Replace UPS with On-Board Battery Backup on IT Equipment
Emergency Power for IT Equipment: On-Board Battery Backup vs. UPS
Using on-board battery backup for each server rather than using centralized uninterruptable power supplies.
Item ID: 491
Power Systems--Backup Power
Technical Advisory Group: 2013 Information Technology TAG (#8)
Average TAG Rating: 2.18 out of 5
TAG Ranking Date: 10/25/2013
TAG Rating Commentary:
- A good strategy. It looks like the market is somewhat limited to larger, more specialized IT operations. Should be pursued to see if there is a more widespread application.
- Equipment not avail to do easily
- Would rate higher, but I do not see this equipment as available at a significantly wide market. Better than DC power distribution.
- Not an ET.
Uninterruptible power supplies (UPS) can be invaluable in ensuring that access to data is maintained during power outages while backup generators start up. However, there are inherent inefficiencies in UPS systems. According to a study funded by Lawrence Berkeley National Laboratory (LBNL) [LBNL page 1], the State of California alone spends $100 million on 1 billion kWh of electricity for data centers, mostly due to conversion losses in data center UPSs. According to the LBNL report, typical UPS systems have inefficiencies of 86-90% depending on their load. Other sources note that high-efficiency UPS are 92-95% and older models can be much lower efficiency. (LBNL, 2005 Pg 1-2)
A good amount of data center innovation is driven by the biggest companies with enough buying power to convince suppliers to modify their designs, and a green image, but such innovations can be emulated by smaller data center designers and operators later if not sooner.
Starting in 2009, Google chose to switch over from a centralized UPS to onboard, lead-acid, 12-volt battery backups on each of its custom-built servers. These batteries are 99.9% efficient and were a key element in Google achieving a PUE of under 1.2. (Fontecchio, 2009) (Fehrenbacher, 2009) Two months later, Facebook announced that it intended to follow suit and estimated that it will reduce its power distribution losses from 35% to 15% and save millions of dollars in UPS costs. (Miller, 2009) (Miller 2, 2009)
If this technology catches on, data centers can cover only part of their power distribution with UPS and charge clients less who are willing to use servers with backup batteries.
Baseline Description: 70 kVA conventional UPS
Baseline Energy Use: 218794 kWh per year per unit
According to the California Energy Commission: "UPSs in a data center environment typically operate at or below 50 percent of their rated active power output. This is primarily because they are often part of a redundant system in which they share the load with another UPS but must be sized to carry the full load. The measured efficiency of a given UPS also decreases by 1 to 2 percent when meeting nonlinear loads such as the switch mode power supplies used in low-end servers...Double conversion, also called “online,” is the most common configuration because it is capable of completely isolating sensitive IT loads from unconditioned utility power." According to field tests by the Lawrence Berkeley National Laboratory, a 70 kVA UPS of this type at 38% load would be about 85.2% efficient with a power factor of 80%. It would consume an average of 70 kVA * .38 * 24 * 365 * .8 / .852 = 218,794 kWh per year. (LBNL, 2005 Pg 36)
Manufacturer's Energy Savings Claims:
Manufacturers of server backup batteries are not listing efficiencies. Apparently their target audience is home owners with one or two computers whose main goal is reliability and price rather than data center operators looking to replace UPS with more efficient backup batteries. The batteries used by Google and Facebook don't seem to be advertised commercially yet. However, research highted in the synopsis above notes that the efficiency of UPS ranges from 86% to 95% and Google reported that the backup batteries they used were 99.9% efficient, so potentially this indicates energy savings potential of 5-14%. Google also noted that these batteries may be 20% more efficient than their UPS. (Fontecchio, 2009) Facebook reported their backup efficiency dropped by 20. (Miller, 2009) So it's fair to say that the potential for savings is significant.
Best Estimate of Energy Savings:
"Typical" Savings: 2%
Energy Savings Reliability: 3 - Limited Assessment
Energy Use of Emerging Technology:
214,418.1 kWh per unit per year
Energy Use of an Emerging Technology is based upon the following algorithm.
Baseline Energy Use - (Baseline Energy Use * Best Estimate of Energy Savings (either Typical savings OR the high range of savings.))
Currently no data available.
Currently no data available.
Simple payback, new construction (years): N/A
Simple payback, retrofit (years): N/A
Cost Effectiveness is calculated using baseline energy use, best estimate of typical energy savings, and first cost. It does not account for factors such as impacts on O&M costs (which could be significant if product life is greatly extended) or savings of non-electric fuels such as natural gas. Actual overall cost effectiveness could be significantly different based on these other factors.