For more than 50 years, Auerbach Publications has been printing cutting-edge books on all topics IT.

Read archived articles or become a new subscriber to IT Today, a free newsletter.

This free newsetter offers strategies and insight to managers and hackers alike. Become a new subscriber today.



Interested in submitting an article? Want to comment about an article?

Contact John Wyzalek editor of IT Performance Improvement.


Electric Energy and Applicable Metrics

Daniel Minoli

A key factor in the design of green networks/green IT (or any green system for that matter) is energy consumption in general and electrical energy consumption in particular.

Two key units in this context are volts and watts. Voltage is the difference in electrical potential between two points; 1 watt (W) is the unit of electrical power, equal to the flow of 1 amp (A) at a potential difference of 1 volt (V); also equal to 1 Joule per second (J/s) and to 1/746 of 1 horsepower. Distribution is the process of delivering electric energy from points on the transmission system (usually a substation) to institutional and end-user consumers. Distribution uses a grid of conductors and equipment that distributes, transports, and delivers electricity to customers. Electric energy is carried at high voltages along transmission lines; electric energy is, in turn, reduced in voltage at a substation and delivered over primary distribution lines. The distribution line is comprised of one (or more) circuit(s) of a distribution system that is (are) direct-buried, placed in conduit, or strung on poles, or a combination thereof. The costs to support, operate, and maintain the local delivery system of wires and equipment that carries electric energy from the transmission system to the customer's premises are included in the rates and are usually priced in cents per kilowatt-hour for energy-only customers and in dollars per kilowatt for demand-billed customers. This is also known as Distribution Service.

The Electricity Service is the network of generating plants, wires, and equipment needed to produce or purchase electricity (generation) and to deliver it to the local distribution system (transmission). This service is priced in cents per kilowatt- hour for energy-only customers, and in dollars (euros, etc.) per kilowatt and in cents per kilowatt-hour for demand-billed customers.

Some related nomenclature follows [MGE200901]:

Energy Charge: That part of the charge for electric service based upon the electric energy (kWh) consumed or billed.

Energy Costs: Costs, such as fuel, related to and varying with energy production or consumption.

Energy, Electric: As commonly used in the electric utility industry, it means kilowatt-hours. Energy, Off-Peak is the energy supplied during periods of relatively low system demand as specified by the supplier. Energy, On-Peak is the energy supplied during periods of relatively high system demand as specified by the supplier.

Fuel Cost Adjustments: A provision in a rate schedule that provides for an adjustment to the customer's bill if the cost of fuel at the supplier's generating stations varies from a specified unit cost.

Power, Firm: Power or power-producing capacity intended to be available at all times during the period covered by a commitment, even under adverse conditions.

Power, Interruptible: Power made available under agreements that permit curtailment or cessation of delivery by the supplier.

Power, Nonfirm: Power or power-producing capacity supplied or available under an arrangement that does not have the guaranteed continuous availability feature of firm power. Power supplied based on the availability of a generating unit is one type of such power.

Primary Voltage: The voltage of the circuit supplying power to a transformer is called the primary voltage, as opposed to the output voltage or load-supply voltage, which is called secondary voltage. In power supply practice, the primary voltage is almost always the high-voltage side, and the secondary is the low-voltage side of a transformer, except at generating stations.

Service Area: Territory in which a utility system is required or has the right to supply electric service to ultimate consumers.

Service, Customer's: That portion of conductors usually between the last pole or manhole and the premises of the customer served.

Service Drop: The overhead conductors between the electric supply, such as the last pole, and the building or structure being served.

Service Entrance: The equipment installed between the utility's service drop, or lateral, and the customer's conductors. Typically consists of the meter used for billing, switches and/or circuit breakers and/or fuses, and a metal housing.

Service Lateral: The underground service conductors between the street main and the first point of connection to the service entrance conductors.

Single-phase service: Service where the facility has two energized wires coming into it. Typically serves smaller needs of 120 V/240 V. Requires less and simpler equipment and infrastructure to support than three-phase service.

Step-down: To change electricity from a higher to a lower voltage.

Step-up: To change electricity from a lower to a higher voltage.

Substation: A collection of equipment for the purposes of switching and/or changing or regulating the voltage of electricity. Service equipment, line transformer installations, or minor distribution and transmission equipment are not classified as substations.

Three-Phase Service: Service where the facility has three energized wires coming into it. Typically serves larger power needs of greater than 120 V/240 V.

Transformer: An electromagnetic device for changing the voltage level of alternating-current electricity.

Transmission: The act or process of transporting electric energy in bulk from a source or sources of supply to other principal parts of the system or to other utility systems.

Data center and telecom providers are interested in service reliability. In this context, reliability is the guarantee of system performance at all times and under all reasonable conditions to ensure constancy, quality, adequacy, and economy of electricity. Reliability includes the assurance of the continuous supply of electricity for customers at the proper voltage and frequency.

