NVIDIA Application Lab at Jülich
The NVIDIA Application Lab at Jülich is jointly operated by Jülich Supercomputing Centre (JSC) and NVIDIA since July 2012. The main focus of this exascale lab is to
- Enable scientific applications for GPU-based architectures
- Provide support for their optimization
- Investigate performance and scaling
A broad application portfolio encompassing computational neuroscience, high-energy physics, radio astronomy, data analytics and others has been established.
Other topics of the lab:
- Parallelisation of applications on multiple GPUs
- Application requirements analysis
- Kepler and CUDA feature analysis
- Training in GPU programming and optimisation
Contact: Dirk Pleiter email@example.com
- GPUMAFIA: high-performance subspace clustering on GPU
- Halloc: fast and scalable GPU memory allocator
Presentations and News
- CUDA Dynamic Parallelism (Part 1)
- CUDA Dynamic Parallelism (Part 2)
- CUDA Dynamic Parallelism (Part 3)
- Accelerating a C++ CFD code with OpenACC
- CUDA-aware MPI (Part 1)
- CUDA-aware MPI (Part 2)
- CUDA Spotlight: Pierre Wahl
- Lab presentation at SC2013
- Lab presentation at SC2012
- NVIDIA press release
- NVIDIA Application Lab at Jülich: 1st Annual Workshop (July 2013)
NVIDIA Application Lab Kick-off Workshop (September 2012)
- A. Adinetz et al., "Triplet Finder: On the Way to Triggerless Online Reconstruction with GPUs for the PANDA Experiment", Proceedings International Conference on Computational Science 2014.
- J. Kraus et al., "Benchmarking GPUs with a parallel Lattice-Boltzmann code", Proceedings 25th International Symposium on Computer Architecture and High Performance Computing 2013.
- A. Adinetz et al., "Computation of Mutual Information Metric for Image Registration on Multiple GPUs", Proceedings Heteropar 2013.
- A. Adinetz et al., "GPUMAFIA: Efficient Subspace Clustering with MAFIA on GPUs" Proceedings Euro-Par 2013.
- A. Adinetz et al., "NVIDIA Application Lab at Jülich" inSiDE, Vol. 11, No. 1 (2013).