Expanding across time to deliver bandwidth efficiency and low latency
Not on the map. Why?
Abstract (arXiv)
Datacenters need networks that support both low-latency and high-bandwidth packet delivery to meet the stringent requirements of modern applications. We present Opera, a dynamic network that delivers latency-sensitive traffic quickly by relying on multi-hop forwarding in the same way as expander-graph-based approaches, but provides near-optimal bandwidth for bulk flows through direct forwarding over time-varying source-to-destination circuits. The key to Opera's design is the rapid and deterministic reconfiguration of the network, piece-by-piece, such that at any moment in time the network implements an expander graph, yet, integrated across time, the network provides bandwidth-efficient single-hop paths between all racks. We show that Opera supports low-latency traffic with flow completion times comparable to cost-equivalent static topologies, while delivering up to 4x the bandwidth for all-to-all traffic and supporting 60% higher load for published datacenter workloads.
Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). First posted on arXiv 29 Mar 2019.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2022: 1; 2023: 1; 2024: 0; 2025: 4; 2026: 1.
7 papers.
- 2023Rail-only: A Low-Cost High-Performance Network for Training LLMs with Trillion Parameterscited by 13
- 2025MixNet: A Runtime Reconfigurable Optical-Electrical Fabric for Distributed Mixture-of-Experts Trainingcited by 7
- 2025InfiniteHBD: Building Datacenter-Scale High-Bandwidth Domain for LLM with Optical Circuit Switching Transceiverscited by 7
- 2022Efficient Direct-Connect Topologies for Collective Communicationscited by 5
- 2025LUMION: Fast Fault Recovery for ML Jobs Using Programmable Optical Fabricscited by 0
- 2025Morphlux: Transforming Torus Fabrics for Efficient Multi-tenant MLcited by 0
- 2026Assistants, Not Architects: The Role of LLMs in Networked Systems Designcited by 0
- 2026Assistants, Not Architects: The Role of LLMs in Networked Systems Design28 Apr 2026
- 2025Morphlux: Transforming Torus Fabrics for Efficient Multi-tenant ML20 Jul 2025
- 2025LUMION: Fast Fault Recovery for ML Jobs Using Programmable Optical Fabrics29 May 2025
- 2025InfiniteHBD: Building Datacenter-Scale High-Bandwidth Domain for LLM with Optical Circuit Switching Transceivers6 Feb 2025
- 2025MixNet: A Runtime Reconfigurable Optical-Electrical Fabric for Distributed Mixture-of-Experts Training7 Jan 2025
- 2023Rail-only: A Low-Cost High-Performance Network for Training LLMs with Trillion Parameters22 Jul 2023
- 2022Efficient Direct-Connect Topologies for Collective Communications7 Feb 2022