Technology Blogs by SAP
Learn how to extend and personalize SAP applications. Follow the SAP technology blog for insights into SAP BTP, ABAP, SAP Analytics Cloud, SAP HANA, and more.
cancel
Showing results for 
Search instead for 
Did you mean: 
Former Member

The sustainable management of IT workplaces is important for a modern (eco-friendly) IT strategy. From selecting suitable hardware through recycling and disposal, it is vital to calculate and set standardized and measurable targets. But how can measurement methods contribute to reducing energy consumption?


The number of work center computers installed at German companies has steadily risen to 32 million. As a result, the energy efficiency of IT systems (data centers in particular, but also IT workplace devices) is an increasingly important part of a sustainable enterprise strategy.

Fortunately, however, devices have become significantly more energy-efficient – although, similar to the situation with automobile traffic, not sufficiently energy-efficient to reduce total energy consumption.

During normal operations, the power consumption of high-performance high-end notebooks has increased due to the extremely high computing power. A reduction in power consumption can only be detected with minimal use. Modern TFT monitors are considerably more energy-efficient than five years ago. Some electricity is used in sleep mode, too. As a rule, it is therefore advisable to turn electronic devices off, rather than to put them in standby mode.

As is many other sectors, the manufacturers’ data is usually gathered under laboratory conditions and therefore does not always reflect actual energy consumption.

Most users and companies do not generally know how much energy certain models consume in day-to-day use.


Measuring power consumption: Power management software solutions usually only give an indication

It is relatively difficult to regulate energy consumption through the operating system’s energy settings.

For this reason, various providers offer software solutions for PC power management. The more advanced among these enable network administrators to record the energy savings achieved and to adjust power management to the individual needs of the users by, for example, automatically setting the computer to standby and waking it up again as required.

However, implementing such a solution is – depending on the company’s size and number of employees – sometimes a costly investment.

Furthermore, the underlying parameters are, as a rule, universally defined and only provide an indication, which is often far removed from the actual power consumption.

The big question is therefore: How can you determine the energy consumption of an IT work center yourself?


Initial assessment: prepare well so you can measure as accurately as possible

To initiate your own power management project, the first thing you must get right is the design and planning.

First, the energy consumption of the computers within the company needs to be analyzed. To develop a reliable measurement method, it is important to record the different device types in a company, in order to create the basis for determining the main energy consumers.


  • Which PCs, notebooks, and monitors models are used?
  • How much power do these devices use and what is the age structure?
  • How many of these devices does the company have?
  • What impact do mobile end devices and IT accessories have?
  • In addition, you should determine which user groups there are in the company and how they use the devices.
  • Have all the employees set the same power settings – if they have set any power settings at all?
  • Are different software programs in use?
  • When do the computers run (just eight hours a day or do PCs and laptops run 24 hours a day)?
  • What is the relation of mobile users (sales, consulting, and so on) to on-site users (administration, development, and so on)?


Once you have answered these questions, you can identify different representative activity groups and device groups.

Distribution of IT work center devices

Actually measured average consumption for each device category

Power consumption for each device type. External monitors, desktop PCs, and laptops are responsible for 98% of power consumption, with external monitors alone responsible for 55%.

The analysis: time to get measuring

Now, an optimal test group can be formed to attain average values that are as representative as possible.

Typical test candidates in an IT company are:

Employees who frequently work away from the office, such as sales or consulting employees each equipped with a very lightweight notebook, a tablet, and a smartphone

Developers with a powerful PC, two screens, and a smartphone

It is important that the test group reflects the user group proportions as accurately as possible and that the most-used device types (which should account for approximately 90% of the inventory) are represented.

The test is performed during normal daily operations using an ammeter. The use of a network-aware ammeter, which transmits data over SNMP and saves it to a server, is recommended here.

The results of the measurements are then used to set up a database.

The permanent use of such ammeters at all work centers in the company would, in theory, be the most precise method. However, it is not cost-effective.


The evaluation: consolidating and assessing the measurement results


The consumption values for the various power settings (such as idle and productivity), which are stored in the database, now serve to enable energy consumption to be extrapolated easily.


