I've read through the archives, and gotten a general sense that people
don't like ODBC for UniData.  Can someone clarify exactly why?  I'm
talking with someone who says his app doesn't scale, that once about a
dozen people connect at once, it dies.

And he asked about my UOJ-based web applications, which are amazingly
stable and, really, never crash.  (I'm pretty sure I'd run out of
licenses or run the web server out of memory before anything UOJ related
dies.  And now I'm going to have to dig out the code I used to test it
way back when...)

Since I think the connections for UOJ and ODBC both come in through the
RPC port, what is it that makes ODBC not scale well?  (Or, if I tried to
hold open persistent UOJ connections, would I experience the same thing?
Currently I get in and get out, and probably don't have a bunch of
concurrent UniSessions very often.)

As an aside... UniData is single threaded, correct?  I found several
posts discussing UniVerse, but nothing definitive on UniData.  Is that
part of the issue with ODBC?

Thanks for any insight,
-- 
Wendy Smoak
Application Systems Analyst, Sr.
ASU IA Information Resources Management 
-------
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please visit http://listserver.u2ug.org/

Reply via email to