"Our research group at Indian Institute of Technology Goa (IIT Goa) has developed an in-house CFD solver to simulate incompressible turbulent flows. To port our solver to GPUs, we are working with the unsteady Poisson equation, a model template for our CFD solver. The Poisson equation code is developed in CUDA*, limiting its execution to vendor-specific GPUs. Using Intel's SYCLomatic tool for oneAPI, this code is migrated to SYCL*, thus opening up other vendor and architectural alternatives. After further optimization, the migrated code runs on Intel® Data Center GPU Max Series and achieves up to 1.9x speedup compared to the existing GPU solution. We look forward to using the migrated SYCL code on different platform architectures and migrating our CFD solver to SYCL."
IIT Goa, a premier academic institute in India, used Intel tools to migrate a computational fluid dynamics (CFD) Poisson solver equation from CUDA to multivendor-supported SYCL.
With code migration, the standard equation governing physical processes, such as diffusion and heat transfer, performs better. Plus, a single code base can run seamlessly across heterogeneous architectures.
The solver’s performance improved up to 1.9x on Intel® Data Center GPU Max Series 1550 compared to its performance on NVIDIA* A100 GPU.
Resources
The High-Performance Computing (HPC) Nodal Center at IIT Goa, in association with Intel, aims to enhance HPC education and research in India.