Gregory Stoner retweeted
I'm excited to finally open-source the model from my 2022 paper, “Forecasting Global Weather with Graph Neural Networks”.
Highlights:
• 10-day forecast in <1 min
• Initialize forecasts from ERA5 or IFS analysis
• Scripts for eval, sensitivities, & Hurricane Sandy
Pekka has been doing fantastic job here
This Post is from an account that no longer exists. Learn more
Nice work by Aksel team with AdaptiveCpp
Like C++ standard parallelism offload? #AdaptiveCpp outperforms @NVIDIAHPCDev nvc++ in C++ PSTL offloading for the LULESH app on A100. Perhaps this a good moment to reconsider your use of vendor compilers 😉 (Also, the AdaptiveCpp binary runs runs on AMD/Intel hw too!)
Great Paper on SYCL/ESIMD and how to use it with Sparse GEMMS
PERFORMANCE OPTIMIZATION OF DEEP LEARNING SPARSE MATRIX KERNELS ON INTEL MAX SERIES GPU arxiv.org/pdf/2311.00368.pdf
Great seeing the updates in Chapel for the GPU, it just keep getting better.
Interested in an update about Chapel's support for high-level, vendor-neutral GPU programming? See the following slide deck from our recently published release notes: chapel-lang.org/releaseNotes…
It is pretty amazing how far Chapel has come to support Direct GPU Programing chapel-lang.org/docs/technot… @ChapelLanguage
Learn more about Level Zero api to bring a solid foundation to your accelerator language projects – at Colorado, USA
In this article, Intel's @RobM_A gives an overview of the Level Zero application interface, the basic architecture, and its benefits for low-level access control to compute unit resources. intel.ly/42fbkLF
#oneAPI
It was a great day on Sunday, Also helped me out to switch boots from a Dabello Pantera to Lange RX – at Colorado, USA


























