Ingo's Blog

I’ve been playing a bit with OpenMP recently – in particular, with the pragma omp target based device offloading that OpenMP 5.0 and newer are offering.

I’m aware that not everybody does (yet) know what OWL actually is, but for those that do, and for those: yes, I know I have to write a blog article about OWL at some point in time to introduce it – but this is not the blog post that’ll do this.

For a while now I’ve been telling everybody that my original ‘optix7course’ repo on github (ie, https://github.com/ingowald/optix7course) is a bit deprecated, and everybody should instead look at OWL (which since last week can be found at a new location,…

This article is about a capability – and in particular, about the “how do you actually use that” – that i recently added to Barney, namely, to do data-parallel multi-GPU rendering through Barney’s ANARI interface layer.

After a lot of fixed, pynari 1.2.8 was finally released earlier this week.

I’ll have to write a bit more later on the “how” of it, but for now, just a quick “first light” of barney rendering the latest “Drosophilia Brain” data set (using four H100 GPUs).

One of the best things about working in graphics is that at least some of the bugs one encounters are – in some way, shape, or form – funny.

Good news; getting closer to the next version of pynari: Feedback to 1.0 was overwhelmingly positive, but two things stood out: First, there seems to be some demand for having a CPU-fallback for the kind of hardware that doesn’t have any CUDA capable GPUs…

I haven’t written much for a while – partly because I was too busy, partly because some of the stuff i worked on I couldn’t write about, and mostly, because much of the stuff that I could in theory write about wasn’t exactly “ready enough” to do so.

(sorry for the delay, this is a bit outdated – this was apparently stuck in my Draft folder, and never sent out ) Finally ….