# STrack: A Reliable Multipath Transport for AI/ML Clusters

arXiv: https://arxiv.org/abs/2407.15266 · HTML: https://mlsanity.com/paper/2407.15266/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jul 2024.

## Built on by (most cited first, 4 of 4)

- [RailS: Load Balancing for All-to-All Communication in Distributed Mixture-of-Experts Training](https://mlsanity.com/paper/2510.19262/) (2025)
- [Load Balancing for AI Training Workloads](https://mlsanity.com/paper/2507.21372/) (2025)
- [OpenURMA: A Clean-Room Open Implementation of the Unified Bus Protocol](https://mlsanity.com/paper/2605.28717/) (2026)
- [Incast-Free MoE Rate-Based Scheduling](https://mlsanity.com/paper/2607.26340/) (2026)

## Abstract (arXiv)

Emerging artificial intelligence (AI) and machine learning (ML) workloads present new challenges of managing the collective communication used in distributed training across hundreds or even thousands of GPUs. This paper presents STrack, a novel hardware-offloaded reliable transport protocol aimed at improving the performance of AI /ML workloads by rethinking key aspects of the transport layer. STrack optimizes congestion control and load balancing in tandem: it incorporates an adaptive load balancing algorithm leveraging ECN, while adopts RTT as multi-bit congestion indicators for precise congestion window adjustment. Additionally, STrack facilitates out-of-order delivery, selective retransmission, and swift loss recovery in hardware for multipath environment. The extensive simulation comparing STrack and RoCEv2 demonstrates that STrack outperforms RoCEv2 by up to 6X with synthetic workloads and by 27.4% with collective workloads, even with the optimized RoCEv2 system setup.