# Accelerating Geo-distributed Machine Learning with Network-Aware Adaptive Tree and Auxiliary Route

arXiv: https://arxiv.org/abs/2404.11352 · HTML: https://mlsanity.com/paper/2404.11352/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Apr 2024.

## Built on by (most cited first, 5 of 5)

- [TPI-LLM: Serving 70B-scale LLMs Efficiently on Low-resource Edge Devices](https://mlsanity.com/paper/2410.00531/) (2024)
- [TAH-QUANT: Effective Activation Quantization in Pipeline Parallelism over Slow Network](https://mlsanity.com/paper/2506.01352/) (2025)
- [Taming Latency and Bandwidth: A Theoretical Framework and Adaptive Algorithm for Communication-Constrained Training](https://mlsanity.com/paper/2507.17346/) (2025)
- [Learning In Chaos: Efficient Autoscaling and Self-Healing for Multi-Party Distributed Training](https://mlsanity.com/paper/2505.12815/) (2025)
- [Distributed LLM Pretraining During Renewable Curtailment Windows: A Feasibility Study](https://mlsanity.com/paper/2602.22760/) (2026)

## Abstract (arXiv)

Distributed machine learning is becoming increasingly popular for geo-distributed data analytics, facilitating the collaborative analysis of data scattered across data centers in different regions. This paradigm eliminates the need for centralizing sensitive raw data in one location but faces the significant challenge of high parameter synchronization delays, which stems from the constraints of bandwidth-limited, heterogeneous, and fluctuating wide-area networks. Prior research has focused on optimizing the synchronization topology, evolving from starlike to tree-based structures. However, these solutions typically depend on regular tree structures and lack an adequate topology metric, resulting in limited improvements. This paper proposes NetStorm, an adaptive and highly efficient communication scheduler designed to speed up parameter synchronization across geo-distributed data centers. First, it establishes an effective metric for optimizing a multi-root FAPT synchronization topology. Second, a network awareness module is developed to acquire network knowledge, aiding in topology decisions. Third, a multipath auxiliary transmission mechanism is introduced to enhance network awareness and facilitate multipath transmissions. Lastly, we design policy consistency protocols to guarantee seamless updates of transmission policies. Empirical results demonstrate that NetStorm significantly outperforms distributed training systems like MXNET, MLNET, and TSEngine, with a speedup of 6.5~9.2 times over MXNET.