What is colocation? Almost every business now depends on digital infrastructure. Applications, customer records, communications platforms, operational systems and business data all need somewhere secure and reliable to run.
For some businesses, that infrastructure sits in a server room at their own premises. Others have moved almost everything into public cloud platforms. Many use a combination of physical hardware, private cloud and public cloud services.
Colocation provides another option.
Rather than keeping servers and other critical equipment inside an office, warehouse or operational site, a business can install its hardware in a purpose-built data centre. The colocation provider supplies the physical environment, power, cooling, connectivity and security needed to keep that equipment operating.
This gives the business continued control over its own hardware without requiring it to build, maintain and protect a data centre environment itself.
This guide explains what colocation is, how colocation hosting works, how it compares with cloud and on-premise infrastructure, and when it could make sense for your business.
What is colocation?
Colocation is an infrastructure model in which a business places its own servers, storage and networking equipment inside a third-party data centre.
The business usually continues to own or lease the IT hardware. The colocation provider supplies the secure physical space and supporting infrastructure needed to operate it.
Depending on the colocation service, this can include:
- Rack space
- Electrical power
- Backup power systems
- Cooling
- Physical security
- Internet and private network connectivity
- Environmental monitoring
- Fire detection and suppression
- Remote technical assistance
- Infrastructure monitoring and management
The simplest colocation definition is therefore:
Colocation allows a business to run its own IT equipment from a specialist data centre instead of keeping it at its own premises.
You may also see colocation described as colo, server colocation, data centre colocation or colocation hosting. These terms generally refer to the same underlying model.
Colocation is more than simply storage space for servers. A data centre is designed specifically to support technology that may need to operate continuously. Its power, cooling, connectivity and security systems are therefore considerably more advanced than those normally available in an office server room.
What is colocation hosting?
Colocation hosting refers to the service through which physical IT equipment is hosted within a provider's data centre. A business might colocate:
- One or two individual servers
- A private cloud platform
- Storage arrays
- Backup appliances
- Network switches and routers
- Security appliances
- High performance computing equipment
- Several complete racks of infrastructure
- A full production or disaster recovery environment
The amount of space required will depend on the size and power demands of the equipment.
Smaller deployments may only require a few rack units. A rack unit, commonly written as U, is a standard measurement used to describe the vertical space occupied by equipment inside a server rack.
Larger environments may require a quarter rack, half rack, full rack or several racks. Businesses with more complex requirements may also use a private cage or dedicated area within the data centre.
The colocation provider will normally assess the space, power, connectivity and resilience requirements before recommending an appropriate configuration.
How does colocation work?
A colocation arrangement usually begins with an assessment of the existing infrastructure.
The provider and customer identify which systems will move into the data centre, how much physical space they require, how much power they consume and how users, offices and cloud services will connect to them.
The process can be broken down into several main stages.
1. Infrastructure assessment
The first stage is understanding what is currently in place by identifying:
- The servers and other devices being used
- Their dimensions and rack requirements
- Current and peak power consumption
- Cooling requirements
- Application dependencies
- Network connections
- Security requirements
- Backup and recovery arrangements
- Future capacity requirements
This assessment helps avoid moving an existing problem into a new location. It can also identify hardware that should be upgraded, consolidated, virtualised or retired before migration.
2. Data centre and rack selection
The business then selects the data centre location and amount of space it requires.
Location can influence latency, connectivity options, accessibility, resilience and price. A business may also choose two geographically separated sites if it needs a secondary recovery environment or wants to reduce its dependence on a single location.
Rack allocation will depend on the volume of equipment, power density and projected growth. It is generally sensible to account for future capacity rather than selecting a configuration that is already close to its limit.
3. Connectivity design
Equipment inside a colocation data centre still needs to communicate with employees, customers, business sites and other technology platforms.
Connectivity may therefore include:
- Dedicated internet access
- Ethernet services
- Private wide area networks
- SD WAN
- MPLS
- Direct cloud connectivity
- Cross connects to carriers or other services
- Resilient connections using separate routes or providers
This is a critical part of colocation design. Moving infrastructure into a highly resilient facility will not deliver the intended result if access still depends on a single network connection.
