Re: Realtime SQL Trace like MS SQL Server

From: DA Morgan <damorgan_at_x.washington.edu>
Date: Mon, 21 Mar 2005 06:39:23 -0800
Message-ID: <1111415759.371118_at_yasure>


Thomas Hase wrote:

>>Mark Powell posted a good explanation of exactly this to a remarkably 
>>similar question on c.d.o.server in the last couple of days. 

>
> Is there a link to this topic?

www.google.com
click on groups

>>However, doing 
>>so regularly is a bad idea on a system 

>
> I want to use this feature in a test environment
> (show my first post)
> because realtime tracing increase the performance in all dbms.

On its face your statement is not valid. Tracing always adds a load. The question is rather one of tuning increasing performance and that is best accomplished after the fact ... not realtime.

Unless I am misunderstanding your intent.

>>that is doing any real work. I'm 
>>interested in what the business need is that you are attempting to address 
>>by monitoring all SQL in realtime, it would seem likely that there are other 
>>ways of doing it.

>
> I have to port an access2k-mssql2k (ODBC) app to access2k-ora9i
> (ODBC).
> The app is not 100% client server. There are some large browsingforms.
> If I open an nativ access table with 100.000DS MSSQL-ODBC is 10 time
> faster as ora (the ms driver creates automaticly stored procs for
> msssql). Its a known problem, that oracle odbc drivers are slow.

Also a known problem that MS Access is a dog. Almost any other front-end would be superior and far more compliant with Sarbanes-Oxley if that is a concern.

HTH

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu
(replace 'x' with 'u' to respond)
Received on Mon Mar 21 2005 - 15:39:23 CET

Original text of this message