Bridge: Optimizing Collective Communication Schedules in Reconfigurable Networks with Reusable Subrings

TL;DR

Bridge introduces a reconfiguration strategy for optical networks that reuses links across multiple steps, significantly improving collective communication performance in AI/ML and HPC workloads.

Contribution

It leverages Bruck's pattern to enable sparse reconfiguration and reuse of optical links, reducing delays and enhancing throughput for collective primitives.

Findings

Reduces All-to-All completion time by 3x to 10x over static baselines.

Outperforms existing reconfiguration strategies for AllReduce, with up to 1.5x speedup.

Exceeds bandwidth-optimal Ring algorithm by 1.5x to 6.6x on certain workloads.

Abstract

Optical circuit-switched networks have emerged as an appealing alternative to electrical fabrics as they can reconfigure the network topology at runtime, reducing communication cost and improving bandwidth utilization. Yet exploiting optical reconfigurable networks for collective communication comes with a fundamental trade-off: each reconfiguration incurs non-negligible delay, communication must pause while the fabric reconfigures, and the benefit of a new topology depends on future traffic. The central question is therefore when reconfiguration is worth its cost. While prior work has demonstrated the benefits of reconfiguration, existing strategies use optical links only to optimize the current step, without reusing them for future steps. In this paper, we present Bridge, a reconfiguration strategy for important collective communication primitives used in AI/ML and HPC applications, namely All-to-All, AllReduce, Reduce-Scatter, and AllGather. Bridge exploits the structure of Bruck's communication pattern to support efficient sparse reconfiguration. The key idea is to reduce propagation and transmission delay by directly connecting immediate communication partners and preserve efficient reachability to future peers through connected subrings. As a result, optical links can be reused across multiple subsequent steps, allowing the benefit of reconfiguration to amortize beyond a single step. Our evaluation shows that Bridge reduces All-to-All completion time by typically $3\times$ to $10\times$ over static baselines even with millisecond-scale reconfiguration delays. For AllReduce, Bridge uniformly outperforms existing reconfiguration strategies, delivers up to $1.5\times$ speedup, and exceeds the bandwidth-optimal Ring algorithm by $1.5\times$ to $6.6\times$ on low to moderate-sized workloads.