Re: RAC 'gc blocks lost' in stress test

From: Martin Klier <usn_at_usn-it.de>
Date: Thu, 20 Mar 2008 15:12:52 +0100
Message-ID: <47E270E4.3030500@usn-it.de>


Hi,

Yong Huang schrieb:
> We're using Swingbench to stress test an 8-node 10.2.0.3 RAC running on RH
> Linux (2.6.9-55.ELsmp x86_64). The total non-idle active sessions is between
> 200 and 300 (select count(*) from gv$session where status = 'ACTIVE' and
> wait_class != 'Idle'). At this stress level, we start to see 'gc blocks lost'
> statistic going up, sometimes as fast as 10 in 2 hours.
> gv$cr_block_server.fail_results also goes up.
in load times, my 4-node-RAC (10gR2 64bit on Linux) has between 58 and 140 lost gc blocks. I never had any troble with that. But you can decrease that number by setting the parallel_execution_message_size to a value as large as your architecture supports it. Target: 65535.

Regards
Martin Klier

-- 
Usn's IT Blog for Linux, Oracle, Asterisk
www.usn-it.de

--
http://www.freelists.org/webpage/oracle-l
Received on Thu Mar 20 2008 - 09:12:52 CDT

Original text of this message