In most jurisdictions in the United States, public utilities operate as controlled monopolies but are subject to regulation. They are obligated to charge fair, nondiscriminatory rates and to provide reliable service to the public. In return, they are generally free from direct competition and are permitted to get a fair return on investment. The Federal Energy Regulatory Commission (FERC) is an agency within the U.S. Department of Energy that has broad regulatory authority. Nearly every facet of electric (and natural gas) production, transmission and sales conducted utilities, corporations, or public marketing agencies is under FERC control.

Demand represents the rate at which electric energy is delivered to (or by) a system; it is typically expressed in kilowatts at a given instant in time or averaged over any specified period of time. Average demand is the demand on an electric system over any interval of time (this is derived by dividing the total number of kilowatt- hours by the number of units of time in the interval.) Maximum annual demand is the largest demand that occurred during a (15-minute) interval in a year.

Consumption of electrical energy is typically measured in kilo-, mega-, or gigawatt- hour: a kilowatt-hour (kWh) is the basic unit of electric energy equal to 1 kilowatt (1,000 watts) of power supplied to or taken steadily from an electric circuit for 1 hour. One kilowatt-hour equals 1,000 watt-hours. Clearly, 1 megawatt-hour (MWh) equals 1 million (1,000,000) watt-hours; 1 gigawatt-hour (GWh) is 1 billion (1,000,000,000) watt-hours, or 1 million (1,000,000) kilowatt-hours, or 1 thousand (1,000) megawatt-hours. Also see Appendix 3B for some related parameters.

Load management, defined next, may be advantageous for the user. Load management is the reduction in electric energy demand during a utility's peak generating periods typically resulting in lower overall costs to the user. Load-management strategies are designed to either reduce or shift demand from on-peak to off-peak times (by way of contrast, conservation strategies may primarily reduce usage over the entire 24-hour period). Actions may take the form of normal or emergency procedures. Many utilities encourage load management by offering customers a choice of service options with various price incentives. Related concepts include the following:

Load Curve: A plot showing power (kilowatts) supplied versus time of occurrence that captures the varying magnitude of the load during the period covered.

Load Factor: The ratio of the average load in kilowatts supplied during a given time period to the peak or maximum load in kilowatts occurring in that period. (Load factor, in percent, is derived by multiplying the kilowatt-hours (kWh) in the period by 100 and dividing by the product of the maximum demand in kilowatts and the number of hours in the period.)

Load Shifting: Entails moving load from on-peak to off-peak periods to take advantage of time-of-use or other special rates; examples include use of storage water heating and cool storage.

The following concepts cover some of the typical billing-related factors related to electrical energy, based in part on reference [MGE200901]:

Base Rate: That part of the total electric rate covering the general costs of doing business unrelated to fuel expenses.

Conjunctive Billing: The combination of the quantities of energy, demand, or other items of two or more meters or services into respective single quantities for the purpose of billing, as if the bill were for a single meter or service.

Connection Charge: An amount to be paid by a customer in a lump sum or in installments for connecting the customer's facilities to the supplier's facilities.

Customer Charge: An amount to be paid periodically by a customer for electric service based upon costs incurred for metering, meter reading, billings, etc., exclusive of demand or energy consumption.

Demand, Billing: The demand upon which billing to a customer is based, as specified in a rate schedule or contract. It may be based on the contract year, a contract minimum, or a previous maximum and therefore does not necessarily coincide with the actual measured demand of the billing period.

Demand Charge: That part of the charge for electric service based upon the electric capacity (kW) consumed and billed on the basis of billing demand under an applicable rate schedule.

Demand Costs: Costs that are related to and vary with power demand (kW), such as fixed production costs, transmission costs, and a part of distribution costs.

Demand, Customer Maximum 15-Minute: The greatest rate at which electrical energy has been used during any period of 15 consecutive minutes in the current or preceding 11 billing months.

Demand, Instantaneous Peak: The demand at the instant of greatest load, usually determined from the readings of indicating or graphic meters.

Demand Interval: The period of time during which the electric energy flow is averaged in determining demand, such as 60-minute, 30-minute, 15-minute, or instantaneous.

Demand, Maximum: The greatest demand that occurred during a specified period of time, such as a billing period.

Firm Obligation: A commitment to supply electric energy or to make capacity available at any time specified during the period covered by the commitment.

Fixed Costs: Costs that do not change or vary with usage, output, or production.

Minimum Charge: A provision in a rate schedule stating that a customer's bill cannot fall below a specified level.

Primary Discount: A discount provision that is available to customers who can take delivery of electrical energy at primary distribution voltage levels. The transformer equipment discount is also available to customers taking primary voltage service who own their own transformers and transformer equipment.

Rate Case: The process in which a utility appears before its regulatory authority to determine the rates that can be charged to customers.

Rate Class: A group of customers identified as a class subject to a rate different from the rates of other groups.

Rate Level: The electric price a utility is authorized to collect.

Rate Structure: The design and organization of billing charges to customers.

Rates, Block: A certain specified price per unit is charged for all or any part of a block of such units, and reduced/increased prices per unit are charged for all or any part of succeeding blocks of such units, each such reduced/increased price per unit applying only to a particular block or portion thereof.

