Rust-CUDA Alternatives
Similar projects and alternatives to Rust-CUDA
-
-
rust-ndarray
ndarray: an N-dimensional array with array views, multidimensional slicing, and efficient operations
-
Scout APM
Less time debugging, more time building. Scout APM allows you to find and fix performance issues with no hassle. Now with error monitoring and external services monitoring, Scout is a developer's best friend when it comes to application development.
-
-
-
-
ReShell
A modern shell language built for the future :rocket: (by ClementNerma)
-
-
SonarQube
Static code analysis for 29 languages.. Your projects are multi-language. So is SonarQube analysis. Find Bugs, Vulnerabilities, Security Hotspots, and Code Smells so you can release quality code every time. Get started analyzing your projects today for free.
-
-
Pandas
Flexible and powerful data analysis / manipulation library for Python, providing labeled data structures similar to R data.frame objects, statistical functions, and much more
-
-
-
mintable
🍃 Automate your personal finances – for free, with no ads, and no data collection.
-
MLPP
A library created to revitalize C++ as a machine learning front end. Per aspera ad astra.
-
-
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
-
-
Rust-CUDA reviews and mentions
- There is framework for everything.
-
C++ is making me depressed / CUDA question
And here's an example on how to add two floats using Rust-CUDA: https://github.com/Rust-GPU/Rust-CUDA/blob/master/examples/cuda/gpu/add_gpu/src/lib.rs
I can't say whether or not it works well, and seems maybe fiddly to install, but Rust-CUDA might work for you.
-
Does Rust still depend on LLVM?
Kinda - SPIR-V is an intermediate representation that can be consumed by graphics drivers and compiled down by them to the GPU's native code. In that sense it's more similar to emitting LLVM IR than e.g. x86 machine code. Similarly, Rust-CUDA emits NVVM IR, which is a subset of LLVM IR.
-
State of GPGPU in 2022
Rust-GPU look promising, also Rust-CUDA, also I see OpenCL 3.0 wrapper
-
Silverblue loads nouveau instead of installed nvidia
rpm-ostree install https://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm https://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm sudo rpm-ostree uninstall akmod-nvidia sudo rpm-ostree install akmod-nvidia-470xx rpm-ostree kargs --append=rd.driver.blacklist=nouveau --append=modprobe.blacklist=nouveau --append=nvidia-drm.modeset=1 sudo systemctl reboot sudo rpm-ostree install nvidia-settings-470xx xorg-x11-drv-nvidia-470xx-power sudo rpm-ostree uninstall xorg-x11-drv-nvidia-cuda sudo rpm-ostree install xorg-x11-drv-nvidia-470xx-cuda sudo rpm-ostree install xorg-x11-drv-nvidia-470xx-devel sudo rpm-ostree install xorg-x11-drv-nvidia-470xx-cuda-devel sudo rpm-ostree install cuda-samples sudo rpm-ostree install vulkan-tools vkmark mesa-vulkan-devel sudo rpm-ostree install libshaderc-devel sudo rpm-ostree install clang clang-tools-extra libstdc++-devel sudo rpm-ostree install glib2-devel glib-devel avahi-gobject-devel sudo rpm-ostree install cairo-devel pango-devel gdk-pixbuf2-devel sudo rpm-ostree install graphene-devel gtk4-devel cairo-gobject-devel sudo systemctl reboot modinfo /usr/lib/modules/$(uname -r)/kernel/drivers/video/nvidia.ko | grep ^version find /usr/lib/modules -name nvidia.ko -exec modinfo {} \; sudo lspci -v | grep -A 20 VGA git clone https://github.com/Rust-GPU/Rust-CUDA.git git clone https://github.com/vulkano-rs/vulkano.git git clone https://github.com/Relm4/relm4.git glxgears glxinfo glxgears glxinfo vkcube vkcubepp ./teapot ./triangle ./occlusion-query ./interactive_fractal
- [P] C++ Machine Learning Library Built From Scratch by a 16-Year-Old High Schooler
-
Writing Vulkan SPIR-V shaders in C++?
Is Rust CUDA, "writing [...] GPU code fully in Rust" (from its "about" description), but still early in development, relevant?
- Announcing Rust-CUDA 0.3; Hardware raytracing, cuBLAS, cuDNN, atomics, and more
-
Rust and Scientific/High-Performance Computing
I'm looking for some good sources I can read to see arguments for and against. I'm relatively new to Rust myself, but I am looking at the Rust-CUDA project (and have contacted the developer). I am primarily interested in Rust for this task because of what it offers in terms of memory safety, though I realize that some of the tools/algorithms rely heavily on shared memory between threads. Really, any good reads that you folks could offer would be greatly appreciated.
-
[Media] Follow-up to my previous path tracing in rust post, after a lot of work and trial and error, Rust can now do Hardware Raytracing on the GPU using OptiX, sharing the exact same core rust code with the CPU and GPU renderers.
After nearly a year of work in total, i am happy to announce i was able to bring hardware raytracing to Rust CUDA using the OptiX SDK. OptiX is an SDK based on CUDA which provides fast raytracing pipelines which are hardware-accelerated on Turing and Ampere GPUs. OptiX is half a CPU-side C API, and half a GPU-side C++ library i had to recreate in rust using inline assembly.
-
What's everyone working on this week (4/2022)?
I finally got rust-cuda to generate optix-compatible code, this is the first picture rendered using hardware raytracing 100% in rust. Rust GPU code and Rust CPU code. I will be making a follow-up to my path tracer in rust post once i make it use OptiX. You can find the (very early) code for this picture here.
-
Rust CUDA and shared memory
I am not very familiar with GPU programming, but you might get some answers from here https://github.com/Rust-GPU/Rust-CUDA
We actually have static and dynamic shared memory now. Rust is perfectly able to express CUDA shared memory. What it is NOT able to do is do so very safely. shared memory is the pinnacle of "holy crap this is unsafe", it is basically impossible to check static shared memory usage for safety. Our best bet is: - Encouraging users to run compute-sanitizer (new cuda-memcheck) - Removing as much raw shared mem usage as possible by making abstractions.
-
Enzyme: Towards state-of-the-art AutoDiff in Rust
Would you be interested in collaborating and making it part of Rust CUDA? cuDNN is my next target after cuBLAS but it is a lot of work for one person. I would like to keep all library wrappers inside of one org/repo so there is no ambiguity about what will likely be the most complete and/or most maintained.
Stats
Rust-GPU/Rust-CUDA is an open source project licensed under Apache License 2.0 which is an OSI approved license.
Popular Comparisons
Are you hiring? Post a new remote job listing for free.