Data centre setup & management
A data centre built for your own systems — and run properly after go-live.
We design, build and manage server rooms and on-premises data centres for organisations that host their own software, databases and records — engineering power, cooling, network, compute, security and monitoring as one system, then operating it under a structured management agreement.
When to call us
Problems this solves
- Business systems run on servers in an office cupboard with no cooling, backup power or security.
- Data must stay on premises for regulatory, contractual or performance reasons.
- An existing server room is full, overheating or a single point of failure.
- Nobody is responsible for monitoring, patching and testing the environment once it is built.
Capabilities
What Data Centre Setup & Management covers
- 01
Assessment & capacity planning
Workload, power, cooling and growth assessment, with a site survey that sizes the room for today and the next several years.
- 02
Room design & preparation
Layout, rack elevations, raised floor or overhead distribution, access routes, environmental and structural requirements.
- 03
Power infrastructure
Utility and generator integration, automatic transfer switching, redundant UPS (N+1 or 2N), battery systems and metered rack PDUs.
- 04
Precision cooling & containment
Precision or in-row cooling sized to the heat load, with hot- and cold-aisle containment for efficiency.
- 05
Racks, cabling & network core
Rack build-out, structured copper and fibre, redundant core switching, firewalls and diverse internet links.
- 06
Compute, storage & virtualisation
Servers, shared storage and virtualisation clusters to host in-house applications, databases and file services.
- 07
Physical security & fire protection
Biometric access control, CCTV, early-warning smoke detection and clean-agent gas suppression that protects equipment.
- 08
Monitoring & DCIM
Temperature, humidity, leak and power sensors with alerting and dashboards — including AI-assisted anomaly detection.
- 09
Backup & disaster recovery
On-site and off-site backup, replication and recovery plans — off-site copies can run on MavenHost dedicated servers or VPS.
- 10
Ongoing data centre management
Monitoring, preventive maintenance, patching, capacity planning, change control and recovery testing under a management agreement.
Build options
The right facility for the systems you host
We size the facility to your workloads and criticality, not the other way round.
Server room
1–4 racks
A dedicated, cooled and secured room for an organisation’s core servers, storage and network.
- Single-site organisations
- Business systems, file and email servers
- Replacing an office “server cupboard”
On-premises data centre
Multiple rack rows
A purpose-built facility with redundant power and cooling for organisations hosting many critical systems.
- Hospitals, universities and agencies
- Large databases and records systems
- Data that must stay on premises
Edge & micro data centres
Self-contained racks
Enclosed racks with integrated cooling, power and monitoring for branches, plants and remote sites.
- Branch offices and depots
- Industrial and roadside sites
- Local processing for cameras and sensors
Hybrid
On-premises + hosted
Critical systems on your own infrastructure, with off-site backup, disaster recovery or public services hosted on MavenHost.
- Off-site backup and recovery
- Public websites and portals
- Burst capacity and testing
Engineering scope
Every layer engineered as one system
A data centre is only as reliable as its weakest layer. We design power, cooling, network, compute, security and monitoring together.
Room & site preparation
- Layout, rack elevations and clearances
- Raised floor or overhead cable and power distribution
- Floor loading, access routes and environmental sealing
Power
- Utility and generator integration with automatic transfer switching
- Redundant UPS — N+1 or 2N — and battery systems
- Metered and switched rack PDUs on dual feeds
Cooling
- Heat-load calculation per rack and per room
- Precision, in-row or close-coupled cooling
- Hot- and cold-aisle containment and blanking
Racks, cabling & network
- Structured copper and fibre with full labelling
- Redundant core switching and firewalls
- Diverse internet and inter-site links
Compute, storage & virtualisation
- Servers sized to your applications
- Shared storage (SAN/NAS) with snapshots
- Virtualisation clusters with failover between hosts
Physical security
- Biometric or card access with audit logs
- CCTV with AI video analytics
- Locked racks and visitor procedures
Fire detection & suppression
- Early-warning (aspirating) smoke detection
- Clean-agent gas suppression that protects equipment
- Integration with the building fire alarm
Monitoring & DCIM
- Temperature, humidity, leak and door sensors
- Power and UPS monitoring with alerting
- AI-assisted anomaly detection and capacity trends
Hosting in-house systems
A dependable home for your own software and data.
The facility is only half the job. We set up the platform your applications run on, so in-house systems are secure, recoverable and easy to change.
Production and staging separated
Separate environments so changes are tested before they reach the systems your staff rely on.
Databases built for reliability
Database servers with replication, scheduled backups and documented restore procedures.
Virtual machines and containers
Applications deployed on virtual machines or containers, so capacity can be reallocated without new hardware.
