Data Centre did not have an Energy Management System
So we “Winged” it and saved $3000 a month
“Energy Management by Observation”
One of our most important clients facilitates a decent size, Tier 2, data centre with about 12 Computer Room Air Conditioning Units or CRAC units as the technicians like to call them. We provide a Lon works Building Management System, nothing elaborate, just the basics, but no energy meters or enegry management specific engineering.
The Site Manager was under duress to find more funds to extend the BMS and mechanical Electrical, but without additional capital expense.
We found the monthly energy bill was the vicinity $12,000 to $15,000, so we had to put the “cart before the horse” so to speak, that is, find the savings BEFORE we put the Energy Management System in. The Energy Bill was metered around the 4th of each month.
So we did a few numbers regarding heating load in the room and cooling capacity of the CRAC units We decided to experiment.
Road test the theory
On the 3rd of the month we turned off a specific CRAC unit in a “low maintenance” area. We informed the nerds (sensitive bunch these) and monitored the system carefully for the remainder of the day. We took temperatures under the floor, above the racks, and monitored the equipment’s own temperature sensor on the disks. No change of any real significance….. so we left the CRAC unit off over night. There should be less heat load anyway, but we were cautious.
We grew complacent and left the CRAC unit off for the month, no issues, no nerds complaining ( we gave them pizza).
We did the same process the following month. We ran a spread sheet and keep in touch with the nerds to ensure all systems were OK on the inside where we couldn’t measure directly ourselves with a handheld device
At two CRAC units off we saved circa $3000 a month.
The third and last will be turned of next month, this should bring the month savings to $4000 (to be conservative)
Now the Maths
5 years = 60 months
60 x $4000 = $240,000.
This cant be right, we started in February, we did March and skipped April, now to do May, 2013. My concern is the summer months.
Lets reduce that 60 months by 15 months to allow for summer. there fore 45 months at $4000 which is $180,000.
This is still conservative, but when we extrapolate it over the five years the figure is enough to add major improvements. We expect it go to the bottom line, which is where it belongs..
How much energy is being consumed by equipment and supportive peripheral gear that we don’t actually need?
How much Energy is literally being wasted daily in your Facility?
Our costs for our involvement in this was circa $3000 +