Sisef
Computational Scientist – Earth System Science
Sisef, Salt Lake City, Utah, United States,
Computational Scientist – Earth System Science
The Oak Ridge National Laboratory is seeking an outstanding Computational Scientist for the Computational Earth Sciences group in the Computational Science and Engineering Division (CSED) to accelerate development of the US Department of Energy’s (DOE’s) Energy Exascale Earth System Model (E3SM) on the world’s largest supercomputing platforms. In conjunction with the related Exascale Computing Project (ECP) effort, the successful candidate will design and develop advanced numerical methods, algorithms and software tools for achieving optimal performance that leverage accelerator-based supercomputing platforms (e.g., Summit, Frontier, Aurora, and Perlmutter). CSED focuses on transdisciplinary computational science and analytics at scale to enable scientific discovery across the physical sciences, engineered systems, and biomedicine and health. It develops community applications, data assets, and technologies and provides assurance to build knowledge and impact in novel, crosscut-science outcomes.The candidate will be a staff scientist with an initial term of two years within the CES Group of the Advanced Computing Methods for Physical Sciences Section in CSED. The CES Group is focused on the conduct of world-class research and development in Earth system modeling, model-data integration, large scale data analytics and machine learning, and model benchmarking at DOE’s Leadership Class Computing Facilities (LCFs) and its National Energy Research Scientific Computing Center (NERSC). The CES Group has specific strengths in numerical methods, simulation, and analysis focused on terrestrial hydrology and biogeochemistry, atmospheric and ocean dynamics, aerosols, regional climate, ice sheets and sea level rise, and the global carbon cycle.Major Duties/Responsibilities:ResearchDesign or redesign and implement scalable algorithms for simulation of dynamical, physical, chemical, and biological phenomena within Earth system models (ESMs) on unique hybrid (CPU/GPU) supercomputing architectures.Conduct performance analyses and testing of ESM kernels and components, as well as fully coupled ESMs, on various computing platforms.Develop and implement numerical algorithms aimed at improving the accuracy and performance of ESMs on target architectures.Collaborate with a diverse team of Earth system and computational scientists, both within CESG and across DOE Labs and partner universities, to address applied mathematics and performance issues in E3SM and its components on selected computing platforms.Author peer reviewed papers, technical reports, and proposals focused on next generation algorithms and optimization of ESMs on high performance computing platforms.Basic Qualifications :Ph.D. degree in Applied Mathematics, Computer Science, Computational Science, Earth System Science, Atmospheric Science, Oceanography, or a related field;Experience in model development, algorithm development, and/or performance optimization.Preferred Qualifications:Previous experience with large scale, distributed memory, parallel models;Experience with FORTRAN, C/C++, and Python;Familiarity and parallel programming experience on CPUs and GPUs with MPI, OpenMP, OpenACC, CUDA, including performance-portable programming models such as Kokkos;Knowledge of commonly used data file formats and conventions (e.g., CF, netCDF, HDF);Experience with high performance computing, advanced statistical and machine learning methods;Strong motivation to act as a computational liaison for enabling cutting-edge Earth system simulation within multi-disciplinary teams;Ability to report regular progress and collaborate with diverse teams at multiple research institutions; andExcellent verbal and written communication skills.
#J-18808-Ljbffr
The Oak Ridge National Laboratory is seeking an outstanding Computational Scientist for the Computational Earth Sciences group in the Computational Science and Engineering Division (CSED) to accelerate development of the US Department of Energy’s (DOE’s) Energy Exascale Earth System Model (E3SM) on the world’s largest supercomputing platforms. In conjunction with the related Exascale Computing Project (ECP) effort, the successful candidate will design and develop advanced numerical methods, algorithms and software tools for achieving optimal performance that leverage accelerator-based supercomputing platforms (e.g., Summit, Frontier, Aurora, and Perlmutter). CSED focuses on transdisciplinary computational science and analytics at scale to enable scientific discovery across the physical sciences, engineered systems, and biomedicine and health. It develops community applications, data assets, and technologies and provides assurance to build knowledge and impact in novel, crosscut-science outcomes.The candidate will be a staff scientist with an initial term of two years within the CES Group of the Advanced Computing Methods for Physical Sciences Section in CSED. The CES Group is focused on the conduct of world-class research and development in Earth system modeling, model-data integration, large scale data analytics and machine learning, and model benchmarking at DOE’s Leadership Class Computing Facilities (LCFs) and its National Energy Research Scientific Computing Center (NERSC). The CES Group has specific strengths in numerical methods, simulation, and analysis focused on terrestrial hydrology and biogeochemistry, atmospheric and ocean dynamics, aerosols, regional climate, ice sheets and sea level rise, and the global carbon cycle.Major Duties/Responsibilities:ResearchDesign or redesign and implement scalable algorithms for simulation of dynamical, physical, chemical, and biological phenomena within Earth system models (ESMs) on unique hybrid (CPU/GPU) supercomputing architectures.Conduct performance analyses and testing of ESM kernels and components, as well as fully coupled ESMs, on various computing platforms.Develop and implement numerical algorithms aimed at improving the accuracy and performance of ESMs on target architectures.Collaborate with a diverse team of Earth system and computational scientists, both within CESG and across DOE Labs and partner universities, to address applied mathematics and performance issues in E3SM and its components on selected computing platforms.Author peer reviewed papers, technical reports, and proposals focused on next generation algorithms and optimization of ESMs on high performance computing platforms.Basic Qualifications :Ph.D. degree in Applied Mathematics, Computer Science, Computational Science, Earth System Science, Atmospheric Science, Oceanography, or a related field;Experience in model development, algorithm development, and/or performance optimization.Preferred Qualifications:Previous experience with large scale, distributed memory, parallel models;Experience with FORTRAN, C/C++, and Python;Familiarity and parallel programming experience on CPUs and GPUs with MPI, OpenMP, OpenACC, CUDA, including performance-portable programming models such as Kokkos;Knowledge of commonly used data file formats and conventions (e.g., CF, netCDF, HDF);Experience with high performance computing, advanced statistical and machine learning methods;Strong motivation to act as a computational liaison for enabling cutting-edge Earth system simulation within multi-disciplinary teams;Ability to report regular progress and collaborate with diverse teams at multiple research institutions; andExcellent verbal and written communication skills.
#J-18808-Ljbffr