Re: Performance of ODBC vs. other access methods

That got me thinking that gee, if they dont gear
the system down then we would not have to be so focused on tuning our
database and could be taught instead how to write SQL procedures.

I guess no matter what speed my system is, what platform I am running or 
how smart the database engine is I always think about tuning the DB for 
performance. 
This is just "database indexing 101" thought process regardless of the 
database. 

Flashback......I can remember back when I first started using OPNQRYF how 
I could bring a good system to its knees just by not having the right 
indexes :-)

I understand your soapbox discussion on the i5 vs. p5 cost/perf disparity. 
 Over time the market will take care of the value disparity if there 
really is one. 

If you really think the p5 is a better value for your business, go ahead 
and see what it's like to move from an iSeries to a p5.  Won't catch me on 
that boat :-)

Regards,
Richard Schoen
RJS Software Systems Inc. 
"Providing Your....iNFORMATION NOW!"
Email: richard@xxxxxxxxxxxxxxx
Web Site: http://www.rjssoftware.com
Tel: (952) 898-3038
Fax: (952) 898-1781
Toll Free: (888) RJSSOFT
-------------------------------------------------------------------
message: 6
date: Wed, 6 Sep 2006 21:39:38 -0400
from: "Steve Richter" <stephenrichter@xxxxxxxxx>
subject: Re: Performance of ODBC vs. other access methods

On 9/6/06, richard@xxxxxxxxxxx <richard@xxxxxxxxxxx> wrote:

Steve, are you alluding that other databases are smart enough to create
their own permanent indexes and views based on what the DB thinks the
access path should be ?

no, Richard, I am referring to what is needed on our system. I
attended an IBM SQL seminar recently and the topics were all on
performance tuning. That got me thinking that gee, if they dont gear
the system down then we would not have to be so focused on tuning our
database and could be taught instead how to write SQL procedures.  The
other recent event that pertains to ODBC performance is IBM just about
doubled the performance, keeping the price the same, of the p5 in
August. Maybe DB2 does not have dynamic index creation and retention
because we share a common DB2 code base with the p5 and that system is
so fast and low priced that they dont need that feature.

-Steve