FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Debian > Debian User

 
 
LinkBack Thread Tools
 
Old 07-10-2008, 06:55 AM
andy
 
Default OT: Energy efficiency difference btw GNU/Linux, Mac & MS

Hello all

As part of my studies I must draw up a spec for providing a hypothetical
building with power sourced solely from renewables (solar, PV, wind).
This building is an educational establishment for about 20 people using
computers. The budget is (naturally) tight.


Logically, before one powers a building, one needs to ensure that the
existing loads are the most efficient that they can be so that the
supply is not being wasted by hungry loads.


What I want to find out is whether anyone here knows of any
studies/reports that identifies whether or not there is a difference in
the energy efficiency among GNU/Linux systems, Mac and Microsoft.


I can easily make the argument that licensing and maintenance costs
would be cheaper using GNU/Linux, as well as recommending either a
system of laptops and/or a system of thin clients.


Does anyone have any thoughts on this issue?

Thanks

Andy

--

"If they can get you asking the wrong questions, they don't have to worry about the answers." - Thomas Pynchon, "Gravity's Rainbow"


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 07-10-2008, 05:27 PM
Andrew Sackville-West
 
Default OT: Energy efficiency difference btw GNU/Linux, Mac & MS

On Thu, Jul 10, 2008 at 07:55:56AM +0100, andy wrote:
> Hello all
>
> As part of my studies I must draw up a spec for providing a hypothetical
> building with power sourced solely from renewables (solar, PV, wind).
> This building is an educational establishment for about 20 people using
> computers. The budget is (naturally) tight.
>
> Logically, before one powers a building, one needs to ensure that the
> existing loads are the most efficient that they can be so that the
> supply is not being wasted by hungry loads.
>
> What I want to find out is whether anyone here knows of any
> studies/reports that identifies whether or not there is a difference in
> the energy efficiency among GNU/Linux systems, Mac and Microsoft.

I bet it's pretty hard to find a reasonable, non-biased study about
this, but if you find one, I'd be intrigued.

>
> I can easily make the argument that licensing and maintenance costs
> would be cheaper using GNU/Linux, as well as recommending either a
> system of laptops and/or a system of thin clients.

ISTM that regardless of who's software is more efficient, arguably the
best method is thin clients, from an energy perspecitve. This is based
on the assumption that you will have a few 24/7 machines anyway. And
that points, at least in my mind, a little bit towards OSS because of
the inexpensive virtualization options. A few physical machines
running at nearly full capacity seems to me to be more energy
efficient than a bunch of machines running at lower loads.

But that is all idle speculation around the water cooler.

A
 
Old 07-11-2008, 01:35 AM
Ron Johnson
 
Default OT: Energy efficiency difference btw GNU/Linux, Mac & MS

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 07/10/08 12:27, Andrew Sackville-West wrote:
[snip]
>
> ISTM that regardless of who's software is more efficient, arguably the
> best method is thin clients, from an energy perspecitve. This is based
> on the assumption that you will have a few 24/7 machines anyway. And
> that points, at least in my mind, a little bit towards OSS because of
> the inexpensive virtualization options. A few physical machines
> running at nearly full capacity seems to me to be more energy
> efficient than a bunch of machines running at lower loads.
>
> But that is all idle speculation around the water cooler.

Second that. There's no hard drive to spin up every 5 minutes, the
monitor will of course be configured to power off after N number of
minutes of inactivity, and similarly the CPU can be configured to go
into deep slowdown mode during inactive times.

Thus, after-hours, all the clients will be sipping power.

- --
Ron Johnson, Jr.
Jefferson LA USA

"Kittens give Morbo gas. In lighter news, the city of New New
York is doomed."
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkh2uO0ACgkQS9HxQb37XmexzwCgtWQb02axTb j3Ui7VbxMgymdq
RfoAn10OPt8wO/El1acczqlaGeDy490/
=Wkr+
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 07-11-2008, 01:48 AM
 
Default OT: Energy efficiency difference btw GNU/Linux, Mac & MS

>
>
>
>---- Original Message ----
>From: andrew@farwestbilliards.com
>To: debian-user@lists.debian.org
>Subject: Re: OT: Energy efficiency difference btw GNU/Linux, Mac & MS
>Date: Thu, 10 Jul 2008 10:27:55 -0700
>
>>On Thu, Jul 10, 2008 at 07:55:56AM +0100, andy wrote:
>>> Hello all
>>>
>>> As part of my studies I must draw up a spec for providing a
>hypothetical
>>> building with power sourced solely from renewables (solar, PV,
>wind).
>>> This building is an educational establishment for about 20 people
>using
>>> computers. The budget is (naturally) tight.
>>>
>>> Logically, before one powers a building, one needs to ensure that
>the
>>> existing loads are the most efficient that they can be so that the
>
>>> supply is not being wasted by hungry loads.
>>>
>>> What I want to find out is whether anyone here knows of any
>>> studies/reports that identifies whether or not there is a
>difference in
>>> the energy efficiency among GNU/Linux systems, Mac and Microsoft.
>>
>>I bet it's pretty hard to find a reasonable, non-biased study about
>>this, but if you find one, I'd be intrigued.
>>
>>>
>>> I can easily make the argument that licensing and maintenance
>costs
>>> would be cheaper using GNU/Linux, as well as recommending either a
>
>>> system of laptops and/or a system of thin clients.
>>
>>ISTM that regardless of who's software is more efficient, arguably
>the
>>best method is thin clients, from an energy perspecitve. This is
>based
>>on the assumption that you will have a few 24/7 machines anyway. And
>>that points, at least in my mind, a little bit towards OSS because
>of
>>the inexpensive virtualization options. A few physical machines
>>running at nearly full capacity seems to me to be more energy
>>efficient than a bunch of machines running at lower loads.
>>
>>But that is all idle speculation around the water cooler.
>>
>>A


