Hot, warm or cold? Choosing a DR site

The loss of critical data can cripple operations and result in significant financial and reputational damage. Whether it be a hardware failure, natural disaster or cyber-attack – choosing the right Disaster Recovery (DR) site is crucial to minimise disruption.

DR sites are specialised locations or environments – either physical or cloud-based – to help businesses recover swiftly in the event of a disaster. Pre-cloud, a DR site would need not only infrastructure, but also dedicated, on-site teams for recovery.

For a business based in two or more locations, with one data centre in London and another in Manchester for example, the smaller data centre may have historically doubled as a DR site. In recent years however, cloud-based recovery solutions have become a much more practical solution.

In this blog post we will explore the different types of recovery site (including how they have changed over the years), their advantages and how to choose the best site for your business.

Types of Disaster Recovery site

Your choice of site is dependent on your unique needs for data security, recovery speed and cost. These needs include the criticality of your operations, recovery time objectives (RTO) and budget constraints.

Recovery sites are typically defined as cold, warm or hot – with increasing temperature correlating with faster recovery speeds. While the proliferation of cloud-based DR has blurred the lines between each somewhat, it is important to understand their differences.

Cold sites

Definitions vary, though all agree that cold sites are the most basic form of recovery site.  Some describe them as an empty space with minimal infrastructure in place. Others describe them as a site where infrastructure exists, but is not set up or configured in any way.  This could be a dev/test environment that can be wiped and repurposed for DR at a time of need.

Whichever definition you use, in the event of a disaster you are responsible for supplying and configuring hardware, installing software and restoring backups to restart operations. While cold sites are a cost-effective solution, they require significant time to become operational.

Key features:

• Minimal infrastructure (no pre-installed hardware or software)
• Requires significant setup
• High recovery times
• Lowest cost

Warm sites

Warm sites offer a middle ground between cold and hot sites. They come with some pre-installed infrastructure, such as networking equipment and servers, but data is not replicated or continuously updated.

In practice, the infrastructure is there for you to restore your backups to when you need it. Warm sites offer a shorter recovery time than cold ones but are more expensive due to the additional equipment and maintenance. 

Servers need to have sufficient capacity to support the environment you need to recover.  There is also an overhead in keeping servers and switches updated and patched to the same level as the production environment. 

Key features:

• Partial infrastructure in place
• Must be scaled to support a growing production environment
• Requires data and software installation
• Faster recovery time than cold sites
• Moderate cost

Hot sites

Hot sites are fully equipped data recovery sites with all necessary hardware, software and live data replication in place.

In the event of a disaster, you can switch operations to a hot site almost immediately with minimal downtime. This makes hot sites ideal for businesses that cannot afford prolonged outages. However, the comprehensive setup comes with a high cost.

Key features:

• Fully operational with replicated data
• Minimal to zero downtime
• Best for mission-critical operations
• Highest cost

How does this relate to the cloud?

It has always been a challenge to determine the best recovery site for your needs. Either you risk not having enough protection in the event of a disaster, or you overcompensate and face a drastic increase in maintenance cost and spend.

Fortunately, the advent of cloud computing has enabled on-demand infrastructure and Disaster Recovery as a Service (DRaaS) to quickly become the preferred choice for many organisations. Low-RTO DR is now much more accessible in the cloud.

As an example, DRaaS delivers the functionality of a hot site but at much lower cost by enabling you to provision infrastructure only when you need it. Likewise, some managed backup solutions enable restoration from backups with RTOs closer to cold or warm sites.

Some key benefits include:

Scalability and flexibility: Cloud environments can quickly scale resources to handle increased traffic during a disaster, providing flexibility when faced with fluctuating demands.

Faster recovery times: Cloud-based DR uses automated processes for quicker recovery, unlike traditional sites that require manual reconfiguration.

Enhanced security and compliance: Cloud DR offers robust security features and compliance support, often surpassing the capabilities of physical sites.

Cost efficiency: Cloud DR operates on a pay-as-you-go model, eliminating the need for costly infrastructure and reducing maintenance costs.

Making your decision

In an era where data loss can have catastrophic consequences, investing in a robust data recovery strategy is not optional – it’s essential. Whatever solution you may be considering, the right choice depends on your specific business needs and objectives.

If you need the fastest recovery times at a reasonable cost, then DRaaS is the way to go. But sometimes you need additional contingency.  With ransomware attacks increasingly able to spread to secondary sites, outlast replication and infect DR systems – sometimes full recovery from backups is your only option.

Last year we launched Jump-Start, a cloud-based solution that delivers recovery times and functionality somewhere between cold and warm DR sites. Despite having no infrastructure in place, you get instant provisioning – meaning you can start recovery as soon as you need it.

While recovery time is longer than conventional DR, the process is accelerated by bringing backups and the recovery environment together. Deploying in the cloud via IaC also ensures that the recovery environment is isolated, secure and unaffected by issues to production.

In September 2024, Jump-Start won Cloud DR and Continuity Product of the Year at Computing's Cloud Excellence Awards.