Rates, Demand: Any method of charging for electric service that is based upon, or is a function of, the rate of use, or size, of the customer's installation or maximum demand (expressed in kilowatts) during a given period of time such as a billing period.

Rates, Flat: The price charged per unit is constant, does not vary due to an increase or decrease in the number of units.

Rates, Seasonal: Rates vary depending upon the time of year. Charges are generally higher during the summer months when greater demand levels push up costs for generating electricity. Typically there are summer and winter seasonal rates. Summer rates are effective from June 1 through September 30. During all other times of the year, winter rates are effective.

Rates, Step: A certain specified price per unit is charged for the entire consumption, the rate or price depending on the particular step within which the total consumption falls.

Rates, Time-of-Use: Prices for electricity that vary depending upon what time of day or night a customer uses it. Time-of-use rates are designed to reflect the different costs an electric company incurs in providing electricity during peak periods when electricity demand is high and off-peak periods when electricity demand is low.

Retail Wheeling: An arrangement in which retail customers can purchase electricity from any supplier as opposed to their local utility. The local utility would be required to allow the outside generating company to wheel the power over the local lines to the customer.

Summer Peak: The greatest load on an electric system during any prescribed demand interval in the summer (or cooling) season.

Tariff: A schedule of prices or fees.

Tariff Schedule: A document filed with the regulatory authority(ies) specifying lawful rates, charges, rules, and conditions under which the utility provides service to the public.

Unbundling: Itemizing some of the different services a customer actually receives and charging for these services separately.

Variable Costs: Costs that change or vary with usage, output, or production. Example: fuel costs.

These concepts can all be applied to the analysis of the efficiency of a NE or data center component. A typical datasheet or hardware guide may include a number of energy metrics, such as [ECR200801]

Component-based Consumption Estimate (watts): This metric allows a customer to estimate a power draw in a "customized" configuration by adding together the configured parts (components) with known power ratings. This can be a fairly precise estimate, assuming the vendor has published accurate power numbers for all base, optional, and physical interface modules in the current board revisions. The availability and usability of such data is vendor dependent and requires a thorough knowledge of system structure and operation.

Maximum Power Consumption (watts): This metric can also be used as an upper boundary estimate for the power draw. However, this metric tends to penalize modular systems designed for the optional high-power components. For example, an Ethernet switch designed with Power over Ethernet (PoE) modules in mind will have much higher maximum power draw than a fixed copper- port model, yet both systems may yield identical consumption in a pure 1000 BaseT mode of operation. In addition, a maximum power consumption estimate can change without notice when new modules are introduced and old modules are withdrawn from production.

Power System Rating (amps or watts), aka Power Supply Rating: This metric reflects the site preparation requirements recommended by the vendor. It can potentially be used as an estimate for consumption, but with a possible error margin. In some cases, vendors outfit their platforms with high-capacity power supplies in planning for future system upgrades; the actual system consumption may be a fraction of what the power supplies can deliver.

Typical (Average) Power Draw (watts): This metric tends to underestimate the power consumption range. Motivated to demonstrate the low current draw, vendors are free to report this metric with underpowered configurations, components, or load profiles that yield the best results; omission of a published test methodology typically signifies these and similar issues. In the lack of a public disclosure on measurement conditions, "typical" or "average" power draw cannot be reasonably used to rate the device against any other platform.

Another metric of interest is the power availability metric. As implied earlier, availability refers to the percentage of time that a system is available, on-line, and capable of doing productive work. It is typically described as an annual percentage, or number of "nines." A system with "three nines" availability is 99.9% available, which translates into 8.8 hours of downtime annually. "Fives nines" availability— the standard many data centers aspire to—translates into less than 6 minutes of downtime annually. With power availability, an additional factor should be considered: the availability of conditioned power [LIE200901]. The key question is: Is it acceptable for systems being protected to operate on unconditioned utility power for short periods of time? Answering this question requires trading off the increased risk of operating on unconditioned power as a function of the added cost of UPS redundancy, which is required to minimize or eliminate the time protected equipment is exposed to utility power. Clearly, if the cost of downtime is low, the investment in redundant systems may not have a sufficiently high ROI (Return on Investment). If the cost of downtime is high, omitting redundancy could result in significant losses.


[ECR200801] ECR, Energy Efficiency for Network Equipment: Two Steps Beyond Greenwashing, Technical Report, August 10 2008, Revision 1.0.2, 10-08-2008.

[MGE200901] Madison Gas and Electric, Electric Glossary, Madison Gas and Electric, Madison, WI, 2009.

Read more IT Performance Improvement

This article is an excerpt from:

In recent years, socio-political trends toward environmental responsibility and the pressing need to reduce Run-the-Engine (RTE) costs have resulted in the concept of Green IT. Although a significant amount of energy is used to operate routing, switching, and transmission equipment, comparatively less attention has been paid to Green Networking. A clear and concise introduction to green networks and green network operations, Designing Green Networks and Network Operations: Saving Run-the-Engine Costs guides you through the techniques available to achieve efficiency goals for corporate and carrier networks, including deploying more efficient hardware, blade form-factor routers and switches, and pursuing consolidation, virtualization, and network and cloud computing.