Sort by
Refine Your Search
-
Category
-
Program
-
Employer
- Forschungszentrum Jülich
- Technical University of Munich
- DAAD
- Leibniz
- Nature Careers
- Fraunhofer-Gesellschaft
- Max Planck Institute for Multidisciplinary Sciences, Göttingen
- Heidelberg University
- Helmholtz-Zentrum Dresden-Rossendorf - HZDR - Helmholtz Association
- Humboldt-Universität zu Berlin
- Max Planck Institute for Innovation and Competition, Munich
- Max Planck Institute for Radio Astronomy, Bonn
- NEC Laboratories Europe GmbH
- 3 more »
- « less
-
Field
-
science or engineering field; a completed PhD is an advantage Very good knowledge of software development with C++ and Python, as well as experience in GPU programming (e.g., CUDA or comparable frameworks
-
model in collaboration with partner institutions such as the German Climate Computing Center (DKRZ) and German Weather Service (DWD), including GPU porting. They will perform production runs of ICON and
-
to decompose the task into generating individual 2D horizontal layers separately in order to save GPU memory resources. Your Qualifications / Experience: completed MSc university degree in mathematics
-
programming skills in Python and popular frameworks (e.g., PyTorch). Familiarity with GPU-accelerated environments, virtualization tools, and prototyping using real testbeds (e.g., SDR). We expect a diploma in
-
Master’s theses Requirements: Master’s or PhD degree with above-average results in Applied Maths (analysis, numerics, modeling) or in a comparable program with a strong math. focus and knowledge in
-
and GPU servers for the delivery of the PC exercises, and jointly supervising the PC exercises during the course What you contribute Student on a STEM degree programme Good knowledge of at least one of
-
is of advantage: Knowledge of parallel programming and HPC architectures, including accelerators (e.g., GPUs) Experience in modelling and simulation, ideally in the field of energy systems Experience
-
managing supercomputer resources Strong skills in algorithm development for large sparse matrices Excellency in programming GPU accelerators from all major vendors Very good command of written and spoken
-
of spikes by a model Develop proxy apps representing the different processing stages of spiking network simulation code (targeting CPU and accelerators such as GPU or IPU) Systematic benchmarking of proxy
-
program embedded in a large-scale, nationally funded research consortium with access to unique multimodal clinical datasets - State-of-the-art GPU infrastructure for training and fine-tuning large