Article: Nehalem Performance Preview
By: Jack (jumpingjack6.delete@this.verizon.net), April 10, 2009 7:31 pm
Room: Moderated Discussions
slacker (s@lack.er) on 4/10/09 wrote:
---------------------------
>Jack (Jumpingjack6@verizon.net) on 4/10/09 wrote:
>---------------------------
>>Peak power will dictate various thermal cooling designs and such, but total energy
>>consumed determines energy efficiency. A server can have 20% higher peak power
>>but if it completes the task in 50% less time will be more energy efficient.
>
>Only guaranteed to be true if you're strictly looking at the time span to complete
>the task on each server. Otherwise, you need to add the idle energy consumption
>when the faster server is sitting idle. This is what shot the Harpertown platform
>in the foot for workloads which you can imagine.
>
>David: consider writing a short article explaining energy vs. power, and where each metric is relevant.
True, I simplified it. The total energy over a long period depends on many factors, in terms of energy savings the sooner a task completes and gets to idle, caveat -- idle needs to be as low or lower than the competing or existing solution to be of benefit.
The concept is straight forward. A box that demonstrates peak power that is higher can actually be more energy efficient.
jack
---------------------------
>Jack (Jumpingjack6@verizon.net) on 4/10/09 wrote:
>---------------------------
>>Peak power will dictate various thermal cooling designs and such, but total energy
>>consumed determines energy efficiency. A server can have 20% higher peak power
>>but if it completes the task in 50% less time will be more energy efficient.
>
>Only guaranteed to be true if you're strictly looking at the time span to complete
>the task on each server. Otherwise, you need to add the idle energy consumption
>when the faster server is sitting idle. This is what shot the Harpertown platform
>in the foot for workloads which you can imagine.
>
>David: consider writing a short article explaining energy vs. power, and where each metric is relevant.
True, I simplified it. The total energy over a long period depends on many factors, in terms of energy savings the sooner a task completes and gets to idle, caveat -- idle needs to be as low or lower than the competing or existing solution to be of benefit.
The concept is straight forward. A box that demonstrates peak power that is higher can actually be more energy efficient.
jack
Topic | Posted By | Date |
---|---|---|
Nehalem review up | David Kanter | 2009/04/07 03:43 AM |
Nehalem review up | noone | 2009/04/07 06:48 AM |
Strange jbb on Harpertown | Henrik S | 2009/04/07 08:29 AM |
Strange jbb on Harpertown | David Kanter | 2009/04/07 11:19 AM |
Strange jbb on Harpertown | Henrik S | 2009/04/07 09:33 PM |
Strange jbb on Harpertown | Chris | 2009/04/08 12:54 AM |
Strange jbb on Harpertown | Henrik S | 2009/04/08 02:40 AM |
Nehalem review up | Vincent Diepeveen | 2009/04/07 08:34 AM |
Nehalem review up | Jack | 2009/04/09 04:51 PM |
Nehalem review up | Vincent Diepeveen | 2009/04/10 01:58 AM |
Nehalem review up | Michael S | 2009/04/10 03:45 AM |
Nehalem review up | EduardoS | 2009/04/10 07:01 AM |
Nehalem review up | Michael S | 2009/04/10 07:56 AM |
Nehalem review up | Eugene Nalimov | 2009/04/10 09:12 AM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/10 10:10 AM |
Choice of C compiler doesn't matter much for Java... | EduardoS | 2009/04/10 02:49 PM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/11 07:13 AM |
Choice of C compiler doesn't matter much for Java... | EduardoS | 2009/04/11 11:30 AM |
Large pages | David Kanter | 2009/04/11 02:02 PM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/11 11:06 PM |
Choice of C compiler doesn't matter much for Java... | Paul | 2009/04/12 01:53 AM |
Choice of C compiler doesn't matter much for Java... | iz | 2009/04/12 02:59 AM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/12 07:37 AM |
Choice of C compiler doesn't matter much for Java... | EduardoS | 2009/04/12 08:08 AM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/12 09:25 AM |
Choice of C compiler doesn't matter much for Java... | EduardoS | 2009/04/12 05:24 PM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/12 10:18 PM |
Thread costs | David Kanter | 2009/04/13 12:12 AM |
Thread costs | Henrik S | 2009/04/14 02:08 PM |
Choice of C compiler doesn't matter much for Java... | Michael S | 2009/04/11 08:53 AM |
Choice of C compiler doesn't matter much for Java... | Henrik S | 2009/04/11 11:08 PM |
Nehalem review up | Vincent Diepeveen | 2009/04/11 04:50 PM |
Nehalem review up | Michael S | 2009/04/11 05:12 PM |
Nehalem review up | Vincent Diepeveen | 2009/04/12 03:01 AM |
Nehalem review up | Michael S | 2009/04/12 05:07 AM |
Nehalem review up | rwessel | 2009/04/07 02:01 PM |
Nehalem review up | slacker | 2009/04/08 09:11 AM |
Energy vs. power | David Kanter | 2009/04/08 10:11 AM |
Energy vs. power | Vincent Diepeveen | 2009/04/10 02:08 AM |
Energy vs. power | slacker | 2009/04/10 09:26 AM |
Energy vs. power | RagingDragon | 2009/04/10 10:19 AM |
Energy vs. power | David Kanter | 2009/04/10 11:47 AM |
Energy vs. power | Jack | 2009/04/10 04:44 PM |
Energy vs. power | slacker | 2009/04/10 07:00 PM |
Energy vs. power | Jack | 2009/04/10 07:31 PM |
Energy vs. power | David Kanter | 2009/04/11 12:16 AM |
Nehalem review up | rwessel | 2009/04/08 02:32 PM |
Minor font issue | gpriatko | 2009/04/07 04:35 PM |
Minor HTML issue | David Kanter | 2009/04/07 09:38 PM |
Minor HTML issue | David Kanter | 2009/04/07 09:39 PM |
Good work, i look forward to linux and SP2 numbers (NT) | PiedPiper | 2009/04/08 01:52 AM |
Nehalem review up | Joe Chang | 2009/04/10 03:59 AM |