Faster Scientific Simulations with New FP64 Algorithms in NVIDIA cuBLAS
Faster Scientific Simulations with New FP64 Algorithms in NVIDIA cuBLAS
NVIDIA cuBLAS releases new FP64 algorithm, significantly enhancing scientific simulation performance by optimizing DGEMM/ZGEMM high-precision matrix multiplication. The algorithm achieves speed improvements while maintaining high precision, suitable for high-performance computing and AI applications in fields such as molecular chemistry and materials science.
Many scientific simulations, from molecular chemistry to materials science, spend the majority of their execution time computing high-precision dense general...
Many scientific simulations, from molecular chemistry to materials science, spend the majority of their execution time computing high-precision dense general matrix multiplications (GEMM). These operations are called DGEMM for real 64-bit floating-point numbers and ZGEMM for complex 64-bit floating-point numbers. cuBLAS is widely used by HPC and AI applications to accelerate GEMM and other Basic…
Source: NVIDIA Technical Blog · developer.nvidia.com