Hi,

llothar wrote:
> On 6 Apr., 13:56, post2goo...@yahoo.com wrote:
>
>>You know, I thought that nothing could be louder
>>than the Dell Poweredge 1850. The T1000 makes
>>it seem like the Poweredge is turned off. And before
>>someone says "it's designed to be in a server area",
>>let me add that if you put a bunch of these things in
>>a rack, we'd be talking mandatory OSHA hearing
>>protection required, if not being tagged an EPA
>>superfund site. It's a good machine, but not
>>NEARLY powerful enough to justify the noise.
>>
>>But enough ranting. What I wanted to ask was
>>whether anyone had tried to replace the fans
>>on the T1000 with quieter aftermarket ones? Or
>>found some other way to quiet these things down.
>>Thanks.
>
>
> Well it's the tradition of Sun to give a bullshit on noise.
>
> I was pretty surprised that most of the Fan's inside the
> Sun Desktop Machines already have a "Silent Fan" label.
>
> And yes, Sun Fanboys will surely answer here too. But taking
> a 72W CPU machine with 2 10k/rpm Disk and producing this
> noise let look the Sun Engineers as complete idiots.
>
> I can build a very cool Quad Core with 130Watt and 4 Raptors
> which is very quiete compared to this machine and safe
> Using new Fan Technology - just a few bucks more expensive
> then.
>
What is the " new Fan Technology"? Or are you refering to bigger heatsinks?

Making a machine boot and run is easy, making it work in all specified
environments is much harder and takes some testing.

I am not defending Suns noise, but the performance on the machines is
very good from cooling point of view.

> The real problem is that even the Ultra 20/40 for desktops
> are unuseable noisy.
>
They are maybe not for desktop usage I would say, atleast not the u45 I
just received!

I would put the u45 next to a workgroup server due to it's expandability
and cooling possibilities, it has 3 big fans cooling everything exept
for the PSU which has two small fans running very quite. .


/michael

> True. Maybe over-engineering?

Maybe not care ?

On 7 Apr., 01:55, Casper H.S. Dik <Casper....@Sun.COM> wrote:
> "llothar" <llot...@web.de> writes:
> >Well it's the tradition of Sun to give a bullshit on noise.
>
> For servers; the current desktops don't make much noise.
> (The SunBlade 1000 was the first serious SPARC desktop with
> didn't produce an annoying amount of noise)
>

Come on, if you think that the Blade 1000 (i tried it with a 750Mhz
and a dual 900 Mhz)
has an acceptable noise rate then you have a real damaged brain.

I can't even work concentrated on a problem when the system is in the
next room.
Through 20 cm of finest german concrete i can here these beast.

On 8 Apr 2007 08:54:44 -0700
"llothar" <llothar@web.de> wrote:

> On 7 Apr., 01:55, Casper H.S. Dik <Casper....@Sun.COM> wrote:
> > "llothar" <llot...@web.de> writes:
> > >Well it's the tradition of Sun to give a bullshit on noise.
> >
> > For servers; the current desktops don't make much noise.
> > (The SunBlade 1000 was the first serious SPARC desktop with
> > didn't produce an annoying amount of noise)
>
> Come on, if you think that the Blade 1000 (i tried it with a 750Mhz
> and a dual 900 Mhz)
> has an acceptable noise rate then you have a real damaged brain.

It's not the most silent of systems, but it's quite acceptable - on par
with, for example, a Fujitsu-Siemens Scenic E600. A lot depends on the
disk drive(s). In my Blade 2000 I use drives with fluid bearings that
are almost noiseless. The E600 is actually at this moment far noisier,
because its disk drive is producing a very audible whine (the type of
noise disk drives make when their bearings are worn out).

> I can't even work concentrated on a problem when the system is in the
> next room.
> Through 20 cm of finest german concrete i can here these beast.

Now you're being quite ridiculous. The only time that would be the case
is when the fan control isn't working and the fans are going full
blast. You did configure the OS properly?

--
Stefaan A Eeckels
--
Never explain by malice what can be adequately explained by stupidity.
However:
Sufficiently advanced stupidity is indistinguishable from malice.

"Frank Cusack" <fcusack@fcusack.com> wrote in message
news:m2veg9dkag.fsf@sucksless.local...
> On Sat, 07 Apr 2007 00:40:30 +0200 Alexander Skwar
> <usenet@alexander.skwar.name> wrote:
>> · Frank Cusack <fcusack@fcusack.com>:
>>
>>> On 6 Apr 2007 14:31:03 -0700 "llothar" <llothar@web.de> wrote:
>>>> The real problem is that even the Ultra 20/40 for desktops
>>>> are unuseable noisy.
>>>
>>> I hate to risk putting myself into the Sun fanboy category, but the
>>> real problem is that many many many people put these servers into
>>> rooms with inadequate environmental controls.
>>
>> Well, but why is it, that especially Sun hardware is always
>> so extremely loud? Other server hardware (eg. HP, be it Unix
>> or be it Proliant) isn't nearly as loud.
>
> True. Maybe over-engineering?

