As of today, in the internet and communications technology (ICT), isolated single-optimized server systems are often suspect of unused capacities and hence a high energy demand. The project AC4DC plans to develop methods to reduce this energy demand by optimizing the operation of ICT systems as a whole, including end users, office PCs, data centres, data networks and energy providers in terms of energy demand and total costs.

The holistic view of the system not only creates extensive possibilities of energy saving but also enables a load optimization in power networks, especially in smart grids.

To accomplish the project's aims, intelligent forms of management of load in data centres, infrastructure and data storage will be researched. These will be applied to single data centres as well as to a cluster of multiple data centres.

This three-step approach partially enables the parallel achievement of the project's aims. The concepts developed will then be implemented as prototypes and tested in partly isolated environments. This is then used to evaluate the energy saving potential.

The optimization will also reflect influences such as user behaviour, economic requirements and site-specific parameters. The systematic and early involvement of data centre operators assures the project's market-orientation. In addition highly-effective business models and potentials to save energy and material will be identified.

OFFIS is significantly involved in multiple work areas of the project. Analysis of existing data as well as the accomplishment of new measurements in data centres are an essential part to be able to make statements about the energy demand of servers and complete data centres. This also includes the development of new measurement technologies.

In addition new models will be developed and evaluated which allow for a proactive controlling and sharing of services across servers. These models will also take into account the cooling systems for servers and data centres.

Finally an implementation of a prototype of the system will be developed and used to prove that the results are also relevant in practical use.