Matrix Solver is Accelerated by oneAPI and SYCL Using a Common Codebase for Multivendor GPUs
The CPU-specific Fortran code's most time-consuming phase, matrix inversion, was greatly accelerated using Intel oneAPI tools and the SYCL programming environment
It makes it possible to use the OpenMP and SYCL frameworks to offload calculations to GPUs for parallel executions
Vendor-locked to NVIDIA GPUs, the resulting code is accelerated using NVIDIA cuSOLVER library function calls
SYCL has the ability to use accelerated GPUs from many manufacturers, such as AMD, NVIDIA, and Intel
Vendor-locked to NVIDIA GPUs, the resulting code is accelerated using NVIDIA cuSOLVER library function calls
SC23 suggests Fortran-to-oneMKL SYCL API calls to speed up matrix inversion code on multi-vendor GPUs and reduce vendor lock-in