Or maybe the loud noise is to scare the bugs away....

On a more serious note, a rack of X100/V100's sounds like a jumbo jet
engine. That noise adds up pretty quick in a room full of servers.

On Apr 6, 4:56 pm, post2goo...@yahoo.com wrote:
> You know, I thought that nothing could be louder
> than the Dell Poweredge 1850. The T1000 makes
> it seem like the Poweredge is turned off. And before
> someone says "it's designed to be in a server area",
> let me add that if you put a bunch of these things in
> a rack, we'd be talking mandatory OSHA hearing
> protection required, if not being tagged an EPA
> superfund site. It's a good machine, but not
> NEARLY powerful enough to justify the noise.
>
> But enough ranting. What I wanted to ask was
> whether anyone had tried to replace the fans
> on the T1000 with quieter aftermarket ones? Or
> found some other way to quiet these things down.
> Thanks.
>
> -P


If you are paying for support, the first thing to try
would be to make your problem Sun's problem.

..

"Casper H.S. Dik" wrote:
> "llothar" <llothar@web.de> writes:
[snip]
> >And yes, Sun Fanboys will surely answer here too. But taking
> >a 72W CPU machine with 2 10k/rpm Disk and producing this
> >noise let look the Sun Engineers as complete idiots.
>
> Supressing noise in a small package costs money;

I slightly disagree here. For example for RC model planes there are
air-screws available which are much less noisy than normal ones. In
theory the same could be done for a server, too.
Another item which may contribute to the noise is the transfer of the
vibrations to the chassis itself (disclaimer: I am not an enginner, I
may be horribly wrong) which makes the whole thing "swing". Long ago I
modified my old Ultra1 (e.g. removed the normal rails and replaced them
with self-made ones made from a very soft polysiloxane/air-material)
that the hard disk no longer makes the whole chassis go
greeeeeeeeeeeeeeeeeeeee... which greatly reduced the amount of
high-pitched greeeeeee...-noise which was very annoying. I guess the
same could be done for the T1000 to reduce some of the vibrations.

----

Bye,
Roland

--
__ . . __
(o.\ \/ /.o) roland.mainz@nrubsig.org
\__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer
/O /==\ O\ TEL +49 641 7950090
(;O/ \/ \O

On 2007-04-08 21:31:20 +0100, Roland Mainz <roland.mainz@nrubsig.org> said:

> I slightly disagree here. For example for RC model planes there are
> air-screws available which are much less noisy than normal ones. In
> theory the same could be done for a server, too.

If the noise is fan noise as opposed to turbulence noise as air is
driven through small gaps full of bumps.

On Sun, 08 Apr 2007 22:31:20 +0200
Roland Mainz <roland.mainz@nrubsig.org> wrote:

> I guess the
> same could be done for the T1000 to reduce some of the vibrations.

Indeed, but it would not be noticeable in the typical computer room.

Speaking about noticeable, I just installed a couple of Cisco ASA5520
firewalls, and these things make a very obnoxious warbling noise that is
clearly audible in a room with about 30 HP and Dell 1U servers, a
T2000, a bunch of V890s, V440s, A3320s, A6600s, 280Rs, switches, load
balancers, routers, and the air-conditioning.

--
Stefaan A Eeckels
--
When the need is strong, there are those who will believe anything.
-- Arnold Lobel

On 2007-04-08 23:22:39 +0100, Stefaan A Eeckels <hoendech@ecc.lu> said:

> Indeed, but it would not be noticeable in the typical computer room.

It also may not be desirable. One of the things that is bad for disks,
both (I am sure) in terms of performance and (I suspect) in terms of
reliability is vibration, and particularly in this context self-induced
vibration. That means you need to be careful about how you mount
disks, and you probably want them bolted to something solid like a
chassis.

Of course it's not that simple, since if you have a bunch of disks
bolted to a common chassis they can all now "hear" each other, and that
may not be good either, so you have to think about how you make the
chassis and that's something that's way off topic... Finally, this
probably counts more for arrays in OLTP application where you have a
large number of very fast disks all seeking like mad than it does for a
pair of slower system disks which probably seek hardly at all in
typical use.

(Incidentally, before you laugh at this: the performance impact is
apparently a real thing. The mechanism is head-settle time - vibration
can cause heads to take longer to settle after a seek, and this can
cause them to miss a rotational window which means you lose a whole
rotation which is a significant time. There was an article by someone
at NetApp which talked about this (among other things) in ;login: a few
months ago. Obviously more of an issue for arrays though...)

--tim