Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2605.26960

Extreme-Scale Interconnection Networks

Alejandro Cano·Cristina Brinza·Cristóbal Camarero·Carmen Martínez·Ramón Beivide
View original source ↗Read source document ↗Export metadata ↓

Abstract

Extreme-scale data centers are the backbone of next-generation computing, enabling breakthroughs in science, artificial intelligence, and global innovation through unprecedented processing power and scalability. This work examines leaf-spine network topologies that offer extreme scalability--connecting a vast number of endpoints--while delivering strong performance at low cost. It takes as a starting point two alternatives to the widely used Fat-Tree topology: the Orthogonal Fat-Tree and the Random Folded Clos. The resulting Multipass Random Leaf-Spine (MRLS) networks inherit their advantages and surpass Fat-Trees in both throughput and flexibility. To fully leverage the topological properties of these networks, various non-minimal routing strategies are considered. An exhaustive evaluation using an interconnection network simulator provides insight into the trade-offs and scalability of these topologies under realistic conditions, positioning them as a promising solution for extreme-scale systems. The MRLS achieves a 50% speedup against a Fat-Tree for an All2All collective comprising 100k endpoints, and 100% against Dragonfly networks for the same collective.

Explore related subjects

cs.NIcs.DC

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alejandro Cano, Cristina Brinza, Cristóbal Camarero, Carmen Martínez, Ramón Beivide. 2026-05-26. Extreme-Scale Interconnection Networks. https://arxiv.org/abs/2605.26960

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Formal Abstractions for Packet Scheduling

Early programming models for software-defined networking (SDN) focused on basic features for controlling network-wide forwarding paths, but more recent work has considered richer features, such as packet scheduling and queueing, that affect performance. In particular, PIFO trees, proposed by Sivaraman et al., offer a flexible and efficient primitive for programmable packet scheduling. Prior work has shown that PIFO trees can express a wide range of practical algorithms including strict priority, weighted fair queueing, and hierarchical schemes. However, the semantic properties of PIFO trees are not well understood. This paper studies PIFO trees from a programming language perspective. We formalize the syntax and semantics of PIFO trees in an operational model that decouples the scheduling policy running on a tree from the topology of the tree. Building on this formalization, we develop compilation algorithms that allow the behavior of a PIFO tree written against one topology to be realized using a tree with a different topology. Such a compiler could be used to optimize an implementation of PIFO trees, or realize a logical PIFO tree on a target with a fixed topology baked into the hardware. To support experimentation, we develop a software simulator for PIFO trees, and we present case studies illustrating its behavior on standard and custom algorithms.

cs.NI↗
arXiv2026

Decoding Delay Guarantees of Space Regulated Multiple Access Random Wireless Networks using Successive Interference Cancellation

This paper is focused on decoding delay guarantees in wireless networks, where messages have a given signal-to-interference-plus-noise ratio threshold $η_0$ to meet in order to be successfully decoded, and where transmissions should occur within some strict time constraints. Its main contribution consists in quantifying the worst-case transmissions decoding delays in the uplink of cellular and cell-free networks using successive interference cancellation. We show how such decoding delay guarantees can be obtained using spatial network calculus, a new tool introduced recently, and in particular spatial regulation. The results rely on the assumption of absence of fading. We nevertheless outline what this approach will lead to in the fading case for cellular networks.

cs.NI↗
arXiv2026

Toward Composable Network Digital Twins: A Subgraph-Based Latency Prediction Study

Modern networks must support changing topologies, configurations, and performance objectives, motivating fast and reliable performance estimation. Network digital twins (NDTs) enable what-if analysis for performance estimation in such network scenarios, however, existing machine learning-based NDT approaches often rely on entire topology representations, which are inherently monolithic and lack reusability under topological or traffic changes in the network. This paper introduces a composable NDT approach that decomposes networks into subgraphs represented by reusable unit twins that capture subgraph structure, configuration and traffic behaviours. A lightweight composer aggregates unit twin combinations to create NDTs that predict per-route end-to-end latency through an overall topology. Evaluation across controlled synthetic topologies and diverse traffic scenarios, real-world Topology Zoo topologies, and a public NDT challenge dataset demonstrates that the composable NDTs achieve high in-distribution accuracy while remaining stable under out-of-distribution scenarios. Comparison with monolithic full topology NDTs demonstrates that our composable approach achieves reusability, while achieving comparable or superior accuracy.

cs.NI↗