such as where you want to
- > >search through a bunch of trouble calls for matches to a particular
- > >problem. 1-2 years ago, when I inquired to this newsgroup, there was
- > >no RDBMS that could handle text searches efficiently. (Efficiently,
- > >that is, compared to a text processing/retrieval system.) Is this still
- > >the case?
-
- > As far as I know, no one has yet integrated content-based retrieval (CBR)
- > with BLOBs and large text fields, which is what you really want. Remember,
- > though, that BLOBs are relatively new and I would expect that something like
- > this will start happening in the next year or so.
-
- BLOBs are not new to the Object-Oriented database community. An active
- OODB, which allows you to write methods that are stored and executed in the
- database engine, makes solution of this kind of problem quite routine. Such
- solutions that can be easily customized and tuned to match specific problem
- needs, and readily shared between applications.
-
-
- > It might be interesting to the vendors who are reading this group if we
- > were to start a thread on what a CBR-BLOB capability would look like. The
- > way I conceive it, at its simplest, it would involve a new type of index
- > ("create text index"?) which would set up a varchar or BLOB column to be
- > retrieved via CBR operators, along with some new CBR operators (at least
- > contains, and probably more sophisticated stuff than that.)
-
- Or use an OODB and save the need for a new syntax or a new set of
- operators. Most OODB's handle this kind of work with their built-in access
- mechanism.
-
- Marc San Soucie
- Servio Corporation
- Beaverton, Oregon
- marcs@slc.com
-