Omar Sultan - Cisco CCIE No. 1284 Routing and Switching in his recent Cisco Data Center blog entry, made an interesting point that certainly adds more controversy to our blog story:
Cisco loses $2M order to ruthless Nortel energy efficiency calculator
 Nortel asserts their ERS 8610 offers energy savings of 60% over a 6500 equivalent. There is not a lot of detail to the comparison, although there is a footnote that states:
Unless noted, all product comparisons are based on vendor published maximum power ratings. So this seems a comparison of power supplies, not actual draw. |
The salient point is that these power values are for two switches plugged-in and idling with no connections--the assertion is that if there is such a discrepancy at idle, imagine what happens when we actually forward packets.
I guess the moral of the story here is, when it comes to energy efficiency, there is, sadly, no one magic metric to measure goodness. Its a design function like anything else in the data center and its a matter of doing research and balancing the design parameters.
View Omar's entire blog entry.
Omar's point that a comparison at idle, may not accurately measure what happens when packets are forwarded is a good one.
Do YOU agree with Omar that when it comes to energy efficiency, there is no one magic metric to measure goodness?
About Brad Reese on Cisco
Brad Reese is research manager at BradReese.Com, advancing the careers of 1 million certified individuals in the growing Cisco Career Certification Program.
Contact him.
Brad's blogroll
Brad Reese on Cisco archive.
Cisco Subnet
|
|
there is no one magic metric to measure goodness....
But there are lots of ways to measure badness. Typical obsfucation, Omar. When confronted by an inconvenient truth, deflect and dodge, but don't answer the question.
This seems incredibly simple to me. Test the systems under whatever load you prefer and publish the numbers. Then let us compare. Don't tell me to go to your power calculator and plug in 100 different parameters (and oh, by the way, you need a cisco.com account).
And finally, please, please, please stop citing Meircom as an "independent" testing organization. I haven't seen a Meircom test in the last decade that didn't feature a Cisco device winning...
What else will be helpful?
Anon111:
It would seem that Cisco has provided what you are looking for: the Meircom report I linked to in my blog provides power draw at idle, 5% uplink load and 100% uplink load. Since I noted that power draw is proportional to switch load, you can approximate draw for your particular environment.
The Power Calculator and DCAP are tools for our customers to determine actual power and cooling load, instead of working from the max draw, which is a level of granularity that our customers find helpful. Yes, you do need to register with Cisco.com to access them--sorry for not mentioning that. Again, the goal is to give customers greater transparency on actual power draw. While the complexity in inherent, we do try and avoid it, but the reality is that a switch pushing 10GbE will have a higher draw than the same switch pushing GE and most customers have told us they are interested in that level of detail.
The original point in my blog was that measuring the draw of an idling switch, the basis of the Nortel claims, was not all that helpful. While I have engaged in much discussion since that blog post, I have yet to have anyone disagree with that point.
That being said, any feedback on metrics you would like to see is always welcome.
Omar Sultan
SMM - Data Center Switching
Cisco Systems
www.cisco.com/go/dcswitching