Green computing

Under Green IT ( rarely also Green ICT ) refers to efforts, the use of information and communication technology (ICT ) throughout their entire life cycle to make environmentally and resource friendly. This involves the optimization of resource consumption during manufacturing, operation and disposal of equipment ( in Green IT).

Another aspect of Green IT is the energy saving through the use of information technology ( IT), such as when missions are replaced by video conferencing ( Green by IT).

5.1 Refurbished IT

Overview

The data used above terms " sustainability " and "green " but are blurred. Therefore, the GreenIT includes a variety of approaches that address different aspects of the life cycle of ICT.

These include:

Reduction of energy consumption in the use
Reduction of energy - and material use in the manufacture
Reduction of waste heat and pollutant emissions in the use and production
Reduction of contaminants in the products and in the production processes
Recycling and energy-saving disposal
Sustainable design of products and manufacturing durable as possible Hardware
Resource-saving programming software ( Green Software Engineering )
Use of IT to reduce energy consumption another source ( such as transport, heating systems )
Social and ethical aspects, such as healthy and fair working conditions (fair IT)

For political and image reasons, the operation of IT equipment is often viewed with renewable energy sources, such as by Greenpeace

Significant legal standards in the EU RoHS Directive and the WEEE Directive.

History

The beginnings of Green IT are in the year 1992 as by the U.S. EPA, the Energy Star label was launched, on the other hand, the Green IT also roots in the environmental computer science, which of themselves with the use of IT for environmental purposes or use IT has on the environment. Here especially the ETH Zurich and the affiliated EMPA Institute has been active since the 1980s.

Areas of application in industrial use

Data centers

Most of the electricity consumption of ICT is generated in the commercial use of data centers and servers. She spent 2008 in Germany 10.1 terawatt hours of electricity, 1.8 percent of Germany's total consumption, and thus generated approximately 1.1 billion euros costs. The energy density of the centers and thus their cooling requirements are steadily increasing. Studies forecast for Germany a growing demand for electricity by up to 50 % by 2013 and identify major savings through the use of Green IT (2008 ).

Here, depending on the average estimate up to half of the energy requirements for the operation of infrastructure such as uninterruptible power supply (UPS), power distribution, cooling or similar required. Measures to deal with their optimization also called DCIM (Data Center Infrastructure Management) solutions. A rough estimate for the assessment of the cost of the infrastructure provides the Power Usage Effectiveness ratio ( PUE ). In individual cases, the energy expenditure for the infrastructure can largely differ from the average, because various factors, such as high ambient temperatures, restricted construction areas or high security requirements of the data center (if UPSs are duplicated for example) increase this even with modern technology. Thus, a significant higher increase in power consumption is predicted by the shift of computing power in large data centers, because the security requirements by its location in an earthquake zone are significantly higher than in Central Europe for Japan.

A common guideline states that the current round-the- clock server are on average operating at only 10 to 20 percent. Usually one lays out the server capacity using operational tips, but these are rarely achieved. A concept through to Green IT is here the consolidation: This heterogeneous systems are merged and the number of servers and data centers reduced. Another option for data centers with better capacity utilization is virtualization. This allows application programs that ran previously scattered on different machines, bundling in virtual machines on powerful computers. As a result, servers can be much higher, approximately up to 50 percent, are utilized.

Is supported virtualization concepts such as service-oriented architectures (SOA) and Software as a Service (SaaS). SOA down business processes by computing power, which can be the necessary resources in data centers pinpoint. SaaS application programs shifted from local computers to central high-performance machines and uses there the potential for resource sharing. Modern IT management systems can actually required server capacity to predict and curb services as needed from or switch.

The IT service provider, Siemens IT Solutions and Services is that they have increased the utilization of in-house data centers by consolidating and virtualizing more than 80 percent. The total energy consumption had decreased by over 30 percent. At the same time, the number of data centers in Germany was reduced from 70 to 30. In May 2008, the company opened a new data center in Beijing, which houses the infrastructure of all IT and telephone services from over 70 Siemens sites in China and Mongolia. According to Siemens IT Solutions and Services, the servers consume about 37 percent less energy than standard server.

