Executive roundtable series

The Fabric Exchange

AI clusters are only as fast as the network stitching the GPUs together, and most enterprise networking was not built for AI-scale east-west traffic.

For network, data centre, infrastructure and technology leaders.

Region
APAC, six cities
Dates
October to November 2026
Access
By invitation
The premise

The fabric is the bottleneck.

// the GPUs are waiting on each other, not on work

Buy the accelerators, then discover the job is gated by how quickly they can talk. East-west traffic at AI scale behaves nothing like the north-south patterns enterprise networks were designed and sized around.

That is the starting point for this series. Not a product session, and not a briefing. A table of senior people working through where their fabric limits utilisation today, what visibility they have into it, and what a redesign realistically involves.

Six cities across APAC, between October and November 2026. Each table runs under the Chatham House Rule.

The evidence
1.6Tbps AI fabric platform capacity introduced for cluster interconnect
37.7% Year over year revenue growth reported for the second quarter of 2026

Arista Networks Q2 2026 financial results and product announcements, 2026. Figures pending client verification.

What the table will cover

Three questions on the table.

Each table works through the same three questions, shaped by what the room brings to them.

Utilisation you are actually getting

The first question is measurement: what percentage of theoretical cluster throughput each organisation realises in practice, and whether they can see where the rest goes.

// tension: bought capacity and usable capacity are different numbers

One operating system or many

This session looks at the cost of running different network operating systems across campus, data centre and AI cluster, and whether unifying them is worth the migration.

// differentiator: one operating system from client to AI centre

Cooling as a networking constraint

Liquid cooling changes what can sit in a rack and therefore what the fabric has to reach. The table works through how far physical constraints are now shaping network design decisions.

// context: liquid-cooled platforms for high density clusters
The series

Six cities. One fabric.

Each city is its own table with its own room, confirmed independently. Registration of interest is open now, with dates and venues confirmed city by city.

Customers see networking as the central nervous system for infrastructure from the client to campus to data and AI centers.

Jayshree Ullal Chairperson and CEO, Arista Networks
Before you register

Background reading.

Published Arista Networks material behind the questions on the table. Share your details once to unlock all three.

Platform AI networking and cluster fabric platforms Product EOS, one operating system across the network Resource library Arista technical resources and white papers
Request an invitation

Join us at the table.

Tell us which city suits you and we will come back to confirm your spot.

Require private transfer?
Who is convening this

About the hosts.

Arista Networks

Arista built its business on a single programmable network operating system running everywhere, at a time when the industry assumed each layer needed its own. That decision is why AI fabric was an extension rather than a pivot.

The company is a leader in data-driven, client-to-cloud networking for large AI, data centre, campus and routing environments.

Innovatus Media

Innovatus Media is a Sydney based B2B events agency that runs invitation-only executive roundtable series across APAC, EMEA and North America.

We convene senior decision makers for closed door conversations on the problems they are actually working on, and produce every element of the series end to end, from the room and the table to the follow up.

Questions first

Talk to the team.

Connect for more details on the series.