Collaborative and Autonomic Green computing track: Summary report

摘要

This paper introduces the CAGing track. Topics, program committee, selected papers are explained. Green computing is nowadays a major challenge for a large community. Datacenters consume around 2% of global electricity, this consumption represents 200 billion kWh [1]. Electricity use per server could range from 50W when it is idle from 250W when it is full loaded [2]. The Green500 [3] provides a ranking of the most energy-efficient supercomputers in the world. [4] presents a complete description of the energy comsumption in data centers. Energy consumption has a cost and also an ecological impact. It is then taken into account when anticipated environmental changes are studied. Energy efficiency has been more and more considered by industries, institutions and government. [5] has presented the commercial implications for Data Centre industry by country. The integration of green preoccupations can be done at different levels: hardware, network, operating system, applications, etc. Many works in the litterature address energy management techniques to reduce the energy consumed, the aim is to have a trade off between energy consumed and performance. In parallel, the architecture of computer systems and software is becoming more and more complex and distributed. Grids and clouds are good example of those trends. The size and complexity of such a system can no longer be managed at the human level or with scripts. The use of collaborative methods and autonomic computing [6] could solve many problems: configuration, performance, safety, availability, scalability of solution, accuracy of the solution, etc. This track proposes to focus on the use of collaboration and autonomic approaches in order to adapt, react and adjust entities behavior in order to save energy. Section II presents the track, then section III presents the organisation and to finish selected papers are introduced in section IV.