Re: high log file sync wait
Date: Fri, 30 Nov 2012 10:03:21 +0800
Message-ID: <CAJ-jFMbKH95qQT2xFQevph5xxMqicAMmboYyBB54Ti5bd-yEag_at_mail.gmail.com>
Thanks Martin,
I agree that the time of 40 ms means the log file write operation is slow, but I am trying to figure out which stage is slower and how to fix it. As we can see that the log file parallel write wait is only 6.52 ms which is a reasonable value for I/O latency. thus I don't think it's a purely I/O problem.
Suya
2012/11/29 Martin Berger <martin.a.berger_at_gmail.com>
> Suya,
>
> I don't see any more details to investigate here.
> Log File Sync of 40ms is just 'slow' in current environments.
> This is pure physical write IO - so you must go any way to make your
> physical write IO faster - OR reduce the number of commits in your
> application.
>
> I don't see any other path to resolve this issue - maybe others do?
>
> Martin
>
>
> On Mon, Nov 26, 2012 at 7:36 AM, Suya <suya.huang_at_gmail.com> wrote:
>
>> Hi,
>> I have a four-node RAC system, the workload runs against the database
>> waits
>> on log file sync heavily, average wait time is 82.41 ms, account for 75%
>> DB
>> time. The system CPU resource is rich enough, average 70% idle time.
>>
>> similar testing was done previously, which has log file sync wait with
>> average 40 ms, and because the CPU resource shortage, this time, we added
>> two logical CPU to the system. but see that the log file sync wait event
>> was even worse.
>>
>> any ideas on what else I should check in the AWR report?
>>
>> Thanks!
>> Suya
>>
>>
>> --
>> http://www.freelists.org/webpage/oracle-l
>>
>
-- http://www.freelists.org/webpage/oracle-lReceived on Fri Nov 30 2012 - 03:03:21 CET