APOS-EU: Application Performance Optimisation and Scalability
The future of high-performance computing means two things: massive parallelism and heterogeneity.
Processor manufacturers are adding more and more computational cores onto their CPUs and HPC vendors are looking to augment these many-core chips with GPU-like accelerators to achieve the next level in number-crunching power. While such developments give scientists the potential to run bigger, faster, and more detailed simulations, there are significant challenges to be overcome before current application codes can exploit these hardware advances.
APOS investigated technologies and develop tools for the software challenges encountered by computational scientists in the face of new trends in high-performance computing architecture development. The project sought to:
- Advance the state of the art in optimisation and scalability methodologies.
- Increase cooperation and collaboration between European and Russian scientists.
- Develop advanced tools for programming heterogeneous supercomputers.
Partners and funding
The complementary APOS-EU and APOS-RU projects were funded as part of the European Commission’s Framework 7 “EU-Russia” call and by the Ministry of Education and Science of the Russian Federation respectively, looking to foster collaborative research between European and Russian scientists. The European project had five partners and is led by EPCC. The Russian counterpart was headed by the Keldysh Institute of Applied Mathematics at the Russian Academy of Sciences in Moscow, one of four partners forming the consortium. The APOS projects started in early 2011 and ran for 2 years
For further information, visit the APOS-EU project website