Re: Transparent HugePages causing crashes?

From: Niall Litchfield <niall.litchfield_at_gmail.com>
Date: Thu, 6 Jun 2013 17:21:19 +0100
Message-ID: <CABe10sav8Uq=YezGXDG4R-2DN_iz3exw9YWm8i8K8i5q_pVZRw_at_mail.gmail.com>

I think the fact that Kevin Closson also sees a significant performance hit for Greenplum workload suggests that it might be well worth taking care with this feature on heavy data processing (of whatever variety) workloads

On Thu, Jun 6, 2013 at 5:04 PM, Rich Jesse <rjoralist2_at_society.servebeer.com
> wrote:

> Karl writes:
>
> > have you seen this blog?
> >
> http://structureddata.org/2012/06/18/linux-6-transparent-huge-pages-and-hadoop-workloads/
>
> Now we're getting somewhere! But does Hadoop allocate memory like Oracle
> does for it's SGA (assuming PGA will not use THP)? In other words, would a
> statically-sized SGA be subject to the THP defrag?
>
> Greg's article is great, but I still question if disabling THP defrag would
> solve the issue, as he has no performance stats specific to it.
>
> At this point, I see enough of a risk that there's no way in hell I'll be
> using THP on a Production box, but now I just want to know.
>
> Thanks!
> Rich
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

-- 
Niall Litchfield
Oracle DBA
http://www.orawin.info


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

Received on Thu Jun 06 2013 - 18:21:19 CEST