4. Migration planning
Once the target environment has been designed, the equipment can be migrated.
This may involve physically moving existing hardware or installing new equipment in the data centre and transferring applications and data to it, considering factors like:
- Acceptable downtime
- Application dependencies
- Data synchronisation
- Testing
- Rollback arrangements
- User communications
- Security controls
- Supplier responsibilities
- Out of hours work
- Business continuity requirements
A staged migration can reduce risk by moving services in a controlled sequence rather than attempting to transfer the entire environment at once.
5. Ongoing operation
Once installed, the equipment continues to operate as the business's own infrastructure.
The colocation provider maintains the surrounding environment, including power, cooling, physical access and facility security. The business remains responsible for its servers, operating systems, applications and data unless it purchases additional managed services.
Some providers offer remote hands support, which allows authorised data centre technicians to carry out agreed physical tasks. These might include checking equipment, replacing a cable, restarting a device or installing approved components.
A fully managed arrangement can go further, with the provider monitoring, maintaining and supporting the infrastructure on the customer's behalf.
What does a colocation provider supply?
The exact scope of colocation services varies between providers, but the main components normally include the following.
Secure rack space
Servers and other equipment are installed in purpose-built racks within a controlled data centre environment.
Access is restricted to authorised people and may be protected through security personnel, electronic access controls, identity checks, monitored entry points and CCTV.
Resilient power
Data centres are built to reduce the risk of power disruption.
This can include diverse power feeds, uninterruptible power supplies and backup generators. The objective is to keep equipment operating if there is a failure or interruption in the primary electricity supply.
Power resilience may be described using terms such as N+1 or N+2.
N represents the amount of capacity required to support normal operations. The additional number represents spare capacity.
For example, an N+1 configuration has the required capacity plus one additional component or system. An N+2 configuration provides two additional units of capacity.
However, these terms should not be considered in isolation. Businesses should understand which systems are covered, how the complete design works and whether resilience extends across power, cooling and connectivity.
Specialist cooling
Servers generate heat and require a controlled environment.
Data centre cooling systems regulate temperature, airflow and humidity to protect equipment and maintain reliable performance. These systems are designed for continuous IT loads rather than the general comfort cooling used in offices.
Connectivity
Colocation facilities usually provide access to high-capacity connectivity from one or more network carriers. This gives businesses more choice when designing internet access, private networks, cloud connections and resilient routes.
The data centre provider may also be able to manage connectivity alongside the colocation environment, reducing the number of separate suppliers involved.
Environmental monitoring
Conditions within the facility are monitored continuously, such as:
- Temperature
- Humidity
- Power usage
- Cooling performance
- Water leaks
- Smoke
- Fire
- Physical access
Monitoring allows potential problems to be detected and addressed before they affect customer equipment.
Physical protection
Purpose built data centres include controls that would be difficult for most businesses to reproduce inside an office.
These may include controlled access zones, reinforced building security, CCTV, security staff, fire detection systems and specialist fire suppression.
Physical security does not replace cyber security. Businesses still need appropriate protection for their networks, systems, accounts, applications and data.
Colocation vs on premise infrastructure
On premise infrastructure is hosted inside a building controlled by the business, such as an office, warehouse, factory or private facility.
This gives the business direct physical access and complete control over the environment. However, it also makes the business responsible for providing and maintaining everything the equipment needs, including:
- Electrical capacity
- Backup power
- Cooling
- Fire protection
- Physical security
- Network connectivity
- Hardware monitoring
- Building maintenance
- Access procedures
- Disaster recovery arrangements
A small server room may initially appear cost effective because the premises already exist. The limitations often become more apparent as the infrastructure grows or becomes more critical to daily operations.
Office power and cooling systems may not have been designed to support a dense, continuous technology load. Equipment may share the building with employees, contractors and visitors. Maintenance work, local power failures, flooding, overheating or connectivity problems can all affect availability.
