The next-gen
edge K8s

Easily spin up and manage K8s and applications across thousands of edge locations... “plug-n-play” style.

Request a demo

Easily deploy and manage thousands of edge locations

Say no to costly field engineer visits or risky upgrades with Palette Edge

Plug`n`play, low touch approach

Just plug your pre-configured bare metal server to power and Internet and let the Palette do the rest.

Risk-free rolling upgrades

Perform rolling upgrades even on single-server configurations due to Palette’s unique failsafe design.

VM and container app support

Deploy and manage both VM-based and container-based application and manage the full stack with consistency.

Time to rethink what edge K8s means

Edge can be challenging, especially when adding the complexity of Kubernetes. With no centralized IaaS-like management, no IT skills on-site and often unreliable connectivity, conventional bare metal approaches just won’t work.

And what’s point of doing edge in the first place if you have to rely on costly field engineer visits for an upgrade or increasing the number of edge hardware for the management plane not to become a bottleneck while scaling?

By extending our Real Metal architecture and customizing it for edge locations with a unique, zero-downtime design, you can now run both containerized and VM-based applications, without worrying about downtime during upgrades or configuration drift.

Palette Edge is purposely-designed for efficiency, engineered to work seamlessly for single-server edge locations. All you have to do is power up and connect your pre-configured device to the Internet and it will do the rest. With an A/B OS partion and multi-node Kubernetes deployment, you can apply OS and K8s updates with easy rollback and zero risk of downtime. And you can run both container and VM-based application on top for maximum flexbility.

An architecture that can actually scale

Palette Edge is designed for infinite scale to tens of thousands of locations.

As opposed to other solutions, all policy enforcement and operations happen locally “at-cluster”, independently of the centralized management plane, meaning no performance impact to the overall architecture.

As with all other Palette editions, you can still choose from a public repo the combination of layers for your K8s infrastructure and application services, or bring your own to manage them consistently across the full stack.

How it works

Wanna see Palette in action?

Are you interested in a quick 30-minute live demo?

Go ahead and submit your information and someone will get back to you shortly.