Projects
Projects: Summary of Projects by RegionProjects in Region Scotland involving University of Edinburgh : EP/G058989/1 |
||
| Reference Number | EP/G058989/1 | |
| Title | Adaptable Processor Architecture and Software for Energy-Efficient Computing | |
| Status | Completed | |
| Energy Categories | Energy Efficiency(Residential and commercial) 10%; Not Energy Related 90%; |
|
| Research Types | Basic and strategic applied research 100% | |
| Science and Technology Fields | PHYSICAL SCIENCES AND MATHEMATICS (Computer Science and Informatics) 100% | |
| UKERC Cross Cutting Characterisation | Not Cross-cutting 100% | |
| Principal Investigator |
Dr C Dubach No email address given Sch of Informatics University of Edinburgh |
|
| Award Type | Standard | |
| Funding Source | EPSRC | |
| Start Date | 01 July 2009 | |
| End Date | 30 June 2014 | |
| Duration | 60 months | |
| Total Grant Value | £1 | |
| Industrial Sectors | No relevance to Underpinning Sectors | |
| Region | Scotland | |
| Programme | Information & Communication Technology | |
| Investigators | Principal Investigator | Dr C Dubach , Sch of Informatics, University of Edinburgh (100.000%) |
| Web Site | ||
| Objectives | ||
| Abstract | Energy consumption is now a major concern for both mobile devices and high-performance data centres. Consumer electronics requires more powerful and versatile mobile devices with extended battery life, while businesses increasingly rely on data centres to run their services. Given this trend and the rising cost of energy, it will become increasingly unsustainable to run such centres. The challenge is to develop new techniques that span the entire range of computer systems and enable truly energy-efficient computing.This proposal aims at investigating novel ways of designing energy-efficient systems by focusing on the core element of these systems; the microprocessor. The key contribution is the development of adaptive software and hardware driven by machine-learning techniques to fit the dynamic behaviour of the application. By accurately predicting the future, we can specialise the software or hardware for greater efficiency. To reach this goal, three main themes will be explored within this proposal: "Adaptable Architecture Design", "Program Modelling and Runtime Software-DrivenAdaptivity" and "Compiler Optimisation and Hardware Interaction".If successfull, each of these themes will lead to major breakthroughs in related research areas. This will bring benefits to both mobile devices and data centres. This will be made possible by using machine-learning in the design process of such adaptable architecture and to control the dynamic adaptivity of both software and hardware | |
| Data | No related datasets |
|
| Projects | No related projects |
|
| Publications | No related publications |
|
| Added to Database | 21/09/11 | |
