VectorWare retweeted
VectorWare is the most ambitious GPU project that I know of - this is how we will cross the performance frontier and overcome the GPU limitations of today.
VectorWare retweeted
Excited to be speaking about Rust on the GPU at frontiers by @paradigm! Apply to come see my talk in the link below 🤙
VectorWare retweeted
.@vectorware is coming!!!!
Introducing CUDA Rust!
CUDA Rust lets you write GPU kernels natively in Rust, not just launch them from it.
Two paths: cuda-oxide for SIMT kernels compiled to PTX, and cutile-rs for Tile-based programming on stable Rust. Both can catch aliasing errors at compile time.
Technical blog: nvda.ws/4hm1bHS
Excited to be working with the NVIDIA team on Rust and GPUs! More VectorWare info coming soon.
Introducing CUDA Rust!
CUDA Rust lets you write GPU kernels natively in Rust, not just launch them from it.
Two paths: cuda-oxide for SIMT kernels compiled to PTX, and cutile-rs for Tile-based programming on stable Rust. Both can catch aliasing errors at compile time.
Technical blog: nvda.ws/4hm1bHS
We're not planning to make it better than the status quo. Generally we launch megakernels once instead of launching many that coordinate. Similar to how on the CPU you don't launch many processes, you launch one process and then do a bunch of heterogenous work on threads and such
We are excited to announce that we can successfully use Rust's portable SIMD on the GPU. This has never been done before.
vectorware.com/blog/simd-on-…
Supporting Rust's `core::simd` enables existing Rust code to work on the GPU and makes GPU programming more ergonomic.
No worries, Rust is not for everyone! We're pretty bullish on github.com/nvidia/stdexec which maps well to our async/await work and CUDA Tile. Someone will likely take what we did here for Rust threads and do it for C++ to explore the tradeoffs.
Agreed, software architected for the GPU will always be better than CPU software ported over. We've done this work mainly to use existing CPU libraries in GPU-native apps where it makes sense (adding some GPU-specific logic in places for perf).
Yep, future posts will talk about SIMD lanes within each warp. Check the pedantic note on "first" if you haven't!
We are excited to announce that we can successfully use Rust's std::thread on the GPU. This has never been done before.
vectorware.com/blog/threads-…
Supporting Rust's std::thread enables existing Rust code to work on the GPU and makes GPU programming more ergonomic.
We are excited to announce that we can successfully use Rust's async/await on the GPU. This has never been done before.
vectorware.com/blog/async-aw…
Supporting Rust's async/await (and futures) enables existing Rust code to work on the GPU and makes GPU programming more ergonomic.
We got an unmodified coremark runnning on the GPU and a GPU warp is surprisingly competitive with a CPU core
We are excited to announce that we can successfully use Rust's standard library from the GPU. This has never been done before.
vectorware.com/blog/rust-std…
Supporting Rust's standard library enables existing Rust code to work on the GPU and makes GPU programming feel normal.
VectorWare retweeted
I did an interview with the @rust_foundation about Rust on the GPU (with a brief mention of my new company @vectorware ):
youtube.com/watch?v=monOq_uH…
We're honored to be one of The Information's 50 most promising startups:
theinformation.com/projects/…
Hello world! We are building the first GPU-native software company. Today we are sharing the thesis, people, and partners behind it.
vectorware.com/blog/announci…