Sort by
Refine Your Search
-
Listed
-
Category
-
Country
-
Employer
- Oak Ridge National Laboratory
- University of Central Florida
- Aalto University
- Argonne
- CEA
- Michigan State University
- National University of Singapore
- Northeastern University
- SUNY University at Buffalo
- Technical University of Denmark
- Technical University of Denmark (DTU)
- University of Florida
- University of Lund
- University of North Carolina at Chapel Hill
- University of Utah
- 5 more »
- « less
-
Field
-
implemented in the Fortran programming language, and it relies on the platform CUDA for parallelization of the computation over several GPUs’ cores, and has interfaces with Matlab and Python for ease of use
-
posséder de solides compétences en calcul scientifique et HPC, notamment en : * Programmation en C++ dans des environnements parallèles (MPI, threads) ; * Programmation GPU (CUDA, HIP ou équivalent
-
skills are very important. Experience with C/C++, Python, Julia, MATLAB, CUDA, or a comparable scientific-computing environment is desirable. Additional Application Materials Required: UCF requires all
-
open source framework for adaptive mesh refinement, written in CUDA and supporting one-sided communication for scalability. The target architecture is a supercomputer with AMD MI300A GPU+CPU. 2
-
). Expertise in the following areas is considered beneficial: scripting (Python, bash, etc.), high-performance programming (Fortran/C/C++, CUDA, etc.), ML libraries (sklearn, Pytorch, etc.). In your CV or
-
-squares solvers, or scalable tensor kernels. GPU performance engineering — CUDA/HIP kernel design, communication–computation trade-offs, or performance modeling on heterogeneous systems. Theory of
-
platform CUDA for parallelization of the computation over several GPUs’ cores, and has interfaces with Matlab and Python for ease of use. You will research determining the ideal microstructure, for
-
stack, lifecycle-managed nodes, TF2, rosbag workflows, and distributed system diagnostics. Experience with CUDA and deployment on NVIDIA edge-GPU platforms (Jetson Orin, Thor, or Spark). Experience
-
(Chroma, CPS, MILC, QUDA, Grid, etc.) is highly desirable. Familiarity with C/C++ and accelerator application programming models such as CUDA, HIP, SYCL, OpenMP, Kokkos or Raja, vectorization; MPI
-
ecosystems such as CUDA and/or AMD ROCm/HIP. Experience with performance profiling and optimization of scientific applications, particularly molecular dynamics and machine learning applications