High-Performance Computing (HPC) Application Performance Engineer per el Barcelona Supercomputing Center
Supercomputers are indispensable tools for solving the most challenging and complex scientific and technological problems through simulations. This position will be integrated into the operations department within the User Support team. The team will work on applications support for the most relevant projects asking for BSC resources, including performance optimization support, code porting, improvement of the scalability and performance of selected European scientific applications and tools towards Exascale and GPU computing optimization.
BSC now hosts one of the largest supercomputers in Europe as part of the EUROHPC Joint Undertaking. MareNostrum 5, a pre-exascale machine, delivers a sustained performance of over 205 PFlops. Its general-purpose partition is one of the largest in the world, with more than 700,000 cores available. Additionally, the accelerated partition features one of the biggest scientific clusters utilizing NVIDIA's new H100 GPU technology, with over 4,400 GPUs. The HPC support team plays a direct role in optimizing the usage of this powerful system.
Key Duties
- Optimize and adapt scientific application codes to new pre-exascale architectures
- Improve the performance of existing parallel codes, improving the serial efficiency and the scalability, changing if necessary the code or helping the developers with their required modifications.
- Generate performance analysis and benchmarks for selected applications and report the results to the application developers.
- Collaborate with other functional groups at the European and International levels on technical matters related to supporting scientific application work.
- Help the HPC users to run their codes in an efficient way and support the research teams to adapt their codes to new machines.
Closing Date: Tuesday, 01 October, 2024
Més posts de Recerca
Comentaris