Another important aspect of Green IT, which offers high potential for optimization, the data center cooling. Between 35 and 50 percent of the total energy costs of a data center is spent on cooling. Through modernization of measuring and cooling technology and by implementing structural measures (eg arrangement in hot and cold aisles, modular design of the cooling technology ), this proportion may, however, be greatly reduced.

However, the correct choice of optimization measures should always be oriented to the status quo of an existing IT environment. For this reason, each " Cooling Cosmetics " in the sense of Green IT, a comprehensive analysis of the thermal actual states in the data center should be preceded. This should an IT infrastructure, such as the air flows over and in the double bottom, cooling air losses, the efficiency of cooling systems, the heat development on the racks are measured in depth to demonstrate such savings can. Based on an analysis of this data then a meaningful action plan should be developed that quantifies precisely the expected savings for each optimization step, or at a mesh of several thermal corrections.

Communication network

Currently, the volume of traffic increases in the ICT networks by 50 % to 100 % per year. This growth will continue over the next 10 years. This is associated with the same time an increase in the energy consumption of approximately 16% to 20 % per year. There are estimates that the ICT equipment and facilities are currently responsible for 2 % of global CO2 emissions. Other estimates come to the conclusion, that this size is closer to 3%.

At the annual traffic growth of 50 % to 100 % in the ICT networks the proliferation of mobile applications has a significant share. This has to do with the nationwide expansion of mobile networks and the development of new services with higher data rates and improved service qualities as well as the development of completely new fields of application together. In particular, developing economies rely on mobile technology due to the mostly non-existing fixed network infrastructure, low investment costs and rapid network expansion.

With regard to the actual network technology must be especially considered in the context of energy efficiency effects caused by the expansion of broadband systems. In Germany we have the following situation:

The copper access technology with its widely branched network to each customer dominates the growth in energy demand. For example, a nationwide installation of VDSL in Germany would produce a power requirement of 450 MW with today's technology.
For the entire mobile broadband network devices including a demand of 520 MW is forecast for 2010. The vast majority of this is caused by the mobile base stations. The associated transport networks are estimated to contribute 25 % to the energy demand. The mobile terminals are practically negligible despite their high number in their energy consumption compared to other components.
Increasing energy demand arises from corporate networks, that is, by the higher bandwidth requirements of the Local Area Networks ( LAN) and the operation of high -speed LANs. First estimates forecast a demand of over 150 MW of such networks in 2010.
Currently, about 50 % of private broadband connections to a wireless LAN router (WLAN ) are operated, thereby already an energy requirement of about 100 MW caused today.
The share of long-distance technique (backbone routers and WDM technology) the total energy requirement of network technology is below 15%, but due to the high integration of the systems, considerable problems due to high energy densities that ultimately only by expensive air conditioning ( which in turn also requires a lot of energy) manageable be.

(see also link to VDE position paper below). See also: Energy Efficient Ethernet

Office applications

The 26.5 million desktop computers in Germany in 2010 consumed in operation about 3.9 terawatt hours of electricity. During the fabrication of these devices fell to a further 9.3 TWh. In this field of application further efficiencies may arise through appropriate developments. Besides the direct optimization of hardware components with regard to power and material consumption, new technologies can also enable other applications of IT. With desktop virtualization can resource-intensive desktop PCs are replaced by simpler thin clients. The lack of processing power of the thin clients is centrally offset by a data center. This is also called Thin Client & Server Based Computing ( TC & SBC). Such data centers can "produce" the resulting power depending on the load and save as compared to desktop PCs energy because they can respond relatively poorly to idle times. The user realizes the ideal case of this change nothing. The physical delivery of computing power can in principle be carried around the world ( Cloud Computing ). In addition, may be extended by TC & SBC the useful life of the devices. In the " classical" solution laptops or desktop PCs are often discarded due to the low processing power, not on the basis of material fatigue or damage. The transfer of computing power in a data center just thin clients remain on the current state of the art, and so have the potential to reduce the resource consumption of the IT production.

