Alphie — System Requirements

1. Supported operating systems

Alphie is intended to run on 64-bit Linux distributions commonly used for servers:

• Red Hat Enterprise Linux 9 and compatible:
  • Rocky Linux 9
  • AlmaLinux 9
• Ubuntu Server 22.04 LTS or Debian 12 (Bookworm)

A minimal, server-style install (no desktop environment) is recommended. SELinux can be enforced or permissive as long as Alphie’s policies are applied correctly.

2. Components on the Alphie controller

An Alphie controller typically runs:

• The Alphie API + UI service (FastAPI backend + compiled binary frontend).
• A reverse proxy (usually nginx).
• A PostgreSQL database (for small and trial setups) or connects to an external PostgreSQL cluster.
• Local log storage under /opt/alphie/logs and job/pipeline artifacts under /var/alphie.

For production Pro/Enterprise environments, we strongly recommend using an external PostgreSQL server or cluster instead of running the DB on the controller.

3. Sizing by edition

The tables below assume a “typical” environment with average-sized playbooks and reasonable logging. Heavy workloads (lots of forks, large inventories, or noisy debug logging) should lean toward the higher end or next tier up.

3.1 Alphie Trial (small lab / PoC)

The trial edition is capped in features (limited runbooks, pipelines, automation sets, runners, etc.), so it does not need a big server. A single VM can run the controller, local PostgreSQL, and a small amount of local logging.

Good use cases: single admin or small team, lab environments, demos, or quick evaluations.

3.2 Alphie Pro (single-controller production)

Pro is aimed at “real” production use for a team or department: more users, more targets, more automation sets, and higher log volume. Here the controller should be treated more like an Ansible Tower/AAP controller in terms of resources, with Alphie still being somewhat lighter.

3.3 Alphie Enterprise (HA / large team)

Enterprise assumes multiple teams, large inventories, heavy logging, and more frequent automation. At this tier, it’s common to separate roles onto different nodes and use multiple runners.

As a rule of thumb, if you expect hundreds of forks across multiple runners, budget at least 1 GB of RAM for every 10–15 active forks on top of the base OS + Alphie overhead.

4. Runner host requirements

Alphie runners execute Ansible, Terraform, and other workload containers. Their sizing depends mostly on what your jobs do, not on Alphie itself.

4.1 Baseline runner (light to moderate workloads)

4.2 Heavy runner (large inventories or big playbooks)

• 4–8 vCPUs
• 16–32 GB RAM
• 100+ GB disk on SSD if:
  • You use large container images or many execution environments.
  • You keep logs and artifacts locally for a long time.

Instead of one giant runner, it’s often better to run multiple mid-sized runners and let Alphie distribute work across them.

5. Disk and log retention planning

Alphie writes job and pipeline logs, per-step artifacts, and optional backups. To avoid surprises:

• Keep at least 40 GB available under /var/alphie on Pro/Enterprise controllers.
• Use a separate filesystem or volume for /var/alphie if possible.
• Enable Alphie’s log cleanup scripts or external rotation to prune old logs and temp files.
• For Enterprise, consider a separate archive node or object storage for long-term log retention.

6. Network & connectivity

• The controller must reach all runners over SSH (or your configured runner transport) on the configured port.
• Runners must reach their targets just like a normal Ansible control node would (SSH/WinRM, HTTPS APIs, etc.).
• Users access the Alphie UI over HTTPS (typically port 443) through a reverse proxy such as nginx.
• If using external PostgreSQL:
  • Open TCP 5432 (or your DB port) between controller and DB nodes.
  • Use TLS + strong authentication for the DB connection where possible.

7. Quick sizing cheat sheet

• Trial: 2 vCPU, 4–8 GB RAM, 40–60 GB disk, all-in-one VM.
• Pro: 4+ vCPU, 16–32 GB RAM, 80–160 GB disk, external PostgreSQL recommended.
• Enterprise: 8+ vCPU, 32+ GB RAM on the controller, external DB cluster, multiple runners.
• Runner: 2–4 vCPU and 4–8 GB RAM for normal work; scale up or add more runners for heavy jobs.

Start with these baselines, monitor CPU, RAM, disk, and job runtimes, and then scale your controller, database, and runners to match your actual workload.