Colocation moves the equipment into an environment designed specifically for that purpose.
The business retains greater control over hardware than it would with many cloud services, while transferring responsibility for the supporting facility to a specialist provider.
Colocation vs cloud
Cloud computing allows businesses to access configurable technology resources as a service. These resources can often be provisioned and adjusted without buying or installing physical hardware.
With public cloud, the provider owns and operates the underlying infrastructure. The customer pays to use the computing, storage, networking or application resources it requires.
With colocation, the customer normally owns or leases the physical equipment. The provider supplies the data centre environment in which that equipment runs.
This creates several practical differences.
Ownership
In a colocation model, the business retains direct ownership or control of its hardware. In a public cloud model, the infrastructure is owned and operated by the cloud provider.
Cost model
Colocation costs are linked to rack space, power, connectivity and support, alongside the cost of buying and maintaining hardware.
Cloud services are usually consumption or subscription based. Costs may vary according to usage, storage, data transfer, service tier and configuration.
Scalability
Public cloud resources can often be increased or reduced rapidly. Colocation capacity can also grow, but adding physical computing resources may require new hardware, installation work and additional rack space or power.
Control
Colocation gives the business greater control over physical equipment, hardware configuration and infrastructure lifecycle.
Cloud gives the provider more control over the underlying platform, although the level of customer control depends on the service being used.
Management responsibility
With colocation, the customer remains responsible for its hardware and software unless managed services are included.
With cloud, the provider takes responsibility for more of the underlying infrastructure. The customer remains responsible for areas such as data, access, configuration and application security, depending on the cloud service model.
Neither option is automatically better.
Public cloud can be effective for services requiring rapid scalability, geographic reach or flexible consumption. Colocation can be more appropriate for existing hardware, consistent workloads, specialist systems, private infrastructure or environments requiring direct hardware control.
Many businesses therefore use both.
How colocation supports hybrid cloud
Hybrid cloud combines different infrastructure environments within a coordinated architecture that will encompass some, or all, of:
- Public cloud
- Private cloud
- Colocated hardware
- Software as a service
- Edge infrastructure
- Remaining on-premise systems
Colocation can provide a stable physical foundation for hybrid cloud.
A business might run core databases or established applications on its own colocated hardware while using public cloud for development, analytics, customer portals, backup or services with variable demand.
The environments can then be connected using private or managed network services.
This approach allows each workload to run in the environment that best suits its performance, security, control, integration and commercial requirements.
It also avoids treating infrastructure strategy as a choice between keeping everything physical and moving everything into public cloud.
What are the benefits of colocation?
The value of colocation depends on the starting environment and business requirements. However, several benefits commonly influence the decision.
Greater infrastructure resilience
A purpose-built data centre can provide more resilient power, cooling and connectivity than a standard office server room. This reduces dependence on local building infrastructure and creates a more appropriate environment for systems that need to remain available.
For particularly critical services, colocation can support architectures spread across two locations. Applications and data can be replicated so that a secondary environment is available if the primary site is disrupted.
Improved physical security
Moving equipment into a controlled data centre reduces exposure to unauthorised access, theft, accidental damage and general building activity. Access can be restricted, monitored and recorded according to formal procedures.
More predictable operating conditions
Servers perform best in stable environmental conditions. Specialist cooling, power management and monitoring provide a more consistent operating environment than improvised server rooms or converted office spaces.
Better use of internal resources
Maintaining a server room involves more than looking after the servers. IT and facilities teams may need to manage cooling, power, access, connectivity, fire protection and environmental monitoring. Colocation transfers much of this facility responsibility to the provider. In turn, internal teams can then focus more of their time on applications, users, data and business improvement.
Access to better connectivity
Data centres often provide access to multiple carriers and high-capacity network services.
This can improve connectivity options and make it easier to introduce diverse connections, private cloud links or resilient network routes.
