The paper introduces REACT, a system that dynamically tunes communication collectives in distributed AI training to mitigate congestion without requiring network infrastructure changes. REACT operates at the application layer, detecting congestion via flow statistics and adjusting the pattern of data exchange—such as selecting different aggregation nodes in an AllReduce tree—while preserving the semantics of the communication. Evaluations on a shared academic GPU cluster show that REACT improves algorithm bandwidth by 13%–38% under congestion, with simulations indicating potential gains up to 75%.
By Eashan Gupta, Yongzhou Chen, Apoorve Mohan, Pavlos Maniotis, Abdullah Kayi, Radhika Mittal
arXiv:2608. 07733v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) are widely used across domains such as natural sciences, social network analysis, chip design, and recommendation systems.
By Liad Gerstman, Aditya Dhakal, Dejan Milojicic, Avi Mendelson
arXiv:2605. 01989v2 Announce Type: replace Abstract: Distributed machine learning (ML) training has become a necessity with the prevalence of billion to trillion-parameter-scale models.
By Zechen Ma, Zixi Qu, Jinyan Yi, David Lin, Yashar Ganjali
arXiv:2607. 28633v1 Announce Type: cross Abstract: Disaggregated LLM inference creates a datacenter networking problem that no existing system solves correctly.
By Sanjeev Rao Ganjihal
arXiv:2608. 06441v1 Announce Type: new Abstract: Full-graph GNN training delivers high accuracy but scales poorly on multi-server clusters due to heavy, irregular inter-node embedding exchanges.
By Guofan Yu, Sitian Chen, Zhenheng Tang, Xiaowen Chu, Amelie Chi Zhou
arXiv:2608. 06046v1 Announce Type: cross Abstract: AI training workloads are growing rapidly, making their time, energy, and infrastructure costs increasingly important.
By Yutong Zhao, Noga H. Rotman, Gianni Antichi, Ran Ben Basat