TechKnowSurge
TechKnowSurge Catalog Videos Hot, Warm, and Cold Sites
VideoSecurityFree

Hot, Warm, and Cold Sites

Hot, warm, and cold sites represent three tiers of disaster recovery readiness, each balancing recovery speed and cost differently. Understanding the tradeoffs helps organizations choose the right failover strategy for their risk tolerance and budget.

⚑ Complete this video to capture a CTF flag worth 1 point.

About this video

Geographic redundancy is a foundational element of business continuity planning, but having a secondary site means little without a clear strategy for how ready that site needs to be. The three standard models — hot, warm, and cold sites — each represent a different point on the spectrum between recovery speed and operational cost, and the right choice depends on an organization's tolerance for downtime and data loss. A hot site is a fully operational mirror of the primary environment, with live servers, replicated databases, and current data running in parallel at all times. When the primary site goes down, failover to a hot site can happen almost immediately with little to no data loss, making it the strongest option for mission-critical systems. The tradeoff is cost — maintaining a complete duplicate environment means paying for every resource twice. A cold site offers the opposite approach. It may have minimal infrastructure in place, sometimes nothing more than a reserved physical space and access to stored backup data. Bringing a cold site online can take days, and because data replication is infrequent or absent, recovery may involve significant data loss. The appeal is cost efficiency, as organizations are not maintaining duplicate hardware and services continuously. A warm site falls between these two extremes, typically featuring some pre-installed infrastructure and regular data transfers, so it can be activated faster than a cold site but still requires meaningful setup time before it is fully production-ready. Evaluating these options requires weighing recovery time objectives and recovery point objectives against the budget available for redundancy infrastructure.

What you'll learn

What's covered

Hot, Warm & Cold Sites

Key terms

Disaster Recovery
DR
The process and procedures for recovering IT systems and data following a disruptive event.
Failover
The automatic switching to a redundant system or component when the primary one fails.
Recovery Time Objective
RTO
The maximum acceptable time to restore a system or service after a disruption.
Recovery Point Objective
RPO
The maximum acceptable amount of data loss measured in time, defining how far back data must be recoverable.
Hot Site
A fully operational duplicate facility with live systems and current data that can immediately assume workloads if the primary site fails, providing the fastest possible disaster recovery time.
Warm Site
A partially equipped recovery site with some infrastructure and recent data copies that requires additional setup before becoming operational, balancing cost and recovery speed.
Cold Site
A recovery site with little to no pre-installed equipment or live data that can take days to become operational but carries the lowest ongoing cost.
Redundancy
The duplication of critical components or systems to increase reliability and availability.

Topics

Disaster Recovery Business Continuity Hot Site Warm Site Cold Site Recovery Time Objective Recovery Point Objective

Transcript

Geographical separation and the choice of site

We may have geographical separation between our sites, but another choice that we need to make is whether that extra site is going to be a hot site, a warm site, or a cold site.

I'm looking at Amazon Web Services, or the AWS site, where it's showing geographical locations of their different data centers. Here we see in the United States we have one here in the US West, which is closest to me. So I had a bunch of servers that were running here, but I wanted some geographical separation, so I had another site that I set up with a duplicate of some of my equipment over here in the US East, the Northern Virginia region right here. Everything would be replicated or duplicated over there.

But the question is, how much do I want replicated or duplicated? And how fast do I want to turn that up? It depends on what model we have and how much expense we want to have. If I set duplicates of everything up in this other region, that's going to be twice as expensive, because I have a duplicate of everything. But maybe I just want to do a subset of what I have from a service perspective.

The example application

So what do these look like? It's a hot site, a warm site, and a cold site.

Here's the application that I have running. We have several servers that are running this, so maybe we're doing some sort of N+1: we have two servers, but we are adding another server for redundancy. Another thing, we have this database server on the back end, and we have some sort of redundancy with it. This gives us a lot of redundancy, so we could have a server fail, or we could have the database fail, or we could have something fail, and still be up and running.

But what if this whole site comes offline? In this region there could be a volcano eruption, or there could be an earthquake, and so what we want is to fail over to a site that's located somewhere else.

Hot site

Now the question is, do I set up a whole duplicate of everything, with live data and live servers and my database, all of that up and running, so that way as soon as something happens with this we fail over to this hot site, and it could be virtually no or very little downtime to do this failover? So this is a really fast failover, but obviously we're paying twice as much because we need duplicates of everything.

Cold site

Now let's look at the other extreme. We have a cold site. In this cold site, maybe we don't have all of the equipment. Maybe we just have certain pieces, certain aspects of the equipment, or maybe we just have the site planned out with the ability to get the equipment in fairly quickly.

So a cold site just means that it's really not anywhere close to being production worthy — we would have to spin things up, and it could take days to spin everything up, to get the equipment in place, get the data all restored. Maybe over on this site we're just shipping the data over here, so if we need to do a restore, then it's available right there to do a full restore. That's what a cold site is.

Warm site

Then we have a warm site. A warm site is something in between. Maybe we're shipping over copies of our data over here so it's ready to be implemented, and maybe we have a few servers that are purchased, maybe the internet connectivity that goes over there. What we need to do is just then spin up the databases, install the data or do a restore of the data, and now this warm site is up and running.

Comparing cost, data loss and time

The hot site is obviously going to be the most expensive. The warm site is going to be less than that, and the cold site is going to be the least expensive.

But with the cold site, we could be losing quite a bit of data, because we might not be shipping our records over there as often. Versus the hot site, it's instantaneous, so everything is already live over there, and flipping it over means that we're going to have very little to no data loss with it.

Another thing is how much time it takes to flip things over. It takes very little for this hot site to flip over to it. It takes quite a bit of effort and time to get over to the cold site. And the warm site is somewhere in between.

About TechKnowSurge

TechKnowSurge builds IT and cybersecurity professionals through hands-on, concept-first training built around real understanding — not memorization. Free interactive tools, structured programs, and 25+ years of real-world experience, all in one place.

Explore free tools and programs →