TEXT - Towards EXascale ApplicaTions
TEXT is an EC funded research project into new parallel programming models, in particular StarSs, aiming at scaling real-world applications to Exascale performance level.
Duration: 24 months, starting June 2010. (Update: the project has been extended by 3 month to a total of 27 months)
TEXT is centered on the vision that the key component to support high productivity and efficient use of a system is the programming model. The StarSs/SMPSs model provides the necessary support for asynchrony and heterogeneity as well as allowing incremental parallelization, modularity and portability of applications. By combining it with MPI we can propagate its characteristics to the global application level. This is also a way to leverage and provide a smooth migration path for the huge number of applications today written in MPI.
The aim of the TEXT project is to install the MPI/SMPSs environment at HPC facilities of project partners and demonstrate how seven real and relevant applications/libraries, taken from various scientific domains, can be improved using this programming model. At HLRS, in particular, we will focus on the Lattice-Boltzmann code lbc. In addition HLRS will develop debugging and correctness-checking tools for StarSs/SMPSs, namely the debugger Temanejo.
- BSC (Spain) (coordinator)
- HLRS (Germany)
- FZJ (Germany)
- EPCC (United Kingdom)
- FORTH (Greece)
- Univ of Manchester (United Kingdom)
- Univ de Pau (France)
- Univ Jaume I (Spain)
- IBM Zurich (Switzerland)
People involved at HLRS
- StarSs debugger AYUDAME/TEMANEJO
- see also the liveDVD of the Project HPC Europa2 which provides all tools for StarSs
- Article on TEXT published by The Exascale Report (link; subscription required)