NVIDIA Performance Libraries
NVIDIA Performance Libraries (NVPL) are a collection of essential mathematical libraries optimized for Arm 64-bit architectures. NVPL allows you to easily port HPC applications to NVIDIA Grace™ CPU platforms to achieve industry-leading performance and efficiency.
Key Features
Seamlessly Port to Grace CPU
NVPL math libraries are drop-in replacements for standard C and Fortran mathematical APIs. This allows existing HPC applications to be easily ported to Grace-based systems with no source code changes.
Optimized for Arm Architecture
NVPL is specifically tuned for Arm 64-bit CPUs and ensures mathematical applications achieve peak efficiency on the Grace microarchitecture. NVPL allows you to fully harness NVIDIA silicon in the data center.
Use Standard Math Subroutines
Many HPC applications rely on standardized mathematical APIs like BLAS, FFTW, and LAPACK that are critical to application performance. NVPL provides optimized implementations of these essential libraries for the Grace CPU.
NVPL Libraries (Beta)
NVPL BLAS
Industry standard Basic Linear Algebra Subprograms (BLAS) optimized for the NVIDIA Grace CPU architecture with single- and multi-threaded interfaces.
.png)
NVPL LAPACK
Dense direct linear solvers and eigen-solvers for computer vision, linear optimization, and more on NVIDIA Grace CPUs.
NVPL FFT
Fast Fourier Transforms (FFT) for applications including computational physics and quantum chemistry.
NVPL TENSOR
Tensor contraction, reduction, and elementwise operations for deep learning and inference on Grace CPUs.
NVPL RAND
High-performance random number generation (RNG) for fast and high-quality random numbers.
NVPL SPARSE
Sparse linear algebra subroutines for accelerating machine learning, fluid dynamics, and more.
NVPL ScaLAPACK
A LAPACK extension designed for distributed memory parallel computing environments.