Initial release: NCCL Mesh Plugin for direct-connect RDMA topologies

- Enables NCCL over multi-subnet mesh topologies
- 8+ GB/s bandwidth over 100Gbps RDMA
- Successfully tested with distributed LLM inference (Mistral-7B)
- Custom subnet-aware NIC selection
- Background handshake thread for deadlock-free connection setup
This commit is contained in:
autoscriptlabs 2026-01-09 14:09:33 -05:00
commit 031bc48953
13 changed files with 3074 additions and 0 deletions

1508
src/mesh_plugin.c Normal file

File diff suppressed because it is too large Load diff