freq. scaling vs power consumption

[Eoin_Mcguinn at Dell.com]

At idle while running the NOP thread, the CPU isn't loaded, so the base
consumption whether at 300~MHz or 3000MHz is similar. 
Specifically, what watt-readings are you getting at what speed configs?

-----Original Message-----
From: linux-poweredge-bounces at dell.com
[mailto:linux-poweredge-bounces at dell.com] On Behalf Of michalwd1979
Sent: 20 November 2007 05:42
To: linux-poweredge-Lists
Subject: freq. scaling vs power consumption

Hi,
I wanted to tell about some strange results that I got yesterday. I had
an opportunity to accurate measure the power that my 2850 uses and the
results are not as expected. I have 2x 3GHz XEON cpus, 4GB ram, 3xHDD. I
compiled my kernel with p4-clockmod freq. driver and
/sys/devices/system/cpu/... shows that scaling is working. Max freq.
3000MHz min freq. 375MHz. The system is considerably slower with lowest
frequency, so I assume that processors speed really change. However
power used by the server does NOT change! I can see the difference with
no-load vs full-load (about 70W), but not with 3000 vs 375MHz. Does
anybody have any idea why? In that way using any frequency scaling (on
this machine) is not useful at all. 
I also heard that freq. scaling on servers is not the best idea, but I
am not sure why. Does anybody using freq. scaling on his/her servers and
can share with his/her opinions?

Regards,
Michael W.

_______________________________________________
Linux-PowerEdge mailing list
Linux-PowerEdge at dell.com
http://lists.us.dell.com/mailman/listinfo/linux-poweredge
Please read the FAQ at http://lists.us.dell.com/faq