Corporate #FLOSS at its worst: #NVIDIA controls the #Thrust library and its #CUDA, #OpenMP and #TBB backend. #AMD provides rocThrust, that is just Thrust with the CUDA part stripped an a new backend for #ROCm / #HIP. Nobody* is working on a backend for #SYCL
#Intel provides its own #oneAPI alternative as #oneDPL, which is NOT a drop-in replacement.

This is why we can't have nice things.

*there's a dead project here
https://github.com/wdmapp/syclthrust

A question:

Today, a one-week course about #MPI and #OpenMP came to an end. The participants were pleased, the (external) lecturer too. I helped organize it.

My institution, however, thinks to abandon on-site courses entirely. Hybrid and online will have to do.

Is there any study material available which compares the teaching of complex stuff online vs. on-site?

#HPC #OpenScience

Interested in learning more about programming GPUs? Join this (free) online PRACE course organized by SURF covering #OpenACC #OpenMP #HIP and #CUDA in December 2022. I'll be among the instructors.

https://events.prace-ri.eu/event/1466/

#GPU #PRACE

Adding #FreeBSD support to #GPUSPH has been an interesting exercise, and a low-effort one too (possibly because of our existing support for #macOS and #Android, that helped weed out a number of #GNU-isms).

We do still have an issue with the programing locking up HARD (locks up any pre- or post-attached debugger, and even prevents a clean shutdown) when using thread affinity or #OpenMP —I wonder if it's an issue with it being in a VM or something else.

After the lecture, I am already a bit wary of investing codes in GPU. It looks powerful for sure, but the vendor lock-in is a problem. I should also check out if there are any #PRACE courses in #OpenACC / #OpenMP