NVIDIA makes cuDSS generally available for GPU-accelerated scientific simulation
Overview
NVIDIA releases cuDSS, a CUDA-X library, in general availability, moving the sparse-system solving step of engineering and scientific simulations onto GPUs.
NVIDIA says developers can integrate cuDSS into existing engineering, scientific and optimization software, and can use it through C, C++, Python and Fortran interfaces.
NVIDIA says the library scales from workstations to large clusters. Applied Materials reports speedups of up to 10x over CPU-only approaches, and Dassault Systèmes reports an 8-10x speedup in its Abaqus solvers. These figures are the companies' own reported results.
Written by AI from the articles below · updated Oct 21, 8:10 AM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- NVIDIA BlogOfficialNVIDIA releases cuDSS for GPU-accelerated engineering and scientific simulations
AINVIDIA says cuDSS, a CUDA-X library, is now generally available and moves the sparse-system solving step of engineering and scientific simulations onto GPUs. Applied Materials reports up to 10x speedups over CPU-only approaches, and Dassault Systèmes reports an 8-10x speedup in its Abaqus solvers. cuDSS can scale from workstations to large clusters and is accessible through C, C++, Python and Fortran interfaces.
Heat Trend
Not enough continuous observations to show a trend yet.