Example of evaluating the measurement results for PCs/notebooks. The models/manufacturers named here are purely examples. The SLEEP and OFF values are taken from the manufacturers’ technical data sheets and then included in the subsequent calculation.

At this point, it is worth comparing the information from the manufacturer with the measurement results, because, depending on each user’s behavior, differences of several per cent may arise. By making this comparison, you gain an impression of the size of the deviation and can therefore draw on these results in the future.

With the findings gained from the measurements, you can now derive the following simplified formula to determine energy consumption.


Energy consumption (E) =

Performance “average value of measurements” (PON,avg,model_PC) *

Active hours a month (WH4weeks) +

Average value from the power states “OFF” and “SLEEP” based on manufacturer’s information (POFF,avg,modelPC) *

Inactive hours


Applied to an older-generation notebook, the formula is: Energy consumption (E) = 34.83 W * 140 h + 1.24 W * 532 h = 5535.88 Wh

The example refers to a calculation period of four weeks and an active operation time of 35 hours a week (40 hours, with one hour’s break each day). An alternative would be to take a statistical average value as a basis (WH4weeks * active ratio) or, if available, to use technical records from a system management tool such as SCCM.

The calculation of the monitors’ energy consumption is performed in the same way.

Example of evaluating the measurement results for monitors. The models used are examples only.

Tablets and smartphones can be included in the calculation using the number of charging cycles.

Because the energy consumption of accessories such as headsets, keyboards, and mouse devices is very low compared with PCs, notebooks, and monitors, they are not considered in the calculation. However, it would be possible to include them based on information from the manufacturer.

To find out the total energy consumption of a company’s IT work centers, the following method is recommended: You apply the formula to the devices in the test group to determine consumption for each device type/manufacturer. In the last step, you multiply the result by the use/non-use in hours and the number of devices in the company.


The measures: reducing energy consumption long-term


The findings from the analysis can serve as the basis for setting up your own database with characteristic energy consumption values, but they can also be used to scrutinize or optimize power management software solutions that have been deployed or a Windows power plan that has been defined.


You gain a sense of the extent to which the manufacturer’s information is based on real working or use situations, or the extent of discrepancies in this respect.


You soon recognize that replacing old devices and seldom or never used devices and accessories can be a meaningful way of reducing the power consumption of an IT work center environment or of contributing to the intelligent use of work centers. As a result, the life cycle of an employee’s work center can be optimized in the long term.


Of course, this does not mean that old devices should be banished from the workplace in general, because energy and resources are also used up in production and disposal. However, you could, for example, consider moving older functioning computers to work centers that are used less frequently. An employee who is constantly on the move, such as a sales employee who is almost exclusively at the customer, could be assigned an older monitor at his or her rarely used on-site office work center, for example.

In general, transparency is the first step to more sustainable behavior.


  • Let your employees have a share of the action
  • Actively call upon employees to become involved in the measurements
  • Publish the test results and findings internally on your intranet or internal social media channels
  • Organize information events
  • Publish simple tips to save energy
  • Provide examples to make the meaning of a watt clear
  • Only when employees know the impact that their behavior has, can an awareness of the topic be created and their behavior changed in the long term.


And let me take this opportunity to dispel a common myth: For everyone who didn’t know – animated screen savers do not save power. A long time ago, the purpose of screen savers was to prevent phosphor burn-in on old CRT computer monitors. This is not necessary with modern TFT screens and screen savers can even consume more power than normal computer operations, for example, if a graphics-intensive 3D animation is installed. Putting the screen in standby mode or turning the picture off after a short period are better alternatives here.


And lastly: our energy resources are limited

Using an IT system optimally and effectively while reducing energy consumption to a minimum is a complex challenge to which every modern organization should rise – because economics and ecology go hand-in-hand here: Such measures can be marketed well, save money by cutting the amount of energy required, and thus protect the environment.


How do you determine the amount of energy a work center consumes in your company? What other action were you able to take?


I look forward to your questions and to engaging with you.

1 Comment