R. M. Caplan (@sumseq) header
R. M. Caplan (@sumseq) profile picture

Computational Scientist (views mine)

California, USA
Joined March 2011
Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)Photo by R. M. Caplan (@sumseq)
Are you going to #sc26? Come see and learn about PSI's cutting-edge #supercomputing work in the exhibit hall and "Art of HPC" exhibit! #hpc #fortran #gpu #chicago
4
1
8
556
We're giving away a Hogwarts Legacy custom wrapped NVIDIA GeForce RTX 5090 to celebrate its official release with DLSS 5 🪄 To enter: 🟢 Share this post 🟢 Comment #RTXON
12,667
11,347
595
11,278
846,296
swsc-journal.org/articles/sw… Check out our new paper on long-duration flux transport modeling of the solar surface magnetic field, showing (among other things) that random flux emergence can build up polar fields. The simulations used HipFT (github.com/predsci/hipft) which is #poweredbyfortran and uses Fortran standard parallelism to run on GPUs. The simulations were run on an NVIDIA A100 GPU using the NVIDIA HPC SDK compilers, allowing for multi-decade simulations to run in reasonable times.
1
61
R. M. Caplan retweeted
4,805
10,337
4,397
62,977
12,566,350
FortranCon2025 09 Keynote Reid - Reflections on working for 45 years on ... youtu.be/LqtQDLhT4Eg?si=Bdze… via @YouTube
2
105
LUMI-AI is a major step forward for European sovereign AI and AI for Science. Powered by @AMD Instinct MI430X GPUs, it brings modern AI acceleration together with up to 288 TFLOPS of hardware FP64 per GPU. newsroom.amd.com/news/amd-in…
1
1
7
414
Want to run your @fortranlang #Fortran code on GPUs but don't want to add external APIs or directives? Check out our new paper on using Fortran's "do concurrent" to run on NVIDIA, AMD, and Intel GPUs! arxiv.org/abs/2608.20586 Adding @OpenMP_ARB data movement directives can help with efficiency (on Intel GPUs) and for some GPUs are required (AMD consumer GPUs), but for many cases, "Fortran-only" works great - even across multiple multi-GPU nodes! @FortranWiki @fortrantip @nvidiadeveloper @AMDGPU_ @AMDServer @IntelDevTools @inteldevs
5
17
359