2 papers
cs.NI2026
HCCL: Collective Communication for Meta Training and Inference Accelerators
Wesley Bland, Tiago Antunes, Lars Paul Huse +63
We present HCCL, a collective communication library co-designed with Meta's MTIA 300 accelerator, the first Meta chip to integrate backend networking directly on chip package. MTIA…
cs.NI2022
Impact of RoCE Congestion Control Policies on Distributed Training of DNNs
Tarannum Khan, Saeed Rashidi, Srinivas Sridharan +3
RDMA over Converged Ethernet (RoCE) has gained significant attraction for datacenter networks due to its compatibility with conventional Ethernet-based fabric. However, the RDMA pr…