Continued hardware control
Some businesses need or prefer to retain control of their physical equipment. This may be due to application requirements, specialist hardware, data handling policies, existing investments or predictable workload patterns. Colocation provides that control without requiring the equipment to remain at the business premises.
A route away from ageing server rooms
Upgrading an internal server room can require significant investment. A business may need new cooling, power distribution, backup systems, monitoring and physical security before it even considers replacing the IT hardware. Colocation can provide access to an established data centre environment without the business having to design and build one.
Support for growth and change
Rack space, power and connectivity can be expanded as the infrastructure changes.
Colocation can also provide a stepping stone towards private cloud, public cloud or a broader hybrid infrastructure strategy.
Common colocation use cases
Colocation is used across a wide range of infrastructure scenarios:
Moving out of an office server room
A business may relocate critical equipment because its current premises no longer provide sufficient power, cooling, security or resilience. This is common when moving office, reducing property space or reviewing facilities costs.
Hosting established business applications
Some applications continue to perform reliably on dedicated physical or virtualised infrastructure. Moving them into colocation can improve the operating environment without requiring an immediate and potentially disruptive redesign for public cloud.
Supporting private cloud
A business can build a private cloud platform using hardware installed in a colocation data centre. This can provide greater control and isolation while still benefiting from professional data centre facilities.
Creating a disaster recovery site
Colocation can provide a secondary location for replicated systems, backup infrastructure and recovery services. Geographic separation helps reduce the risk that one local incident affects both the primary and recovery environments.
Supporting hybrid infrastructure
Colocation can connect existing private systems with public cloud and Software as a Service (SaaS) platforms. This allows businesses to modernise at a measured pace and place workloads according to their individual needs.
Hosting specialist or high-performance equipment
Some workloads require specific processors, graphics hardware, storage systems or network appliances. Colocation allows businesses to operate this equipment in a controlled environment without relying entirely on standard cloud configurations.
Consolidating infrastructure from several sites
A business with servers distributed across offices or operational locations may consolidate them into one or more data centres. This can simplify management, improve consistency and reduce the number of local environments that need to be maintained.
When does colocation make sense?
Colocation may be suitable when one or more of the following apply:
- Your current server room is reaching its limits
- Critical systems depend on office power and cooling
- The business is relocating or reducing office space
- You want to retain existing hardware
- Some applications are not ready for public cloud
- You require direct control over physical infrastructure
- You need stronger physical security
- You need access to better connectivity
- Your infrastructure has consistent and predictable workloads
- You're creating a private or hybrid cloud environment
- You need a geographically separate disaster recovery site
- Building or upgrading your own data centre would be impractical
- Internal teams are spending too much time managing facilities
- You need a structured route towards infrastructure modernisation
Colocation may be less appropriate for short term workloads, applications that can easily use standard software as a service platforms or environments that need to scale up and down extremely quickly.
The decision should be based on workload requirements rather than a general preference for one infrastructure model.
Why businesses are moving away from server rooms
The role of the office has changed. Hybrid working, distributed operations, cloud applications and greater dependence on digital services mean that business infrastructure no longer needs to sit in the same building as its users.
At the same time, server rooms are becoming harder to justify.
Buildings were not designed as data centres
Many server rooms began with a small number of devices and expanded gradually.
Over time, more equipment was added, power demand increased and cooling became harder to manage. What was once sufficient for a few systems may no longer provide the resilience required by a business that now depends on technology throughout its operations.
Infrastructure requirements are increasing
Modern equipment can have higher power densities, more demanding cooling requirements and faster connectivity needs. Supporting this effectively may require significant upgrades to the building.
Downtime has a wider impact
When systems support customer service, payments, logistics, communications and remote working, infrastructure downtime can affect the entire business. The suitability of the environment therefore matters more than it did when servers supported a smaller number of internal processes.
Office space is valuable
Businesses may prefer to use office space for employees and customer activity rather than power, cooling and server racks. Colocation allows physical infrastructure to be moved without forcing an immediate move to public cloud.
