network-architecture

Tag

Cards List
#network-architecture

@cevenif: Another tool that instantly turns ordinary people into architects is here — drag and drop in the browser, and the data flow between servers, proxies, and databases is all clearly drawn for you without touching a single line of code. In short: can't draw architecture diagrams? No problem, now this barrier has been removed for you too. https://github.co…

X AI KOLs Timeline · 2026-06-27 Cached

Netviz is a browser-based visual tool that allows users to design network architecture diagrams via drag-and-drop, without writing code. Supports adding components such as servers, proxies, databases, connecting them to map data flows, and customizing details.

0 favorites 0 likes
#network-architecture

Goodbye, Leaf-and-Spine Networks?

Lobsters Hottest · 2026-06-10 Cached

A critical analysis of an AWS paper claiming a radical new network design that outperforms leaf-and-spine fabrics. The author argues the idea is not new, compares it to Plexxi's failed approach, and points out flaws in the throughput and randomness claims.

0 favorites 0 likes
#network-architecture

@MaxForAI: http://Z.ai and this ZCube paper from Tsinghua—worth a read for anyone in Infra. Many people's first reaction when talking about AI infra is still GPU, memory, quantization, and inference frameworks. But once you get into long context and Prefill-Decode separation, the network is no longer just a 'supporting role' in the data center. Every...

X AI KOLs Timeline · 2026-05-21

ZCube is a new network architecture that flattens the topology and mixes single/multi-rail access to optimize KV Cache transmission in long-context and PD separation scenarios. In the GLM-5.1 production cluster, it achieved a 33% reduction in switch/optical module costs, a 15% increase in GPU inference throughput, and a 40.6% decrease in TTFT P99.

0 favorites 0 likes
#network-architecture

@Zai_org: https://x.com/Zai_org/status/2057216685040443743

X AI KOLs Timeline · 2026-05-20 Cached

This paper presents ZCube, a novel network architecture developed by Z.ai, Harnets.AI, and Tsinghua University to address topology-induced congestion in Prefill-Decode disaggregated LLM inference clusters. Production deployments on GLM-5.1 coding workloads achieved a 33% reduction in network CapEx, 15% throughput improvement, and 40.6% reduction in TTFT P99 latency.

0 favorites 0 likes
← Back to home

Submit Feedback