Proxmox VE runs VMs well. When you need a genuinely multitenant platform, with quotas per tenant, managed Kubernetes and databases, several sites under one console and an API your automation can trust, edgeContinuum turns OpenStack into that platform.
Proxmox VE is a capable, open virtualization stack, and plenty of teams run it happily until the shape of the problem changes. The searches for a Proxmox alternative tend to start at the same three walls: no real tenancy, so isolating clients or teams means separate clusters and manual discipline; no managed services, so Kubernetes clusters and PostgreSQL are yours to install, upgrade and rescue; and one UI per cluster, so a second and third site multiply the operations instead of the capacity. edgeContinuum keeps the open foundation and adds the platform layer: OpenStack underneath, and a self-service cloud with organizations, quotas and managed services on top.
Up to a point, and the limit is worth knowing before you build on it. Proxmox VE has users, groups, pools and a fine-grained permission system, so you can carve a cluster into areas and delegate them. What it does not model is a tenant as a first-class object: there is no organization that owns projects, carries its own quota envelope and reports its own consumption. So multi-tenant Proxmox in practice means naming conventions, careful ACLs and a spreadsheet, and it holds until someone needs a hard limit or an auditable number.
edgeContinuum makes the tenant the unit. An organization owns projects; projects own resources; roles decide who does what; quotas are enforced by the platform rather than by discipline; and usage is tracked per tenant per managed resource, ready for your own accounting. That is the same model whether the tenants are internal teams or paying clients.
Each facility runs an edgeContinuum agent that pulls its desired state from the control plane and reconciles locally, so a site keeps provisioning, healing and restarting on its own even if the link to the centre drops, then resyncs when it returns. Hundreds of sites are managed from one console without one console per site.
Instead of a hypervisor console, users get a catalog: clusters, databases, virtual machines with networks and firewall rules, and Helm charts or your own internal applications published as one-click templates. What appears in that catalog is your decision, and every request lands inside the quota of the project that made it.
Helm charts and containers instantiated in one click.
Catalogs, routers and firewall rules, managed as a service.
Highly available, lifecycle handled, project scoped.
Your hardware and perimeter, EU-built, air-gap capable.
Book a call and we’ll look at your sites, tenants and services together, or request a free trial and provision a managed cluster on your own infrastructure.