Secure access
VPN or zero-trust access for remote staff and vendors, with role-based permissions and access logs.
Backups that are tested
Local and off-site backups following the 3-2-1 principle, with regular test restores.
Software we build, hosted where it belongs
Custom applications from our software team deployed and supported on the infrastructure we manage.
Resilience levels
Choose resilience to match how critical your systems are
We use the ANSI/TIA-942 rated levels as a design reference, so everyone agrees what “resilient” means before work starts.
TIA-942 Rated-1
Basic
N — no redundant components
Small server rooms; planned downtime acceptable for maintenance.
TIA-942 Rated-2
Redundant components
N+1 UPS and cooling units
Most organisations; a single unit can fail without an outage.
TIA-942 Rated-3
Concurrently maintainable
Redundant power and cooling paths
Critical systems; any component can be maintained without shutdown.
TIA-942 Rated-4
Fault tolerant
2N — fully duplicated paths
Highest criticality; withstands a single failure anywhere in the facility.
| Level | Reference | Redundancy | Suits |
|---|---|---|---|
| Basic | TIA-942 Rated-1 | N — no redundant components | Small server rooms; planned downtime acceptable for maintenance. |
| Redundant components | TIA-942 Rated-2 | N+1 UPS and cooling units | Most organisations; a single unit can fail without an outage. |
| Concurrently maintainable | TIA-942 Rated-3 | Redundant power and cooling paths | Critical systems; any component can be maintained without shutdown. |
| Fault tolerant | TIA-942 Rated-4 | 2N — fully duplicated paths | Highest criticality; withstands a single failure anywhere in the facility. |
Designs are aligned with these levels as a reference; formal certification of a facility is a separate, independent process.
How we deliver
From survey to steady-state operations
- 01
Assess
Inventory of systems to host, criticality, power and cooling load, growth plans and a site survey.
- 02
Design
Resilience level, single-line power diagram, cooling calculation, rack elevations, network design and bill of quantities.
- 03
Build
Room works, power and cooling installation, racks and cabling, security and fire systems — coordinated with any other contractors.
- 04
Commission
Integrated testing including failover of power, cooling and network, followed by as-built documentation and handover.
- 05
Migrate
Planned migration of servers and applications with rollback steps and minimal downtime.
- 06
Operate
Monitoring, maintenance and continuous improvement under a data centre management agreement.
Data centre management
Operated, not just installed.
Most data centre failures come from missed maintenance, untested backups and undocumented changes. Our management agreements cover the routine work that keeps a facility reliable.
Need off-site backup or a recovery site? MavenHost dedicated servers (opens in a new tab) can host your recovery copies.
- Environmental, power and system monitoring with alerting
- Preventive maintenance of UPS, batteries, cooling and generators
- Firmware, hypervisor and operating-system patching
- Capacity and growth planning reviews
- Backup verification and disaster-recovery testing
- Change control, access management and up-to-date documentation
- Incident response under defined priority levels
- Periodic reports on availability, risks and recommendations
Do we need a data centre or is a server room enough?
It depends on how many systems you host, how critical they are and how quickly you are growing. Many organisations start with a well-engineered server room built to the same principles — redundant power, proper cooling, security and monitoring — and expand later.
Can you upgrade our existing server room?
Yes. We assess what is there, then improve power, cooling, cabling, security and monitoring in phases, planned around your working hours to keep downtime to a minimum.
Should we host on premises or use hosting?
Keep systems on premises when data must stay on site, latency to local users matters or existing hardware is still serviceable. Use hosting for public websites, off-site backup and disaster recovery. Most organisations benefit from a hybrid of both, which we can design with MavenHost.
How long does a data centre project take?
Timelines depend on scope, building works and equipment lead times. We provide a schedule with the design, after the site survey.
Industries
Where this is typically used
Resources
Guides and insights
- Blogdata centres
Migrating servers into a new data centre
Inventory, dependencies, move groups, rehearsals and rollback plans make a data centre move uneventful.
By Vicky Ogola 1 min
Read more - Blogdata centres
Fire suppression for server rooms explained
Early detection and clean-agent suppression protect equipment better than water — but design, sealing and safety matter.
By Vicky Ogola 1 min
Read more - Blogsecurity
CCTV storage: how long and how much
Retention period, resolution, frame rate, compression and recording mode determine storage needs. Calculate before buying.
By David Otieno 1 min
Read more
Planning a server room or data centre?
Tell us which systems you need to host, how critical they are and where. We will arrange a site survey and propose a design, resilience level and management plan.
Speak to us directly
Monday – Friday, 8:00 am – 5:00 pm