Crate metrale_comm

Crate metrale_comm 

Source
Expand description

2026-09-26: Collective-communication backends behind one trait, CommBackend: SingleGpuBackend, where every operation is a no-op, and, with the nccl feature, NcclBackend. No other crate calls NCCL.

Owner: metrale-comm. Invariants: none beyond the types.

Re-exports§

pub use nccl_backend::NcclBackend;

Modules§

nccl_backend
2026-09-26: NcclBackend, the multi-rank CommBackend over NCCL. Bootstrap: rank 0 creates an NCCL unique id and sends it over TCP to every other rank, then all ranks call ncclCommInitRank.

Structs§

SingleGpuBackend
2026-09-26: One rank: every collective and point-to-point operation returns Ok without doing anything, rank is 0 and world_size is 1.

Traits§

CommBackend
2026-09-26: Collective and point-to-point operations on raw device pointers and byte counts. A pointer is a u64 (a CUDA CUdeviceptr), so this crate does not depend on metrale-gpu-runtime.