RE: number of transactions per minute/hour

Admittedly without much homework (that I can remember), I've defined a transaction as:

 SELECT SUM(VALUE)
 FROM sys.v_$sysstat
 WHERE NAME IN ('user commits','user rollbacks');

...which of course is only valuable when the delta of two results is measured over time. But I would presume that this would have to take into account all transactions, including those containing "recursive SQL". What is the purpose of the data collection where recursive SQL bothers you? If it's to compare the results of business processes, the overhead of DD calls is valid, isn't it? Just curious...

Rich

-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Christian Antognini Sent: Friday, September 15, 2006 2:31 AM To: exriscer_at_gmail.com
Cc: oracle-l_at_freelists.org; ax.mount_at_gmail.com Subject: RE: number of transactions per minute/hour

Hi

> Not sure but I think it is very tedious to get those figures.... I
> always wondered how benchmark tools gives you informations such as
> transaction per minute, I think those tools dont use v$sysstat for
> these purposes

The reason is quite simple. DML statements or queries are *not* transactions! For that reason Anjo asked the OP to give *his* definition of "USER transaction"... For a regular definition have a look here: http://en.wikipedia.org/wiki/Database_transaction.

Regards,
Chris

--
http://www.freelists.org/webpage/oracle-l
--
http://www.freelists.org/webpage/oracle-l