Ryan Friese

this should hopefully be addressed in the v0.6 release (https://github.com/pnnl/lamellar-runtime/commit/52a3d900a84b655cf694a4a2763b3e9eaea3b124) It worked in my simple test, but if history is anything to go by, that may not be the case...

We don't directly support CUDA as a backend, but we can integrate with other crates that target accelerators. For example, we also develop the [mcl-rs](https://crates.io/crates/mcl-rs) crate, a Rust frontend around...

While I expect there to be some variation between backends, with each level of communication (local -> shared memory -> distributed) introducing slowdowns, the range you are showing is definitely...

I've created a new branch associated with this issue that should better handle oversubscribing threads to cores (assuming this was the actual underlying issue). For what its worth when I...

so far this looks good, all the conflicts look like non issues to me

Thanks for the effort here, its definitely appreciated and needed as we transition this from more research oriented code to stable code. - Why did the serialization/deserialization functions take a...

Hmm lets try having it accept a mutable reference to self and see if that causes any major issues, given that its a public interface we should probably try to...

This looks all good to me! One thing I've ran into in testing is for more complicated tests debug is too slow, so I test in release mode often. Do...