Other applications of the use of video conferencing instead of business travel and the IT control, air conditioning or lighting of office buildings belongs.

Operating Systems

The OS vendor Microsoft has been criticized that the Windows operating system does not have sufficient mechanisms to deal efficiently with energy. Due to the high market share of the operating system, this could have a large impact on the energy use worldwide. According to Microsoft, this situation has changed since Windows Vista, but this is controversial. This problem is largely due to the fact that Windows is no central management of energy management supported up to Vista version by the system administrator. This fact is due to that many companies do not optimize their energy usage on individual computers.

With the introduction of Windows Vista, this failure was partially solved by the introduction of a central energy management. The new energy management will be criticized because of its inflexibility. The main criticism is the lack of ability to dynamically adjust the power management to the current needs. Accordingly, there is on the market some software products that fill this gap.

Application Development

The architecture of software applications has a significant impact on the demand for electricity. Just browser- based applications ( web applications, Rich Internet Application ) will be drastically different depending on the used architecture. GWT World estimates that server up to 50 times longer able to serve clients when modern Ajax frameworks are used instead of classical web architectures. Modern Ajax frameworks allow power saving, but less powerful clients to use, without the user having to wait any longer. See also: efficiency ( computer science )

Application areas in the private use

Households in 2007 were responsible in Germany for about 60 percent of electricity consumption of ICT, a total of 33 terawatt hours. Of this amount, 11.2 TWh to computer itself ( including peripherals ). The largest share have televisions with 15.8 TWh.

Consumers through their purchasing behavior not only short-term cost savings and energy, but also encourage the further development of more efficient appliances. Even the everyday use of these devices is still room for improvement. So still a lot of energy is lost unnecessarily because devices overnight, weekends, or on vacation, not turned off are left on standby. During the period of use the standby function ensures that the energy requirement is automatically throttled when the device - such as a printer, copier or fax machine - just gets no job and therefore " offset even in deep sleep ." However, still a significant amount of energy can be saved when the unit is completely disconnected from the mains when it is not constantly needed. As an example, a printer is, of the ( so if he waits for data to print ) consumed during operation ( ie printing) 20 watts in standby mode at least another 6 watts. Even in the fully off state still 4 watts are consumed.

For home users over a quarter ( 28 %) of the total annual electricity consumption of IT is produced by the stand-by mode, a total of 9.462 billion kWh, also have TV with 3.714 billion kWh for the largest share.

Other areas of application

Traffic management: traffic processes are improved, reducing congestion and emissions. Particularly busy zones can be disabled or restricted to traffic. In addition, the city can make public transport more attractive for citizens through a better traffic management system.
IT systems for the management of power plants with different power generation plants (several lignite-fired power plants, decentralized power generation plants ): The power plants are making it more efficient and resource-saving.
Smart Metering by means of smart meters that automatically measure the cost of electricity and bill. The display lets customers see in real time how much electricity he just where used and how much does it cost him. This transparency can contribute to more economical use of energy.

Refurbished IT

This is IT (mostly computers, notebooks and servers), for example, come from terminated leases. These are reconditioned by the manufacturer or by a dealer, cleaned, tested, and finally re-sold as a used device. Since this usually is business equipment, they can be used due to the high quality components installed without problems for some time. Refurbishing is therefore an environmentally friendly method, existing devices to be used again so as to save resources.

Weaknesses of the approach

Greenwashing The term may be filled by any interested companies for their marketing strategies because, although several protected trademark rights have been defined, but testable criteria for "green" characteristics of products are not defined. Users of the concept therefore run the risk of carrying private marketing strategies of the brand owner unchecked (see also greenwashing ).
Rebound effects through the use of fuel-efficient computers and servers it may happen that the equipment per day are used for longer or more devices that are purchased and used. These effects can make and occur within companies / households saving any efforts to naught, as well as at national and global levels.
Miscellaneous Any IT unless it is used for entertainment purposes per se "green". For example, accounts through online banking trips to the bank, or by e- mail letters can be saved, by electronic bookkeeping much paper is saved and many more savings more.