[Linux-PowerEdge] 15 or 20 package_power_limit events per second on an R320
J. Adam Craig
jacraig at vcu.edu
Tue Jul 2 13:03:25 CDT 2013
You are correct. I've not had the luxury of having the time or material
resources at my disposal to attempt to isolate whether every setting in the
chart is needed. I would presume that the setting to disable
virtualization technology could be left to "Enabled" if you needed it that
way, but again, we've not had occasion to test that in our environment at
the present time.
For what its worth, we also noticed a significant improvement in system
performance after making the changes as outlined on pg. 6 of the document.
A process that had taken several hours to complete was reduced to approx.
*J. Adam Craig*
UNIX Operating Systems Analyst
VCU Computer Center
"Don't be a phishing victim -- VCU and other reputable organizations will
never use email to request that you reply with your password, social
security number or confidential personal information. For more detauls,
On Tue, Jul 2, 2013 at 1:56 PM, Todd Lyons <tlyons at ivenue.com> wrote:
> On Tue, Jul 2, 2013 at 10:18 AM, J. Adam Craig <jacraig at vcu.edu> wrote:
> > Folks --
> > The only "fix" I've managed to find for this issue has been to apply all
> > BIOS changes listed on p. 6 of this guide for "low-latency." Apart from
> > these adjustements, I've found that these messages occur under moderate
> > heavy load on R620, R720, and R820 servers running RHEL 6.x.
> > I've opened multiple support tickets with Dell Support, and consistently
> > been advised to update the BIOS, change from "Performance Per Watt" to
> > "Performance," etc., but these changes (knock on wood) has so far been
> > only viable solution.
> So what you are saying is that in your experience, these do not solve
> the problem:
> 1. Selecting Custom and disabling C1E and CStates.
> 2. Selecting the "Performance" profile.
> In your testing, the only thing that gets rid of the problem is to
> select Custom and pick all of those settings in that chart on page 6.
> The total budget at all receivers for solving senders' problems is $0.
> If you want them to accept your mail and manage it the way you want,
> send it the way the spec says to. --John Levine
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Linux-PowerEdge