Increasingly, energy and power turn from optimization criteria into design constraints, sometimes as critical as, for example, reliability and timing. Furthermore, quanta of energy or specific levels of power can shape the system's action. In this direction, I propose to develop a systematic approach to designing computing systems that are energy-modulated and power-adaptive. This will cover systems where power comes from energy harvesting sources with limited power density and/or unstable levels of power. Initial studies have been performed at Newcastle that include self-powered sensors, variation-resilient logic and memory operating in the dynamic range of Vdd 0.2-1V, and system design techniques for power-adaptation.
Communications
The presentation of the communications of the CHIST-ERA Conference 2012 (keynote and short talks, posters) will be continuously updated in the course of June.
Prof. Alex Yakovlev is professor of computer system design at the University of Newcastle, United Kingdom, and head of the Microelectronics System Design Research group and the head of Signals, Systems and Communications Research group of the Electrical, Electronic and Computer Engineering School.
Poster
Attachment | Size |
---|---|
![]() | 1.02 MB |

Alexander Kipp is deputy head of the Intelligent Service Infrastructures (ISIS) department of the High Performance Computing Center Stuttgart (HLRS) in Germany.
Energy-consumption in IT service centres is gaining a significant attention. Whilst current approaches mainly concentrate on the sensible management of the IT infrastructure, an open, not yet addressed issue is the impact of the application structure to the entire energy consumption. Therefore, we suggest to combine the measurement and steering of IT infrastructures with a structure analysis of applications and services. This proceeding allows to optimise existing software by providing suggestions about energy-wasting code fragments, as well as to provide an extensive support during development and operation.
Poster
Attachment | Size |
---|---|
![]() | 3.76 MB |

Dr. Alexandru-Adrian Tantar and Dr. Emilia Tantar are both post-doctoral researchers at the Computer Science and Communications Research Unit, University of Luxembourg.
In the effort of attaining Carbon-neutral ICT operations at the University of Luxembourg two research directions are followed: energy efficient solutions for the management of data centers and dynamic energy management of the university's facilities. For this purpose a prototype laboratory is about to be developed, aimed at studying strategies that are capable to reduce energy consumption, in a dynamic manner. All the approaches analyzed at this point rely on dynamic paradigms that are capable to deal with different time dependent and stochastic factors, e.g. as determined by the use of renewable energy sources, evolving constraints and requirements.
Poster
Attachment | Size |
---|---|
![]() | 980.85 KB |

Dr. Daniela DRAGOMIRESCU is associate professor at the French engineering school INSA Toulouse. She is part of the Micro and Nanosystems for Wireless Communication group lead by Prof. Robert Plana at LAAS-CNRS laboratory, Toulouse, France; where she is in charge with the Wireless Sensor Network team.
This poster will present the design of an energy efficient architecture for wireless sensor networks (WSN) based on cross layering between MAC layers and PHY layer. To obtain an energy efficient system for WSN it is necessary to change the design paradigm. A new paradigm of cross-layering is necessary and a design from a complete system point of view (routing-software, MAC, PHY layers=hardware, together) is demanded and will be presented. PHY layer use digital IR-UWB ultra low power architecture and design. Digital approach is very important to be able to do cross layering and obtain energy efficient systems. MAC layer is adapted to the IR-UWB PHY layer to obtain an energy efficient system. The advantage of reconfigurability of PHY and MAC layer is used to obtain an energy efficient wireless sensor network. Software layers, like routing (depending on the energy available on the node) are also discussed in this cross-layering design from an energy efficiency point of view.
Poster
Attachment | Size |
---|---|
![]() | 438.94 KB |

Dr. Gunnar Schomaker is within the R & D Division for Energy of the OFFIS Institute for Information Technology, Germany, in charge of the group Energy Efficiency in ICT.
The project AC4DC will develop proactive load and power management methods and software prototypes to reduce the data center energy demand. The aim is to 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 a system of load as power dependent energy consumers not only creates extensive possibilities of energy savings, but also enables a load optimization in power networks, like smart grids.
Poster
Attachment | Size |
---|---|
![]() | 832.1 KB |

Prof. O'Connor is professor at the French engineering school Ecole Centrale de Lyon in the Department of Electronic, Electrical and Control Engineering. He is also head of the Heterogeneous Systems Design group at the Lyon Institute of Nanotechnology, of which he is one of the vice-directors.
The emergence of new research devices based on nanotubes (CNFET) or nanowires (NWFET), offers the opportunity to provide novel logic building blocks, to explore new possibilities for digital design and consequently to reconsider the paradigms of computing architectures to achieve orders of magnitude improvements in conventional figures of merit. This poster will examine circuit technologies using regular structures based on ambipolar CNFETs or NWFETs, and how logic functions can be mapped onto them, particularly in the context of reconfigurable applications.
Poster
Attachment | Size |
---|---|
![]() | 844.78 KB |

Matteo Agostinelli is PhD candidate and works in the Institute of Networked and Embedded Systems at the University of Klagenfurt (Austria) in cooperation with Infineon Technologies and Lakeside Labs.
The need to improve the energy efficiency and dynamic performance (e.g. due to load variations) of DC-DC converters has apparently brought interest in innovative control structures, alternative to conventional linear PID schemes. A few control techniques will be presented and evaluated on mobile applications, such as for smartphones or laptops. The choice criteria of an analog or digital implementation will be also discussed.
Poster
Attachment | Size |
---|---|
![]() | 324.83 KB |
Dr. Süleyman Tosun is assistant professor and researcher in the Computer Engineering Department of the Faculty of Engineering at Ankara University, Turkey.
Network-on-Chip (NoC) is an alternative approach to traditional communication methods for System on Chip (SoC) architectures. Irregular topologies are preferable for the application specific NoC designs since they offer huge optimization space in contrast to their regular counterparts. Generating an energy efficient application specific topology as part of the synthesis flow of a Network-on-Chip architecture is a challenging problem since there may be several topology alternatives, each of which may be superior to the others based on the different objective criteria. In this poster, we present a heuristic and a genetic algorithm based methods for the stated problem. The heuristic method, called TopGen, is a two-phase application specific topology generation algorithm aiming to minimize the energy consumption of the system. TopGen first decomposes the given application into clusters based on the communication traffic. It then maps the clusters onto the routers and connects them in such a way that the communication cost of the network is minimized. The second algorithm, called GATGA, is based on a genetic algorithm, which initially creates a set of solutions and uses genetic operators to reproduce new topologies from them. In this poster, we finally give possible research topics related to energy efficient irregular topology generation for NoC architectures.
Poster
Attachment | Size |
---|---|
![]() | 295.94 KB |
- 1 of 2
- next ›