HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Hours:
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
BladeSystem - General
cancel
Showing results for 
Search instead for 
Did you mean: 

BL460C G6 power usage much higher than DL380 G6

 
BeltwayBenny
Advisor

BL460C G6 power usage much higher than DL380 G6

Using the HP Power Advisor I've configured a BL460c G6 with 2 x X5570, 64GB RAM, 2 drives, 1 mezz card, 60% system utilization. The power advisor says I will use approx 454 watts @ 208v to run this blade.

I also have a blade exactly the same as what I describe above in our chassis. This particular blade is at about 15% CPU utilization and is currently using 436 watts @ 208v.

When I punch up a DL380 G6 with the same specs (power advisor) I see that I'd use only 311 watts @ 208v.

How come the blade is using +120 watts more than the rack mount model?

Ben
1 REPLY
rick jones
Honored Contributor

Re: BL460C G6 power usage much higher than DL380 G6

Any number of possibilities come to mind, including the possibility there is a bug in the power advisor.

Also, the DL380 G6 cannot have completely the same specs as the BL460c G6 if the BL460c G6 has a Mezz card installed.

The power advisor and/or any vendors power estimators are nice and all, but I've developed a preference for direct measurement myself.

Similarly, the built-in power monitors in some kit is nice and all, and certainly will give useful data, is power usage going up, is it going down etc etc, but when making system to system comparisons, I have developed a preference for direct measurement with a dedicated power analyzer.

Some power calculators out there in the world may derive their numbers from benchmarks such as SPECpower_ssj2008 - one common confusion though is that N% load in SPECpower_ssj2008 corresponds to N% CPU utilization in the processors. There are only two points where that might be reasonably ass-u-me-d - at 100% load and at 0% load. This can be a further source of discrepancy when comparing power calculators with live systems.
there is no rest for the wicked yet the virtuous have no pillows