Download the authoritative guide: Enterprise Data Storage 2018: Optimizing Your Storage Infrastructure
Early in my career, I worked for an electric generating and transmission (G&T) utility, also known as a power company, developing and maintaining power plant operations-related software. Drawing on that experience has been rather handy lately with the focus on IT power and cooling topics. Let's take a closer look at some of those issues and other energy-related topics to see why reducing power and cooling to conserve on energy have become such "hot" issues.
Here are some of the needs driving the trend:
- Reducing costs to meet operating expenditures and budgets
- Limited power availability to areas, facilities or room locations
- Lack of adequate UPS and power conditioning or cooling capacity
- Address rising energy costs and energy surcharges for excess kWh usage
- Green environmental initiatives, including reducing carbon dioxide (CO2) emissions
|Figure-1: Average IT data center power and cooling energy consumption.|
Figure 2 shows various power and cooling as well as environmental metrics. For example, a device that uses one watt of electrical power for one hour uses one watt hour of power; if a device uses 1,000 watts of power and runs for an hour, then it uses one kilo watt (1,000 watts). Electrical energy is typically quoted in cents per kWh per month, with a base rate and some higher rates for energy used above a certain number of kWh per month.
Another example: a base rate for a kWh (which changes with energy prices, demand and availability by region) could be 12 cents per kWh and 20 cents per kWh for usage over 1,000 kWh per month. Note that some regions and electrical G&T utilities may also add energy surcharges.
|Kilowatts (kw)||Watts / 1,000||One thousand watts|
|Megawatts (mw)||kW / 1,000||One thousand kW|
|BTU/hour||watts x 3.413||Heat generated in an hour from using energy in British Thermal Units. 12,000 BTU/hour can equate to 1 Ton of cooling.|
|kWh||1,000 watt hours||The number of watts used in one hour|
|Watts||Amps x Volts (e.g. 12 amps * 12 volts = 144 watts)||Unit of electrical energy power|
|Watts||BTU/hour x 0.293||Convert BTU/hr to watts|
|Volts||Watts / Amps (e.g. 144 watts / 12 amps = 12 volts)||The amount of force on electrons|
|Amps||Watts / Volts (e.g. 144 watts / 12 volts = 12 amps)||The flow rate of electricity|
|Volt-Amperes (VA)||Volts x Amps||Sometimes power expressed in Volt-Ampres|
|kVA||Volts x Amp / 1000||Number of kilovolt-ampres|
|kW||kVA x power-factor||Power factor is the efficiency of a piece of equipments use of power|
|kVA||kW / power-factor||Killovolt-Ampres|
|kVA||kW / power-factor||Killovolt-Ampres|
|Carbon Credit||Carbon offset credit||Offset credits that can be bought and sold to offset your CO2 emissions|
|CO2 Emission||Average 1.341 lbs per kWh of electricity generated||The amount of average carbon dioxide (CO2) emissions from generating an average kWh of electricity|
Figure-2: Various power and cooling metrics
IT technology manufacturers provide some combination of information pertaining to electrical energy consumption or heat (BTU/hour) generated for a given scenario. Some vendors provide more information, including worst-case and best-case consumption information, while others provide basic maximum breaker size information. Examples of metrics published by vendors and that should be visible on equipment can include kilowatt (kW), kV, Amps, VAC or BTU.
You can locate your equipment's power usage from vendor supplied documents and specifications as well as from "nameplates" found on the equipment. The "nameplate" electrical values may be in volts, amperes, kilovolt-amperes, watts or some other combination. The nameplate values may exceed the actual power and cooling requirements for the equipment since the values have a built-in safety margin, so you should review vendor specification documents for a more accurate assessment.
Some Facts to Consider
U.S. national average CO2 emission is 1.341lbs per kWh of electrical power. This number can vary depending on region of the country and source of fuel for the power generating station (power plant), with Eastern coal being higher in CO2, followed by Western lignite coal, petroleum and natural gas, hydro, nuclear, thermo and wind, among others.
Energy costs will vary by region and state, as well as from residential to business, based on power consumption levels. IT equipment energy costs need to reflect cooling costs that can be as much as twice those of the actual IT equipment, depending on the PUE (power usage effectiveness) of the data center. A typical gallon of gasoline (octane level will vary) will on average generate about 20 lbs of CO2. Approximately (this number is constantly changing) 78% of CO2 emissions in the U.S. are tied to electrical power generation. Carbon offset credits from organizations such as Terrapass can cost in the range of $250 per lb up to 10tons of CO2 and $200 per pound for larger quantities up to 35 tons of CO2.
