2026/08/01 by Wesley Bland, Tiago Antunes, Lars Paul Huse +63
Computer Science · #cs.NI #cs.DC
12 pages, 17 figures, to be published in the proceedings of "SC '26: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis"
arxiv created 2026/08/01 · arxiv updated 2026/08/04
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 300 includes dedicated message engines (MEs) with near-memory compute (NMC) that fully offload collective execution from the compute grid, enabling large overlap between computation and communication. HCCL uses a compiled communication model in which the host generates a complete description of each collective including dependencies. We describe the control and data path architecture, topology-aware algorithm selection across MTIA 300's asymmetric scale-up and scale-out network, and optimizations for both training and inference workloads. For training, HCCL achieves up to 940 GB/s on intra-rack collectives while introducing less than 0.5% degradation to concurrent compute throughput. For inference, we leverage one-sided communication primitives that bypass the scheduling path to minimize collective latency and describe collective designs that improve compute-communication pipelining for latency-sensitive workloads.