IMHO it would be difficult, from an energy perspective, to
differentiate between any of several O/Ss, each of which ran on an
identical CPU. The major contributors to power dissipation are
usually the power supply, the HDD and to a lessor degree the fan.
The energy consumed by the CPU is considerably less than that of the
power supply (by perhaps a factor of 5) and that of memory even less.
That having been said, however, it is certainly true that the less
memory required the less energy required and the smaller HDD
required, the less energy. Both of these favor the *IX varients.
The last factor would be the speed of operation (the lower the speed,
especially with DRAM, the less energy required), but I can't make a
case one way or the other since most try and run the CPU at maximum.
Larry



--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 07-11-2008, 01:26 PM
andy
 
Default OT: Energy efficiency difference btw GNU/Linux, Mac & MS

owens@netptc.net wrote:


---- Original Message ----
From: andrew@farwestbilliards.com
To: debian-user@lists.debian.org
Subject: Re: OT: Energy efficiency difference btw GNU/Linux, Mac & MS
Date: Thu, 10 Jul 2008 10:27:55 -0700



On Thu, Jul 10, 2008 at 07:55:56AM +0100, andy wrote:


Hello all

As part of my studies I must draw up a spec for providing a

hypothetical


building with power sourced solely from renewables (solar, PV,

wind).


This building is an educational establishment for about 20 people

using


computers. The budget is (naturally) tight.

Logically, before one powers a building, one needs to ensure that

the


existing loads are the most efficient that they can be so that the

supply is not being wasted by hungry loads.


What I want to find out is whether anyone here knows of any
studies/reports that identifies whether or not there is a

difference in


the energy efficiency among GNU/Linux systems, Mac and Microsoft.


I bet it's pretty hard to find a reasonable, non-biased study about
this, but if you find one, I'd be intrigued.



I can easily make the argument that licensing and maintenance

costs


would be cheaper using GNU/Linux, as well as recommending either a

system of laptops and/or a system of thin clients.


ISTM that regardless of who's software is more efficient, arguably


the


best method is thin clients, from an energy perspecitve. This is


based


on the assumption that you will have a few 24/7 machines anyway. And
that points, at least in my mind, a little bit towards OSS because


of


the inexpensive virtualization options. A few physical machines
running at nearly full capacity seems to me to be more energy
efficient than a bunch of machines running at lower loads.


But that is all idle speculation around the water cooler.

A




IMHO it would be difficult, from an energy perspective, to
differentiate between any of several O/Ss, each of which ran on an
identical CPU. The major contributors to power dissipation are
usually the power supply, the HDD and to a lessor degree the fan.
The energy consumed by the CPU is considerably less than that of the

power supply (by perhaps a factor of 5) and that of memory even less.
That having been said, however, it is certainly true that the less
memory required the less energy required and the smaller HDD
required, the less energy. Both of these favor the *IX varients.
The last factor would be the speed of operation (the lower the speed,

especially with DRAM, the less energy required), but I can't make a
case one way or the other since most try and run the CPU at maximum.
Larry





Larry, Ron, Sam & Andrew

Thank you all for your comments. I think what I will do then is to
propose that the site acquire better energy efficiency via a couple of
servers running our fave distro (so much for scientific objectivity!!)
and using thin clients with various energy savings options enabled for
monitors, etc. Sam (via a private email) recommended the XO-1 or
Classmate or Eee PCs as the most suitable candidate machines, so I will
see what more I can find out on those in the meantime.


Once again, many thanks for sharing your opinions.

Cheers

Andy

--

"If they can get you asking the wrong questions, they don't have to worry about the answers." - Thomas Pynchon, "Gravity's Rainbow"


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 07-11-2008, 01:43 PM
Ron Johnson
 
Default OT: Energy efficiency difference btw GNU/Linux, Mac & MS

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 07/11/08 08:26, andy wrote:
[snip]
> monitors, etc. Sam (via a private email) recommended the XO-1 or
> Classmate or Eee PCs as the most suitable candidate machines, so I will
> see what more I can find out on those in the meantime.

While these are neat laptops, they might not have the oomph the
clients need. And they'll need external monitors and keyboards
anyway, so why purchase what you don't need. Also, laptops "grow
legs", but dedicated thin-clients are (a) useless without the
server, (b) can be bolted to the desk, and (c) spec'ed to the
customer's needs.

- --
Ron Johnson, Jr.
Jefferson LA USA

"Kittens give Morbo gas. In lighter news, the city of New New
York is doomed."
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkh3Y4sACgkQS9HxQb37Xmc4uwCfb4hmhrN3zD NolsdvapFAjJZg
N7UAnjRM47GNH9FGCswrAm9c1FNTR3XO
=roUg
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 

Thread Tools




All times are GMT. The time now is 06:21 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org