Xeon dual core run hot and eat power

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Sunner

Elite Member
Oct 9, 1999
11,641
0
76
Originally posted by: Vegito
Originally posted by: Sunner

Power consumption IS a concern for a lot of people in the enterprise.
For one, exelctricity isn't free, and if you tell the beancounters that you can save bucks without actually giving up anything, they'll love you.
Server rooms are built to last, but let's face it, build a room today and you still won't know what you'll be packing into it in 5-10 years.

Sun wouldn't yell about their servers' power consumption unless they knew CFO's/CTO's were listening.

Oh and tasburrfoot78362, my definiton of a server vs a wrokstation is pretty much the same as Vegito's :)


I guess leases are different... we dont pay as much for our electricial bill.. on our lease, we pay 5% of building total.. which is pretty low.. there are other bigger companies than ours here..

I guess power is a problem only if you're actually paying for it..

Yeah, depends on contract, our bill is entirely dependant on how much we use.
And, again, heat/space is a consideration, our cooling can handle quite a bit, but pack 42 1U 2-ways into a rack and you've got an awful lot of heat in a very small space, that's a problem.
 

Pariah

Elite Member
Apr 16, 2000
7,357
20
81
Originally posted by: fuzzynavel
Don't know if this has been posted already but here is a link to Slashdot

400Watts during idle for the Dual core Xeon that gets its ass kicked by opterons

Sounds like a bogus reading to me. 393W idle, and 428W under load? Do Xeon's lack all the power savings features of the standard P4? I'm pretty sure they don't. Look at this link from TechReport:

Power consumption

Dual Xeon 3.4GHz idle: 176W
Dual Xeon 3.4GHz load: 340W

Why would a standard dual Xeon system cuts its power consumption in half idling, while a dual core dual Xeon only sees an 8% decrease? I think GamePC just screwed up.
 

Remedy

Diamond Member
Dec 1, 1999
3,981
0
0
Chris used engineering samples. Which could've lacked that EIST. I haven't talked to him about it yet though.