Re: enq: TX - index contention

Hi,



>My initial test showed I requested 1 ITL and 4 blocks.  I got 4 blocks with

two ITL slots in them.



This behaviour is a bit more than I understand  under "default"; wouldn't be 
more correct to say, that  the allowed range was changed as well?.





> But anybody who looks at the block dump knows better.



I strong skeptic would argue that this is only other presentation of the 
metadata. The best test IMHO is simple to stress a block  with two parallel 
transactions and observe if the concurrency takes place.

If I recall it correctly there is a paper on this topic from Arup Nanda on 
dbazine (unfortunately not online currently) and for sure lot of other stuff 
else.



Regards



Jaromir D.B. Nemec

----- Original Message ----- 
From: "Powell, Mark D" <mark.powell@xxxxxxx>
To: <Les.Hollis@xxxxxx>; <oracle-l@xxxxxxxxxxxxx>
Sent: Tuesday, March 22, 2005 8:21 PM
Subject: RE: enq: TX - index contention



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

Other related posts: