Deployment guide
This guide walks through installing Superphenix end to end: deploying the management plane, installing the operating system, and configuring your availability zones. Whether you are starting with a single hyperconverged cluster or a large decoupled deployment across multiple AZs, use this page for prerequisites and to choose the right installation path.
Requirements
Before you begin, confirm that your environment meets these specifications and follows our recommended patterns:
- Kubernetes: A Kubernetes cluster (v1.28+) to host the Superphenix operator and management plane.
- Helm: Required for the initial operator installation.
- Hardware & network: See Hardware requirements and Network requirements.
- Best practices: See Production recommendations for sizing and high availability.
Installing the Management Plane
Start by deciding where the management plane runs. It installs and operates the rest of Superphenix, manages cluster lifecycle, and provides operator and tenant access across your infrastructure.
Superphenix supports two placement models: the management plane can run inside an AZ or on a dedicated management cluster separate from your workload AZs. Your choice affects connectivity requirements and failure domains.
See Deployment topology for a full comparison.
-
Installing inside an AZ
The management plane runs on one of the workload clusters. Easier to bootstrap, best for single-AZ or isolated deployments. Requires a pre-existing Talos cluster on that AZ.
-
Installing outside an AZ
The management plane runs on a dedicated management cluster, separate from workload AZs. Recommended for multi-AZ orchestration and maximum redundancy. Requires API connectivity to every Superphenix cluster you manage.
Installing the Operating System
Every Superphenix cluster runs on Talos Linux. Talos bootstraps and hosts Kubernetes on your bare-metal nodes; Superphenix then deploys storage, virtualization, and networking on that Kubernetes foundation.
Superphenix supports two provisioning models: bootstrap Talos clusters yourself, or let the Superphenix provision servers declaratively over BMC/IPMI. Manual installation is typically required when the management plane runs inside an AZ; automated installation pairs well with a dedicated management cluster orchestrating greenfield or multi-AZ datacenters.
-
Manual OS installation
Install Talos and bootstrap Kubernetes with
talosctl. Best for labs, first clusters, or when the management plane runs inside an AZ and you need Talos in place before installing the operator. -
Automated OS installation
Let Superphenix provision servers over BMC/IPMI and manage node lifecycle declaratively. Best for greenfield datacenters and multi-AZ deployments with the management plane outside workload AZs.
Installing an Availability Zone
Each availability zone runs the full Superphenix stack on Talos Kubernetes clusters. Choose how storage and workloads are deployed within the AZ. You can install multiple AZs with different topologies under the same management plane.
Superphenix supports both hyperconverged (with the storage and workload running on the same cluster) and decoupled (separate clusters for the storage and the workload) installations.
See Deployment topology for a full comparison.
-
Installing hyperconverged
Storage and workloads run on the same cluster. The simplest model for single-AZ deployments, labs, and environments that need rapid horizontal scaling.
-
Installing decoupled
Dedicated storage and workload clusters operate separately. Better performance isolation and support for shared storage across multiple workload AZs.