Bine Trees: Enhancing Collective Operations by Optimizing Communication Locality
arXiv:2508.17311 · doi:10.1145/3712285.3759835
Abstract
Communication locality plays a key role in the performance of collective operations on large HPC systems, especially on oversubscribed networks where groups of nodes are fully connected internally but sparsely linked through global connections. We present Bine (binomial negabinary) trees, a family of collective algorithms that improve communication locality. Bine trees maintain the generality of binomial trees and butterflies while cutting global-link traffic by up to 33%. We implement eight Bine-based collectives and evaluate them on four large-scale supercomputers with Dragonfly, Dragonfly+, oversubscribed fat-tree, and torus topologies, achieving up to 5x speedups and consistent reductions in global-link traffic across different vector sizes and node counts.
References in corpus (8)
- Gemini: A Family of Highly Capable Multimodal Models
- Synthesizing Optimal Collective Algorithms
- TACOS: Topology-Aware Collective Algorithm Synthesizer for Distributed Machine Learning
- Mitigating Network Noise on Dragonfly Networks through Application-Aware Routing
- Exploring GPU-to-GPU Communication: Insights into Supercomputer Interconnects
- Uno: A One-Stop Solution for Inter- and Intra-Datacenter Congestion Control and Reliable Connectivity
- Bine Trees: Enhancing Collective Operations by Optimizing Communication Locality
- Modeling and Analysis of Application Interference on Dragonfly+