Element 13.15
A community cloud on reused hardware
- Sheet
- 13.15
- Chapter
- 13 Digital infrastructure
- State
- Drafted
- Release
- 16 November 2026
- Editor
- Contributed by Praecise as open research
- Reviewers
- To be named at review
- Safety
- S1, as the chapter
- Licence
- CC BY-SA 4.0
- Resources
- 8
The element
- Requirement
The community's services run as virtual machines or containers across more than one reused machine, so that a machine can be repaired, replaced or lost without the services stopping for long, and the keepers can see and manage every machine from one place.
- How to tell
- A service can be moved to another machine and back in a planned test.
- Every machine is watched: a monitoring system collects metrics and raises alerts by rules (Prometheus).
- Each host has the memory for the platform plus its guests: Proxmox VE recommends at least 2 GB for itself plus guest memory, and roughly 1 GB more per terabyte of used storage with ZFS or Ceph (Proxmox VE System Requirements).
- Management traffic runs on its own network, as the Ceph guidance advises for management and controller traffic (Ceph hardware recommendations).
- Options
- Proxmox VE. Combines KVM virtual machines and Linux containers, with clustering, high availability, live migration and Ceph or ZFS storage, under the AGPL version 3, with paid support available (Proxmox Virtual Environment).
- XCP-ng. A Xen-based platform managed through Xen Orchestra, described as a Xen Project incubation project hosted by the Linux Foundation; community and commercial support (XCP-ng).
- K3s, a small Kubernetes distribution. For services packaged as containers. Its site describes it as built for unattended, resource-constrained and remote places, as a single binary under 70 MB on ARM and x86 (K3s). A server node needs at least 2 cores and 2 GB of memory and an agent 1 core and 512 MB; SSDs are recommended and SD cards are warned against (K3s requirements).
- Remote management. Servers whose management controller supports Redfish, the DMTF standard, can be powered, inspected and repaired over the network (Redfish Developer Hub).
- A single machine. Still a valid start for 30 adults (13.1); the cluster is the next step.
- Settled, open, contested
Settled. Virtualisation and containers let several services share reused machines and move between them (Proxmox Virtual Environment, XCP-ng, K3s). Open. Published sizing for community clouds of 30 or 300 people was not found. Contested. Virtual machines against containers: virtual machines run any operating system and isolate more; containers are lighter but assume a shared kernel and more specialised skills.
The four stages
- Learn: the difference between virtual machines and containers, and what high availability needs (Proxmox Virtual Environment, K3s).
- Design: count services and their memory; size hosts (Proxmox VE System Requirements, K3s requirements); plan a separate management network (Ceph hardware recommendations).
- Build: install the platform on each host, join them into a cluster, and add monitoring from the first day (Prometheus).
- Deploy: rehearse moving services; read alerts daily; record every change.
Resources at each stage
2 Design
Size and choose
- Ceph hardware recommendations Guide, S1
- K3s requirements Datasheet, S1
- Proxmox Virtual Environment Tool, S1
- Proxmox VE System Requirements Guide, S0
3 Build
Build or obtain
- Redfish Developer Hub Standard, S0
- K3s Tool, S1
- Proxmox Virtual Environment Tool, S1
- XCP-ng Tool, S1
4 Deploy
Install, run and measure
- Redfish Developer Hub Standard, S0
- K3s Tool, S1
- Prometheus Tool, S0
- Proxmox Virtual Environment Tool, S1
- XCP-ng Tool, S1
What changes between 30 adults and 300 people
At 30 adults one host, or two with one holding backups, runs every service; monitoring can be a single dashboard. At 300 people three or more hosts form a cluster so that one can fail, the management network is separate, alerts go to a rota, and the platform's own updates are scheduled. No source opened gives server sizing per number of people; sizing follows each service's stated requirements (13.1).
Reference designs
Open for contributions Designs with drawings, specifications and a bill of materials. Add to it
Parts list
Every resource that supports this element. Those marked External carry the label “External resource. Not made or verified by Public Protocol”.
| Resource | Type | Origin | Maturity | Safety | Licence, as stated at its source |
|---|---|---|---|---|---|
| Ceph hardware recommendations | Guide | External | Established | S1 | Not stated on the page opened |
| Redfish Developer Hub | Standard | External | Established | S0 | Not stated on the page opened |
| K3s requirements | Datasheet | External | Established | S1 | Not stated on the page opened |
| K3s | Tool | External | Established | S1 | Not stated on the page opened |
| Prometheus | Tool | External | Established | S0 | Apache 2.0 License |
| Proxmox Virtual Environment | Tool | External | Established | S1 | GNU Affero General Public License, version 3 |
| Proxmox VE System Requirements | Guide | External | Established | S0 | Not stated on the page opened |
| XCP-ng | Tool | External | Established | S1 | Not stated on the page opened |
Depends on, and completes
It is one of 25 elements that complete Digital infrastructure.
Environments
- Islands, remote and harsh places generally: K3s describes itself as made for unattended, remote places (K3s); remote management by Redfish reduces visits (Redfish Developer Hub).
- Arctic cold, hot deserts, tropical humid forest, temperate forest, high altitude, flood plains: no source found specific to the software; the hardware limits are in the server room element.
Safety
State and evidence
Drafted
An editor has written the element's text from the resources listed under it, citing them. It is checked when it is reviewed.
Evidence from a real place or system is recorded here, with what did not work. What the states mean
History
- Drafted in the release of 16 November 2026.