What we deliver
RS Computers designs, deploys and operates OpenStack private clouds for companies that want their own cloud: their own hardware, their own network and their data in a country they choose, without per-socket hypervisor licences. One team covers every layer, from the cables and switches to the APIs your developers use.
- Architecture and sizing: control plane, compute, storage and network design sized for your workloads, with growth and failure domains planned from day one.
- Hardware supply: bare metal compute and storage servers, top-of-rack and spine switches and edge routers, racked, cabled and tested.
- Networking: spine-leaf fabrics with BGP and VXLAN/EVPN, OVN for tenant networks, provider networks, IPv4 and IPv6, and upstream connectivity through our own IP transit.
- OpenStack deployment: containerised with Kolla-Ansible or OpenStack-Ansible, highly available control plane, KVM compute, and the services your users need.
- Storage: Ceph for Cinder volumes, Glance images, S3-compatible object storage and shared file systems.
- Bare metal provisioning: Ironic or MAAS to install and recycle physical servers automatically.
- Monitoring: Prometheus, Grafana and alerting for every node, service and link.
- Kubernetes on top: OKD, Magnum or Cluster API clusters running on your OpenStack.
- Operations: upgrades, capacity planning, incident response and support, or a documented handover to your team.
Where we build your cloud
| Location | What we provide there |
|---|---|
| Prishtina, Kosovo | Our own data centre: rack space, power, bare metal, routers, switches and spine-leaf networking |
| Amsterdam, Netherlands | EU PoP: bare metal, network equipment and IP transit, close to the major Amsterdam exchanges |
| Dublin, Ireland | EU PoP: bare metal and network equipment for clouds serving Ireland and the UK |
| Skopje, North Macedonia | PoP with IP transit and peering at IXP.mk: bare metal and network equipment for the Balkans |
| Albania | PoP: bare metal and network equipment for clouds serving Albania |
| Your own site | On premises in your server room or data centre: we supply and install the hardware, or build on servers you already own |
Clouds can also span several locations, for example production in Amsterdam and disaster recovery in Prishtina, joined over our own network.
How a project runs
- Discovery: your workloads, users, compliance needs, data location and existing hardware.
- Design and quote: node counts, storage layout, network diagram, location and a fixed project quote.
- Hardware and network: supply, racking and cabling of servers and switches, spine-leaf fabric and BGP up to the internet.
- Deployment: OpenStack and Ceph installed, tested for failover, and benchmarked.
- Migration: workloads moved from VMware, other hypervisors or public clouds, one wave at a time.
- Operate or hand over: we run it for you, or train your team and hand over the documentation.
Why OpenStack
OpenStack gives you the self-service APIs of a public cloud on hardware you control: projects and quotas for teams, networks, volumes, load balancers and images on demand, and automation with Terraform or Ansible. It has no licence fees, it runs on KVM, and it is built for scale. For a balanced view of when it fits and when something simpler is better, read our articles on OpenStack advantages and disadvantages, SDN in OpenStack and Ceph storage.