The new IT industry trade organization thegreengrid.org, in addition to seeking new members from the IT server, networking, chip and storage industry sectors, has established some early metrics for use as general guidelines for discussing energy, power and cooling effectiveness in the data center. These metrics include:
- Data center efficiency (DCE)
- Power usage effectiveness (PUE)
- Data center Performance Efficiency (DCPE)
Data center efficiency (DCE) is the indicator ratio of IT data center energy efficiency, defined as IT equipment (servers, disk and tape storage, networking switches, routers, printers, etc.) divided by total facility power. For example, if the sum of all IT equipment energy usage was 1,500 kilowatt hours (kWh) per month, yet the total facility power including UPS, energy switching, power conversion and filtering, cooling and associated infrastructure costs as well as IT equipment, was 3,500 kWh, the DCE would be 1,500 / 3,500 = 0.43. DCE can be used as a ratio, for example, to show in the above scenario that IT equipment accounts for about 43% of energy consumed by the data center, with 57% of electrical energy consumed by cooling, conversion and conditioning or lighting.
Power usage effectiveness (PUE) is the ratio of total energy being consumed by the data center to energy being used to operate IT equipment. PUE is defined as total facility power divided by IT equipment energy consumption. Using the above scenario, PUE = 2.333 (3,500 / 1,500), which means that a server requiring 100 watts of power would actually require (2.333 * 100) 233.3 watts of energy, including both direct power and cooling costs. Similarly, a storage system that required 1,500 kWh of energy to power would require (1,500*2.333) 3,499.5 kWh of electrical power, including cooling.
Another metric that is move involved but has the potential to have the most meaning is data center performance efficiency (DCPE), which takes into consideration how much useful and effective work is performed by the IT equipment and data center per energy consumed. DCPE is defined as useful work divided by total facility power, with an example being some number of transactions processed using servers, networks and storage divided by the energy for the data center to power and cool the equipment.
The importance of these numbers and metrics is to focus on the larger impact of a piece of IT equipment, including its energy consumption, factoring in in cooling and other hosting or site environmental costs. Naturally, energy costs and CO2 (carbon offsets) will vary by geography and region, along with type of electrical power being used (coal, natural gas, nuclear, wind, thermo, solar, etc.) and other factors that should be kept in perspective as part of the big picture.
Many local and regional electrical utilities have rebate and incentive programs associated with reducing your energy footprint and consumption for homes and businesses, including leveraging more efficient and effective IT equipment.
One such energy utility is Pacific Gas and Electric (PG&E), which has programs targeted toward energy demand side management (DSM) for various localities. PG&E is not alone, since other energy utilities leverage DSM as part of their capacity planning and performance management of their resources, which is energy generation via their G&T facilities.
So how much does it cost to power 100TB (raw) of storage, and how much CO2 emissions are generated per year? That, of course, depends on the type of storage, the number and size of the disk drives, the cost per kWh of power, cooling costs and the average number of lbs of CO2 produced per kWh.
One example is a single storage system using 750GB SATA disk drives yielding 144TB of raw storage in a single cabinet footprint, which would require less than 52,560 kWh and cost about $10,512 per year with an emissions footprint of about 39.42 CO2 tons. To account for cooling costs, simply double the above numbers for a worst-case scenario. By comparison, a 1999 Chevrolet Tahoe generates about 7 to 10 tons of CO2 per year, while a Lexus RX333, depending on miles driven, generates about 5-6 tons of CO2 per year, and a 24 cubic foot refrigerator yields about 1.22 tons of CO2 per year.
Vendors are taking different approaches to power and cooling, with some like HP leveraging their services organization to perform energy audit assessments to tune and reduce cooling requirements. By reducing cooling requirements, HP has demonstrated in its own data centers that it can achieve quick reductions in energy consumption. Sun has partnered with PG&E to help its server customers gain energy rebates for using more energy efficient servers. Big Blue (IBM) recently launched its far-reaching "Big Green" initiative, while storage networking switch vendor Brocade spent last week trumpeting its current energy advantage over rival Cisco.
You can expect to hear more about green storage and the various aspects of reducing power consumption and heat output. Vendors are taking several different approaches to achieve energy efficiency and eco-friendly storage, including energy-saving techniques, management software capabilities, storage efficiency and storage disposal. Needless to say, this will not be the last you will be hearing about power, cooling, CO2 emissions and other topics associated with storage, since we are just scratching the surface with this "hot" topic.
We'll devote more time to this important subject in a future article as well as a free webcast on June 4.
Greg Schulz is founder and senior analyst of the StorageIO group and author of Resilient Storage Networks (Elsevier).