One network, every cloud: rethinking multi-cloud networking and security
Every cloud you add multiplies the networking problem. Each one brings its own load balancer stack and security toolchain, plus IP ranges that conflict with those you already have. Most teams respond by adding more layers, and the result is an architecture where no one can answer a simple question: what happened to this request, end to end?
Blame the architecture, not the tools. Most teams coupled where traffic flows with how it's configured, and never realized they were making that choice. Untangle those two, and multi-cloud networking and security stop being a per-environment project and become a single architecture.
That's the idea behind Universal Mesh: high-performance gateways at the boundaries of your environments, all governed by one control plane that discovers services from the platforms you already run. Cloud, on-prem, bare metal, legacy: one fabric and one policy layer, with a single view across all of it. Teams keep their existing workflows, and you get the network you've been trying to build.
In this session, you'll learn:
Why the sidecar-based mesh model breaks down across clouds, and what a gateway-based architecture does differently
How routing by logical names makes overlapping IP space a non-issue, no renumbering project required
A single security policy (mTLS, WAF, access control) enforced at every boundary instead of being rebuilt per cloud
Request tracing across clouds, regions, and clusters without touching application code
A migration path that moves service by service, with no cutover date and no big-bang moment
If you're managing a different networking stack in every environment, this session will show you the way out.
Speakers