Skills are better focused elsewhere
Internal IT teams can add more value by improving services, supporting users, strengthening security and planning future technology. Managing air conditioning, generator testing, power capacity and physical access may not be the best use of those skills.
What should you look for in a colocation provider?
Colocation services should be assessed against your actual operational requirements.
Important areas to consider include:
Location
Consider accessibility, network latency, local risk and geographic separation from your other sites. Regional data centres can also provide a commercially attractive alternative to locations in London and the South East.
Power and cooling resilience
Ask how power and cooling are designed, which components have additional capacity and what happens during maintenance or equipment failure. Look beyond individual labels and understand the architecture as a whole.
Connectivity options
Review available carriers, internet access, private networking and cloud connectivity.
Where resilience matters, ask whether connections can enter the facility through physically diverse routes.
Security
Assess access controls, surveillance, visitor procedures, monitoring and physical protection. Cyber security and managed security services should also be considered as part of the overall solution.
Support
Understand what support is included, how incidents are handled and whether remote hands or fully managed infrastructure services are available.
Scalability
Confirm whether the facility can accommodate additional rack space, power and connectivity as your requirements grow.
Service levels
Review service level commitments, responsibilities, maintenance arrangements and escalation procedures.
Resilience across multiple sites
For critical environments, consider whether the provider can support paired data centres, replication and failover. Using geographically separated facilities can form an important part of disaster recovery and business continuity planning.
Is colocation right for every workload?
No single infrastructure model is ideal for every workload. Some applications are best consumed as software as a service. Others benefit from public cloud flexibility. Established systems may continue to run effectively on dedicated hardware, while sensitive or performance critical workloads may be better suited to private infrastructure.
The strongest approach is often to assess workloads individually, assessing factors like:
- Performance
- Availability
- Security
- Compliance
- Data location
- Integration
- Scalability
- Cost
- Existing investments
- Application lifecycle
- Internal skills
- Recovery requirements
Colocation is therefore not necessarily an alternative to cloud. It can be one component of a broader infrastructure strategy.
Building a practical infrastructure strategy
The first question should not be whether your business should choose colocation, cloud or on-premise infrastructure.
The better question is what each workload needs.
A practical infrastructure strategy starts by understanding the current environment, business priorities and acceptable level of risk. It then matches each system with the most appropriate platform.
For some businesses, that will mean moving almost entirely into public cloud and software as a service. For others, it will involve colocating critical infrastructure, introducing private cloud and connecting it to selected public cloud services.
Some may retain a limited amount of equipment on site while moving their primary production and recovery systems into data centres. The objective is not to follow a particular technology trend. It is to create an infrastructure model that supports the performance, resilience, security and flexibility the business requires.
How Digital Space supports colocation
Digital Space provides enterprise colocation services from secure UK data centre facilities.
Our approach begins with understanding your existing infrastructure, workload dependencies, connectivity and resilience requirements.
We can then design an environment that combines colocation, private cloud, public cloud and managed connectivity where appropriate, considering everything across the entire proposition, from deployment to in-life support:
- Individual rack units
- Full rack deployments
- Private infrastructure
- Hybrid cloud architecture
- Managed connectivity
- Infrastructure monitoring
- Migration support
- Disaster recovery environments
- Ongoing managed services
Our Telford and Newport data centres can support paired resilience architectures, giving businesses the option to place production and recovery infrastructure across geographically separate facilities.
This approach can help reduce dependence on a single site while supporting replication, failover and disaster recovery planning.
Rather than treating colocation as an isolated hosting decision, we consider how infrastructure, networks, cloud services, security and operational support work together.
Compare your infrastructure options
Colocation can provide a more secure, resilient and scalable home for business infrastructure, while allowing you to retain control over the hardware and systems you already depend on.
However, it is only one of several available options.
Understanding the differences between colocation, public cloud, private cloud and on-premise infrastructure will help you make decisions based on your workloads, costs, risks and future plans.
Download our infrastructure comparison guide to see how each model works, where it delivers value and which approach could best support your business.
