# MCCL: Fault-Tolerant Collective Communication for Large-Scale GPU Training

PyTorch Conference North America 2026 · [All sessions](https://sessionize.com/api/v2/s2rq62nk/view/markdown)

- **Session ID:** 1256858
- **When:** Wednesday, October 21, 2026 5:30 pm - 5:55 pm (Pacific Standard Time, UTC-08)
- **Room:** 210A (Concourse Level)
- **Speakers:** Ben Carver

## Description

As AI training workloads scale to tens of thousands of GPUs, hardware failure has become a routine operating condition rather than an exception. Traditional collective communication libraries, which rely on static communicator membership, force costly, full-job restarts whenever a failure occurs. This talk introduces the topic of large-scale training using the Meta Collective Communication Library (MCCL), an open-source, fault-tolerant C++ library designed and developed at Meta to address these challenges.

MCCL redefines the communication lifecycle by treating hardware failure as a first-class design principle, enabling dynamic communicator reconfiguration (shrinking and growing as ranks fail or recover) without interrupting the overall training job. We will demonstrate how MCCL integrates with Torchcomms to provide production-grade, fault-tolerant collective operations, including AllReduce and Send/Recv, while maintaining performance parity with industry standards like NCCL. Attendees will learn about MCCL’s architecture and how to implement robust training loops that recover gracefully from infrastructure instability.

## Speaker

- [Ben Carver](https://sessionize.com/api/v2/s2rq62nk/view/markdown/speaker/92f2fbc0-20d1-49af-9474-c8055566f5bd) — Research Scientist, Meta Platforms