CORTEXA
← Browse
arxivcs.NI2026-07-22

MoX: Efficient MoE Routing on Direct-Connect Topologies

Ori Cohen, Jakob Krebs, Daniel Amir, Mark Silberstein

Optically switched networks suit the regular communication of dense ML models, but MoE introduces sparse, runtime-dependent traffic. We show that efficient offline-optimized routing enables efficient MoE training and inference on direct-connect topologies without the need for MoE traffic matrix or dynamic topology reconfiguration. MoX constructs token-aware multicast trees to reduce bandwidth tax, then uses static, precomputed link weights to balance traffic by solving a restricted multicast tree-packing problem. Using recorded traffic from large MoE models, token-level traces, and ASTRA-sim, we find that MoX accelerates the full MoE block -- dispatch, expert computation, and combine -- by up to 1.8x over min-hop routing. Moreover, it attains nearly ideal packet-switched network performance in random expander topologies. On a 1,024-TPU model of Google's Boardfly topology, MoX reduces the dispatch bottleneck link load by up to 47%. These results show that high-performance MoE on static direct-connect fabrics can be achieved via optimized load-oblivious routing without demand-driven reconfiguration.

View free PDFSource page

Related papers

arxivcs.DCcs.AIcs.NI2026-07-17

Scalable LLM Agent Tool Access in the Cloud

Mingxin Li, Enge Song, Yueshang Zuo, Xiaodong Liu, Rong Wen, Qiang Fu, et al.

LLM agents increasingly rely on tool calling to act on external systems, and the Model Context Protocol (MCP) has quickly become its de facto interface. Operating MCP at cloud scale, however, becomes difficult. On the tool provider side, legacy services are not directly callable…

View free PDFSource page
arxivcs.AIcs.LGcs.NI2026-07-15

AI-Driven Multi-Hop Relay Selection for Smart Urban NR-V2X Networks via Learning-to-Optimize Graph Neural Networks

Giambattista Amati, Federica Mangiatordi, Simone Angelini, Emiliano Pallotti, Pierpaolo Salvo

Reliable and low-latency NR-V2X communications are essential for smart mobility in dense urban environments. However, limited Road-Side Unit (RSU) density, frequent non-line-of-sight conditions, and highly dynamic vehicular topologies often prevent many Connected and Automated Ve…

View free PDFSource page
arxivmath.COcs.NI2026-07-02

Beneš and Shuffle-Exchange Counterexamples

Przemek Chojecki

We give explicit counterexamples to two rearrangeability conjectures for shuffle-type networks. First, for every $N\ge2$ we construct a simple $N$-regular ordered two-stage graph $L_N$ with $F(L_N)=2$ and $R(L_N)\ge N$, refuting the graph-theoretic Beneš inequality $R(L)\le2F(L)$…

View free PDFSource page
arxivcs.NI2026-07-23

Large Language Model Assisted Intent-Based Satellite-Integrated Access and Backhaul FWA for Rural Areas

Anselme Ndikumana, Kim Khoa Nguyen, Adel Larabi, Mohamed Cheriet

Rural areas exhibit low population density and highly variable connectivity needs shaped by both household usage and field operations such as planting, harvesting, and mining. These field activities often occur in isolated locations requiring temporary connectivity, whereas rural…

View free PDFSource page