Develop efficient algorithms for solving large-scale eigenvalue problems arising from nuclear structure calculations; Develop machine learning techniques to refine eigenvector approximation; Implement algorithms in the MFDn nuclear structure calculation software package using multiple programming models (MPI/OpenMP/OpenACC, CUDA, Kokkos etc.); Optimize the performance of the eigensolver by reducing communication overhead, improving load balancing and data locality
Requirements:
Requires a Phd in Mathematics, Computer Science, Physical sciences (physics, chemistry and materials science), Computational Science or Engineering within the last 3 years, with a strong background in numerical methods development and high performance computing; Knowledge of numerical linear algebra and iterative methods for solving large-scale eigenvalue problems; Knowledge of sparse matrix methods
Text:
Computational Mathematics Postdoctoral Scholar Develop efficient algorithms for solving large-scale eigenvalue problems arising from nuclear structure calculations; Develop machine learning techniques to refine eigenvector approximation; Implement algorithms in the MFDn nuclear structure calculation software package using multiple programming models (MPI/OpenMP/OpenACC, CUDA, Kokkos etc.); Optimize the performance of the eigensolver by reducing communication overhead, improving load balancing and data locality Requires a Phd in Mathematics, Computer Science, Physical sciences (physics, chemistry and materials science), Computational Science or Engineering within the last 3 years, with a strong background in numerical methods development and high performance computing; Knowledge of numerical linear algebra and iterative methods for solving large-scale eigenvalue problems; Knowledge of sparse matrix methods
Please click here, if the job didn't load correctly.
Please wait. You are being redirected to the job in 3 seconds.
