On Fri, 23 Jan 2026 18:57:21 -0800
Ron <[email protected]> wrote:

> Quite a few actually *can* deal with this kind of data; it's been a
> fairly mature field for a decade or more already.

Try 60 years and MUMPS.

https://en.wikipedia.org/wiki/MUMPS

The company I work for makes the descendant of MUMPS. We're one of the
best in the world with unstructured, arbitrary data. We succeed where
Oracle and Microsoft and Google and most others fail. I jokingly refer
to us as the biggest name in health care you never heard of, but you
know our customers.

We have a family of classes for handling large amounts of data (> ~3.5
million characters per object) called stream objects. Common examples
of stream objects are x-ray imaging and PDF scans of documents. Stream
objects can be stored in global nodes but the typical use and best
practice is to store these data as files on the filesystem.

Why? Because storing them in the database fills the in-memory index
with unneeded bloat which slows down the process of data retrieval.

At this point you should re-read Mark's posts because they're spot on,
and Mark does a better job at explaining what databases are for and how
they work than I can. Mark is a database person. I'm "just" a sysadmin
for some of the best database people in the world :).

-- 
\m/ (--) \m/
_______________________________________________
Discuss mailing list
[email protected]
https://lists.blu.org/mailman/listinfo/discuss

Reply via email to