Own the control plane
Run PallasDB inside your boundary, wire it to your observability stack, and keep rollout authority close to the operators who carry the pager.
PallasDB is a self-hostable distributed key value store written in Go, designed for teams that want cloud-portable deployment without giving up operational control.
Built for the infrastructure team that wants storage to be explicit, portable, and inspectable.
The page leaves clear product surface for future AWS, GCP, and CRD controller actions while making the core promise unmistakable today.
Run PallasDB inside your boundary, wire it to your observability stack, and keep rollout authority close to the operators who carry the pager.
A Go-native storage layer shaped for direct deployment, predictable packaging, and infrastructure teams that prefer explicit systems.
Reserve clean space for direct deploy buttons without making the product dependent on one cloud story.
The upcoming CRD controller belongs in the main narrative, not buried in an integration footnote.
A distributed store should make placement, replication, and recovery legible before it makes them magical.
Bootstrap nodes, declare topology, and let operators see what joined, moved, or recovered.
AWS and GCP entry points are staged as primary actions while keeping self-hosting at the center.
The controller path is designed for reconciliation, not a decorative Kubernetes checkbox.
The CRD controller can become the operational interface for clusters, upgrades, and placement policy without hiding the underlying distributed system.
Declarative cluster intent
Clear distributed-system mechanics
Cloud deployment without cloud lock-in
Add the AWS, GCP, and controller entry points when they are ready. The action area is already structured for clean cloud-specific deployment paths.