[196228] in North American Network Operators' Group
Re: Temp at Level 3 data centers
daemon@ATHENA.MIT.EDU (Zachary Winnerman)
Fri Oct 13 13:08:02 2017
X-Original-To: nanog@nanog.org
To: Matt Harris <matt@netfire.net>,
David Hubbard <dhubbard@dino.hostasaurus.com>, nanog@nanog.org
From: Zachary Winnerman <zacharyw09264@gmail.com>
Date: Wed, 11 Oct 2017 13:08:33 -0400
In-Reply-To: <CAHdm836QAA3jE4MLEF_yUXtaJfP0qXXdjG4+F1CjkOF+jy2k0w@mail.gmail.com>
Errors-To: nanog-bounces@nanog.org
This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--3RXtqVmkJG37tEAg3WhI5QwLuedGPIt4p
From: Zachary Winnerman <zacharyw09264@gmail.com>
To: Matt Harris <matt@netfire.net>,
David Hubbard <dhubbard@dino.hostasaurus.com>, nanog@nanog.org
Message-ID: <85f956d1-1796-80ea-b40a-74493f0fc100@gmail.com>
Subject: Re: Temp at Level 3 data centers
References: <39086E4C-0BDF-40BF-922A-38A9457F82FF@dino.hostasaurus.com>
<D60397C6.3B0F8%sam@coeosolutions.com>
<37BBD972-0260-4207-8180-DD9C1D7A37FC@neilltech.com>
<0f79a0d2-5ee5-a5d1-aea1-86487e9de60d@gmail.com>
<CAHdm836QAA3jE4MLEF_yUXtaJfP0qXXdjG4+F1CjkOF+jy2k0w@mail.gmail.com>
In-Reply-To: <CAHdm836QAA3jE4MLEF_yUXtaJfP0qXXdjG4+F1CjkOF+jy2k0w@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
That's a good point, though if you are running your breakers that close
I think you have bigger problems, as a power outage, however unlikely,
could cause your equipment to not come back up at all. Software updates
that reboot several servers in quick succession could also cause a
breaker to trip under those circumstances. Unfortunately, there's no way
to tell how close a breaker is to tripping without tripping it. Breakers
may have amp meteres and a rated size, but the actual load before
tripping is +-20% for common models, meaning a 20A breaker may trip as
low as 16A.
On 2017=E5=B9=B410=E6=9C=8811=E6=97=A5 12:58, Matt Harris wrote:
> Another thing to remember - and I've actually seen breakers tripped on
> PDUs due to heat before because of this - is that it's going to spin
> all of your fans harder to keep internal temps down if the ambient
> temp is higher. This will increase your power draw, which means that
> if you're paying for metered power by usage, you're going to pay more
> - those fans really do add up in terms of power. In extreme cases, you
> can draw too much power and trip a breaker on a PDU because every host
> in a rack, or especially those towards the top, are spinning full
> tilt. It's not a good condition and one that you should force them to
> correct.=C2=A0
>
>
> On Wed, Oct 11, 2017 at 11:54 AM, Zachary Winnerman
> <zacharyw09264@gmail.com <mailto:zacharyw09264@gmail.com>> wrote:
>
> I recall some evidence that 80+F temps can reduce hard drive lifeti=
me,
> though it might be outdated as it was from before SSDs were around.=
I
> would imagine that while it may not impact the ability for a server=
to
> handle load, it may reduce equipment lifetime. It also could be an
> indication that they lack redundancy in the case of an AC failure.
> This
> could cause equipment damage if the datacenter is unattended and
> temperatures are allowed to rise.
>
>
> On 2017=E5=B9=B410=E6=9C=8811=E6=97=A5 11:45, Keith Stokes wrote:
> > There are plenty of people who say 80+ is fine for equipment and
> data centers aren=E2=80=99t built for people.
> >
> > However other things have to be done correctly.
> >
> > Are you sure your equipment is properly oriented for airflow
> (hot/cold aisles if in use) and has no restrictions?
> >
> > On Oct 11, 2017, at 9:42 AM, Sam Kretchmer
> <sam@coeosolutions.com
> <mailto:sam@coeosolutions.com><mailto:sam@coeosolutions.com
> <mailto:sam@coeosolutions.com>>> wrote:
> >
> > with a former employer we had a suite at the L3 facility on Canal=
in
> > Chicago. They had this exact issue for the entire time we had
> the suite.
> > They kept blaming a failing HVAC unit on our floor, but it went
> on for
> > years no matter who we complained to, or what we said.
> >
> > Good luck.
> >
> >
> > On 10/11/17, 7:31 AM, "NANOG on behalf of David Hubbard"
> > <nanog-bounces@nanog.org
> <mailto:nanog-bounces@nanog.org><mailto:nanog-bounces@nanog.org
> <mailto:nanog-bounces@nanog.org>> on behalf of
> dhubbard@dino.hostasaurus.com
> <mailto:dhubbard@dino.hostasaurus.com><mailto:dhubbard@dino.hostasa=
urus.com
> <mailto:dhubbard@dino.hostasaurus.com>>> wrote:
> >
> > Curious if anyone on here colo=C2=B9s equipment at a Level 3 faci=
lity
> and has
> > found the temperature unacceptably warm?=C2=A0 I=C2=B9m having th=
at experience
> > currently, where ambient temp is in the 80=C2=B9s, but they tell =
me
> that=C2=B9s
> > perfectly fine because vented tiles have been placed in front of =
all
> > equipment racks.=C2=A0 My equipment is alarming for high temps, s=
o
> obviously
> > not fine.=C2=A0 Trying to find my way up to whomever I can compla=
in
> to that=C2=B9s
> > in a position to do something about it but it seems the support
> staff
> > have been told to brush questions about temp off as much as
> possible.
> > Was wondering if this is a country-wide thing for them or unique
> to the
> > data center I have equipment in.=C2=A0 I have equipment in severa=
l
> others from
> > different companies and most are probably 15-20 degrees cooler.
> >
> > Thanks,
> >
> > David
> >
> >
> >
> > ---
> >
> > Keith Stokes
> >
> >
> >
> >
>
>
>
>
>
> --=20
> Matt Harris - Chief Security Officer
> Main:=C2=A0+1 855.696.3834 ext 103
> Mobile:=C2=A0+1 908.590.9472
> Email:=C2=A0matt@netfire.net <mailto:matt@netfire.net>
--3RXtqVmkJG37tEAg3WhI5QwLuedGPIt4p
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"
-----BEGIN PGP SIGNATURE-----
iQEzBAEBCAAdFiEEOprgcwxSbwkgVk1lEmETIgLkL3IFAlneUBEACgkQEmETIgLk
L3Is0wf/QeczTkbe0Ln6rKETH/yOZomHXZwaOfF3ShImeoY/x4RAUWhZHScpzcWB
ZiGWbiq7rXe4V8bary7UwFCF3m3R3KCn8t1D6bFfGvCoBhmPmKQKDzF0dXz2EMSS
N7WLo5C+Ymnfd0l9Ow0992OA1VSPFTkVK/2QEK/VQk5htXVAkdaK8HSDHxIQaU7W
usASZ+PsoK9GVq5OmGwsZFiKgyhqVKzbR3vexZk8I1DwxzbTiAQW09hfljbfg3ly
iL2unElQTQAdvZHFWPJbs97BVbuD5SmHs4H8JDT67Ie9P8nhIP4tEnq+ZdvZxY5N
ykFXBZjHeJC9w5xHl63ss9ahUMEhNA==
=OFY1
-----END PGP SIGNATURE-----
--3RXtqVmkJG37tEAg3WhI5QwLuedGPIt4p--
