This site has been archived on
The Community Research and Development Information Service - CORDIS
Information & Communication Technologies



FP7: FET Proactive Intiative: Concurrent Tera-Device Computing (TERACOMP)



FET proactive initiative: Concurrent Tera-Device Computing (Extract from Work programme 2009-2010 /Objective 2009.8.1 - Call 4)

Integrated circuits and tightly-coupled systems will integrate up to 1000 billion devices by the year 2020. These will provide orders of magnitude improvement in performance and cost only with much higher concurrency and heterogeneous architectures tuned to specific application kernels. In parallel, device variability and failure rates will reach critical levels and power saving methods will be required at all system levels from transistors to architecture and software.

Target outcome:

Radically new methods and tools for architecture design and programming of chips and systems beyond 2020, including compilers and run-time systems:

  1. Complexity of design and run-time of many-core heterogeneous systems : Radically new concepts, design paradigms, methods and proofs-of-concept to address design, compilation and run-time complexity of computing engines with 100+ heterogeneous cores. Solutions should cover hardware, software and possibly reconfigurable hardware.
  2. Design of dependable systems with faulty components : Methodologies and approaches for the design and construction of dependable systems in the face of critical levels of hardware or software faults and in the face of component variability.
  3. Breakthrough programming paradigms : Radically new design and programming paradigms to enable effective programming of tera-scale ICT Systems with 100+ cores, in terms of scalability, portability and dependability. They would enable high data throughput applications and new algorithms for the management of massive data sets.

The developments should be motivated by the requirements of wide classes of relevant applications in a time scale of 10-15 years. Proof of concept demonstrations should be developed in parallel with foundational advances.

Integrated projects should address at least two of the above topics, STREPs at least one.

Expected impact:

  • Reinforced ability to design, program and manage competitive concurrent computing systems beyond the year 2020, thereby supporting the European systems industry in extending its strengths to diverse future application domains while supporting scalability and portability of applications.Practical info on Call 4.

Short abstract from the Work Programme 2009-2010  (   345KB) 






Timetable Call 4

  • Call launched: 19 November 2008
  • Start of projects: Jan-Febr 2010