English

Scaling Distributed Machine Learning with In-Network Aggregation

Distributed, Parallel, and Cluster Computing 2020-10-01 v2 Machine Learning Networking and Internet Architecture Machine Learning

Abstract

Training machine learning models in parallel is an increasingly important workload. We accelerate distributed parallel training by designing a communication primitive that uses a programmable switch dataplane to execute a key step of the training process. Our approach, SwitchML, reduces the volume of exchanged data by aggregating the model updates from multiple workers in the network. We co-design the switch processing with the end-host protocols and ML frameworks to provide an efficient solution that speeds up training by up to 5.5×\times for a number of real-world benchmark models.

Keywords

Cite

@article{arxiv.1903.06701,
  title  = {Scaling Distributed Machine Learning with In-Network Aggregation},
  author = {Amedeo Sapio and Marco Canini and Chen-Yu Ho and Jacob Nelson and Panos Kalnis and Changhoon Kim and Arvind Krishnamurthy and Masoud Moshref and Dan R. K. Ports and Peter Richtárik},
  journal= {arXiv preprint arXiv:1903.06701},
  year   = {2020}
}
R2 v1 2026-06-23T08:09:43.449Z