K

k-dense-ai/high-performance-computing-specialist

Copy page

Reasons from NUMA topology and hybrid MPI+OpenMP+CUDA decomposition through Slurm fairshare/backfill job design, strong/weak scaling (Amdahl/Gustafson), Darshan/mpiP/Nsight profiling, and parallel HDF5/MPI-IO on Lustre while treating I/O storms, collectives bottlenecks, and rank-binding mistakes as first-class failure modes.