Developer Resources from Intel and Old Dominion University (ODU)*
Old Dominion University (ODU)* overcame vendor hardware lock-in by migrating parallel adaptive GPU algorithm for numerical integration (PAGANI), a deterministic quadrate-based algorithm, and m-CUBES, a probabilistic Monte Carlo algorithm, from CUDA* to SYCL* using AI Tools from Intel.
The SYCL-migrated versions of the numerical integration algorithms perform 90%-95% as well as the CUDA-optimized versions.
Get Started with ODU*
Research Computing Services at ODU* provides specialized environments for advancement of high-performance computing (HPC) research at the university.
Case Study
PAGANI and m-CUBES: An Effortless Migration from CUDA to SYCL for Numerical Integration
"SYCLomatic helped us quickly move the CUDA*-optimized numerical integration code to Intel oneAPI and SYCL*. The SYCL version gave us portability, and we could run the code on Intel® GPUs and CPUs and NVIDIA* GPUs. We demonstrated that the SYCL code provided 90% of the performance as a CUDA-optimized code on NVIDIA V100."
— Old Dominion University
Intel® oneAPI Base Toolkit
Download the Stand-Alone Version
Develop performant code quickly and correctly across hardware targets, including CPUs, GPUs, and FPGAs, with this standards-based, multiarchitecture compiler.