The Center for Efficient Exascale Discretizations (CEED) bake-off problems are a collection of benchmarks representing important compute-intensive kernels and solvers relevant to high-order finite and spectral-element methods, such as those used in the Nek5000 computational fluid dynamics (CFD) code. This talk presents a Data Parallel C++ (DPC++) implementation of the CEED bake-off problems. Benchmark results are given for Intel® CPUs and GPUs. Intel® Advisor is used to conduct cache-aware roofline analysis and understand performance. Finally, batched routines from the Intel® oneAPI Math Kernel Library (oneMKL) Basic Linear Algebra Subprograms (BLAS)-like extensions are explored as a replacement for certain directly programmed DPC++ kernels.

Kris Rowe is an Intel® Software Innovator and an assistant computational scientist in the performance engineering group at Argonne Leadership Computing Facility. He holds a PhD in applied mathematics from the University of Waterloo.

Saumil Patel is an assistant computational scientist in the Computational Science Division at Argonne National Laboratory. He holds a PhD in mechanical engineering from the City College of New York.