Hi there,

I'm currently trying to import data from a product system using an
ODBC driver called AcuODBC.

My problem is that the performance is just killing me. It is currently
taking 35min to import 95,000 rows of data (there are 700 columns). It
equates to approx 1.8M cells of data per min.

I know it must be able to be done quicker, dows anyone have any tips?

Thanks very much in advance.
JPO

I am unfamiliar with the driver itself but is there anything you can do in
its setup to speed things up?
Are you seeing the perf decrease on the INSERT side of the equation?
Do you have fast disks?
Do you have indexes on the destination?
Recovery model of the destination?
Triggers on the destination?

--

Allan Mitchell (Microsoft SQL Server MVP)
MCSE,MCDBA
www.SQLDTS.com
I support PASS - the definitive, global community
for SQL Server professionals - http://www.sqlpass.org


"JPO" <justin.parry-okeden (AT) ge (DOT) com> wrote

Hi there Alan,

Thanks for your points, I went back and checked off everything you
mentioned just in case I missed something. I think I have done
everything I can to speed up the performance:-

- We have a very fast disk set
- LOTS of RAM
- Recovery model set to 'SIMPLE'
- select into/bulkcopy turned on
- There are no indexes on the destination tables
- There are no trigers on the destination tables

I can load data from another MS-SQL DB into the same table approx 15
times faster, and I can load data from a flat file (NOT using
Bulk-Copy) about 5 times faster.

Is it just ODBC or is there something I can do to improve the
performance?

Thanks in advance for any help you can offer.
JPO

Then it sounds like the driver. Have you got any of the advanced options
set on the ODBC drivers?

--
--

Allan Mitchell (Microsoft SQL Server MVP)
MCSE,MCDBA
www.SQLDTS.com
I support PASS - the definitive, global community
for SQL Server professionals - http://www.sqlpass.org

"JPO" <justin.parry-okeden (AT) ge (DOT) com> wrote