Project description
New paradigms for embedded systems, monitoring and control towards complex systems engineering
New tools to reduce design cost of embedded systems
The latest and greatest embedded systems integrate a wide range of very complex functionalities. A smart phone, for example, is capable to communicate through 3G and WIFI connections. It integrates phone services with high performance graphics and sophisticated software applications such as real-time video and audio. To support this application load, new devices make use of recent parallel architectures capable to deliver enough processing power. But these changes have two negative effects. First, the development of parallel software, capable of exploiting multiple processor cores, is much more complex and therefore more expensive than traditional sequential software, which increases the product cost. Second, the increased quality of service requires more energy and hence is associated with a reduction of autonomy. The newest smart phones can run for at most about three days in standby mode while former simple phones did not need a recharge for an entire week. The PHARAON project will develop new techniques and tools that will offer the possibility to reduce the software development cost (25% targeted in the project) and increase the autonomy of embedded systems by nearly 20%.
For the academic partners the project will further strengthen their leading scientific position. For Thales, the main output of the project is the parallelisation of a waveform that will enable the development of new and highly competitive communication devices for e.g. government forces communications. Efficient interconnection of the different forces means more effectiveness in anti drugs, anti terrorism, police border patrol, and fire-fighters organisation. Also for Tedesys a competitive advantage will be provided by the project, as its main output is the implementation of algorithms on an embedded platform while they usually target PC. Thus Tedesys can address new markets and increase their revenues. Vector Fabrics tools will substantially boost their application performance and power modelling capabilities, crucial for optimal parallelization decisions. Subsequently, such intelligent yet affordable tools will empower European SMEs and multinational companies to quickly architect competitive multicore systems, while dramatically reducing Non-Recurring Engineering costs. All PHARAON demonstrators target the improvement of European citizens security and safety, should it be direct (collision avoidance) or indirect (citizens protection)
For a few years, multiprocessors have penetrated the embedded market. Platform providers such as Texas Instrument, STMicroelectronics or ST-Ericsson now propose multicore architectures with scalable performances. These architectures are capable to cope with the rising processing needs, while keeping stable power consumption. They provide developers with more flexibility and offer efficient power monitoring and control features. These evolutions will therefore help to sustain the market growth, but at the expense of programmability.
Hardware architectures evolve faster than software tools, and the specification and mapping of applications onto new heterogeneous multicore architectures becomes more complex. The lack of efficient software tools hinders adoption of new architectures and increases software development costs. The European industry therefore has to face a new challenge and sustain software developments by introducing efficient tools capable to assist designers.
The main objective of PHARAON is to improve competitiveness of the European electronic industry, especially with respect to reducing power consumption and improving performance, by providing new paradigms for multicore architectures programming, monitoring and control, as well as new dynamic power adaptation strategies, algorithms and interfacing standards. Raising the expertise of European industry in system architecture, software development and power management is crucial to ease the transition to multicore platforms. It will enlarge the range of applicability of a hardware platform and plays in favour of re-use, cost and time-to-market reduction, which have become crucial requirements in a worldwide competition.
Fields of science
Not validated
Not validated
Call for proposal
FP7-ICT-2011-7
See other projects for this call
Funding Scheme
CP - Collaborative project (generic)Coordinator
92230 Gennevilliers
France