Development of performance portable spline solver for exa-scale plasma turbulence simulation

This paper describes the development of performance portable spline building kernels on top of Kokkos-kernels. We wish to solve a single matrix equation with multiple right-hand sides. This problem is quite unique and thus neither Kokkos-kernels (direct method) nor Ginkgo (iterative methods) is opti...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis s. 1117 - 1128
Hlavní autoři: Asahi, Yuuichi, Legouix, Baptiste, Bourne, Emily, Padioleau, Thomas, Bigot, Julien, Grandgirard, Virginie, Obrejan, Kevin
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 17.11.2024
Témata:
Conferences
Costs
Ginkgo
High performance computing
Iterative methods
Kernel
Kokkos
Kokkos-kernels
Maintenance
Mathematical models
Performance portability
Plasmas
Sparse matrices
Splines
Splines (mathematics)
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper describes the development of performance portable spline building kernels on top of Kokkos-kernels. We wish to solve a single matrix equation with multiple right-hand sides. This problem is quite unique and thus neither Kokkos-kernels (direct method) nor Ginkgo (iterative methods) is optimized for this. We develop the required solvers in Kokkos-kernels with a batched serial implementation and optimize them using kernel fusion and sparse matrix storage. We demonstrate that our spline solver works efficiently on NVIDIA A100 and AMD MI250X GPUs, while keeping a reasonable performance on CPUs. This effort significantly reduces the development and maintenance cost of spline solvers on exa-scale supercomputing systems.
DOI:10.1109/SCW63240.2024.00154