ZeusK8s
Google GKE

Google GKE, managed the same way as everything else.

Add GKE clusters to the same console, workflow, and service definitions you already use — no separate tooling, no separate mental model.

One workflow that happens to also do Google.

Every cloud you add normally means a new tool, a new set of IAM concepts, and someone who has to learn both. The result is engineers everywhere exiting panic mode and pretending they weren't just googling "how to set up GKE networking." ZeusK8s makes GKE feel identical to EKS and to a cluster in your rack, because the workflow on top is the same — only the execution underneath differs.

Link a Google Cloud project, provision GKE, map workloads to Google service accounts with Workload Identity, and join it into the same global fabric as your AWS clusters. A service you defined once runs on all of them.

Zeus · Clusters
Clusters
6 clusters · 3 providers
Live New Cluster
NameProviderRegionVersionNodesStatus
production-usAWS EKSus-east-11.306Ready
production-euGKEeurope-west31.304Ready
stagingAWS EKSus-west-21.29→1.30 ⚠3Ready
edge-apacGKEasia-east11.302Ready
HQ Denverk3sDenver, COv1.35.55Ready
prox-test01ProxmoxProvo, UTv1.34.1Provisioning
How it works

From zero to running.

01

Connect a Google Cloud project

Link projects with a scoped service account. Zeus manages GKE, DNS (Cloud DNS), and Workload Identity from there.

02

Provision GKE

Pick region, node pools, and architecture. Networking and identity match your other clusters, so GKE behaves like everything else.

03

Map Workload Identity

Bind pods to Google service accounts with no exported keys: the GCP equivalent of IRSA, configured in the same identity UI.

04

Join the global fabric

Connect GKE to the same encrypted overlay as your EKS and private clusters. Services reach each other across clouds by name, by real IP.

The specifics

Built by people who run this in production.

No hand-waving. Here’s what’s actually under the hood: the kind of detail you’d expect from a platform you’re going to trust with production.

Account model
Multi-project linking with scoped service accounts
Networking
Encrypted overlay; one private network across clouds
Identity
GCP Workload Identity: keyless pod auth
DNS
Google Cloud DNS managed alongside Route 53
Service definitions
Identical to EKS: one template, both clouds
Observability
Same live pods / logs / metrics views
Straight answers

Questions you’d actually ask.

Is GKE a first-class citizen or an afterthought?

First-class. The entire value of ZeusK8s is that EKS, GKE, and private clusters share one UX. A service you define deploys to GKE the same way it deploys to EKS: same screen, same steps.

Can EKS and GKE clusters talk to each other?

Yes, that’s the point of the global fabric. Over the encrypted overlay, a pod in GKE reaches a service in EKS by name and real IP, with no NAT and no public exposure.

Do I manage Google IAM separately?

No. Per-service Google identities are configured in the same identity UI as AWS IRSA, so cloud permissions live next to the service that needs them.