Data Center Tiers: What Are They and Why Do They Matter?
Data center tiers indicate the level of uptime and security associated with an off-site data center. Below, we explore what each of the data center tiers (1, 2, 3, or 4) mean and which your business should use.
Blog Post
8 minute read
Sep 11, 2023
Surprisingly, most of the world’s data is new. It’s estimated that around 90% of the world’s data was generated in the last two years alone. All this data has to be stored somewhere, though. Data centers are buildings or dedicated spaces for housing computer systems. Data centers are ranked by their capability and categorized into data center tiers.
There are currently four data center tiers, ranked by performance and uptime. We explore each in depth below.
Modern IT stacks can be as intricate or as streamlined as you’d like. Explore how a provider can help you build the perfect solution for your organization in Impact’s eBook, Does Your Business Need a Managed IT Service Provider?
What Are Data Center Tiers?
Data center tiers are established rankings for the performance of servers that house data and information. The tier of the data center you use dictates the level of security and how much potential downtime you could experience over the course of a year.
This is crucial as downtime is extremely costly in terms of dollars and reputational clout. On average, downtime costs small businesses anywhere from $137-$427/minute while larger companies suffer a cost of $5,600-$9,000/minute.
Data centers are ranked from 1 to 4, with 1 offering the lowest performance and 4 the highest. This means that tier 1 data centers experience the most downtime while tier 4 data centers experience the least.
Cyberattacks are an increasingly relevant threat for SMBs—nearly half of all attacks target small and midsize organizations, a figure that has been rising in recent years.
SMBs have been falling victim more frequently to cyberattack. In fact, small businesses account for 43% of all data breaches. Because of this, ensuring business continuity is a major concern for decision makers and executives. After all, the cost of downtime is significant and oftentimes crippling.
Simply put, where you keep your data matters immensely, and with so many businesses migrating their company data to the cloud, it’s important to know the distinction between the data center tiers that offer to host your most valuable information.
Let's take a look at how the different data center tiers stack up against each other, what it means for SMBs, and the data center tier rating system in its entirety.
What Determines the Data Center Tier Standard?
The Uptime Institute's Tier Certification is the independent measure through which ratings are judged. It determines the criteria for each tier and lists several values that collectively make up what constitutes the standards of data center tiers. Key factors include:
Performance: Standards are performance-based, meaning any solution which meets the requirements for availability, redundancy, and fault tolerance is acceptable.
Technology neutral: Tiers don’t require specific technologies in order to be classified, mostly because new advanced technologies consistently disrupt the digital landscape.
Vendor-neutral: The brand of technology used in a data center isn’t considered as a factor in determining the tier, meaning centers are judged purely on their capabilities.
What Are the Data Center Tier Ratings?
Now that we’ve gone over the purpose of data center tier ratings, let’s take a look at each individual rating and see what you can expect from each tier in terms of uptime and redundancy.
Listed below are the data center redundancy tiers as outlined by the Uptime Institute in their data center tier classification system.
What Is a Tier 1 Data Center? (Basic Capacity)
A tier 1 data center is the lowest-rated. It’s above putting a stack of servers in a closet in your office, but as far as performance goes, tier 1 is at the bottom of what you can get from data center tiers.
This is because a data center adhering to tier 1 standards can only guarantee an uptime of 99.671% and lacks IT equipment that supports redundancy. While this sounds fairly impressive from a percentage standpoint, it actually equates to over a full day of downtime on an annual basis.
Redundancy is a process in which components critical to the functioning of the data center are duplicated and kept as backups and fail-safes should an issue occur. The primary function of redundancy is to improve the reliability of the data center.
Tier 1 data centers require no redundancy, meaning they don’t have to offer basic backup needs, like a simple power and cooling setup and uninterrupted power supply (UPS), and they fall short of the standards of the other tiers.
As such, tier 1 data centers are typically best suited for very small businesses or start-ups that aren’t operating 24/7 and are looking for an affordable option. However, they might need to have scheduled downtime during off-hours for maintenance.
All-in-all, tier 1 data centers can result in up to nearly 29 hours (28.8) of annual downtime, which is just under 1.25 days.
What Is a Tier 2 Data Center? (Redundant Capacity Components)
Tier 2 data centers offer all the capabilities of tier 1, but with added redundancy options. Like tier 1, tier 2 systems will have a single power input, but with additional fail safes for backup.
These fail safes include UPS modules, chillers, pumps, and energy generators. Tier 2 centers offer a marginally higher uptime of 99.741%, which translates to just under 23 hours (22.68) of downtime/year.
Tier 1 and tier 2 can be thought of as similar in most respects, with tier 2 being the more robust of the two. As a slightly better option in terms of uptime and a bit of cooling/power redundancies, tier 2 data centers can also serve some small-to-medium sized businesses that are still on a tight budget.
Both tier 1 and tier 2 data centers lack the performance capabilities found in tiers 3 and 4, but still offer a good balance of performance and affordability for small-to-medium sized businesses.
Tier 2 data centers are not totally redundant, but they are more reliable and secure than tier 1 alternatives.
What Is a Tier 3 Data Center? (Comprehensive Redundancy)
This is where data center tiers start to get a little more serious. SMBs that are more established, have a larger budget, or work with a lot of data generally prefer to use at least a tier 3-rated data center for the far superior redundancy protections offered.
There is a significant jump in uptime from tier 2, as tier 3 offers an annual uptime of 99.982%. This means that your network will experience no more than 95 minutes of downtime/year. That’s, at a minimum, 20 hours of downtime eliminated annually when moving from a tier 2 data center to a tier 3.
The improved uptime is owed to the requirements for more comprehensive redundancy capabilities.
Where tiers 1 and 2 only require one path for power and cooling, tier 3 requires an additional redundancy path for backup which executes in the event of failure.
N+1 redundancy means that there is an additional component for the purpose of supporting a single failure, or planned maintenance on a component. “N” refers to the necessary capacity in order to run the data center.
Tier 3 systems are not totally redundant, due to them often being reliant on or sharing components which are not fully independent to the data center—effectively meaning that it could be adversely influenced by external mishaps. This is the case for tiers 1, 2, and 3.
These types of data centers are favored by SMBs on the larger side or with more spending power whose IT operations need additional failsafes over the basic protections offered by the lower tiers.
What Is a Tier 4 Data Center? (Fault Tolerant)
Tier 4 data center security marks the highest standard for data centers—usually used by large businesses, government organizations, and global enterprises that require constant availability.
They have an uptime of 99.995%, meaning annual downtime of no more than 26.3 minutes/year.
They also feature 2N and 2N+1, fully redundant infrastructure, which is the second major difference between a tier 3 data center and the superior tier 4 data center.
2N redundancy means there is a completely mirrored system on standby, independent of the primary system. This means that should anything happen to a component in the main data center, there is an identical replica for every component ready to pick up the slack. This is by far the most robust form of security that can be employed.
All components are supported by two generators, two UPS systems, and two cooling systems. Each path is independent of the other, meaning that a single failure in one will not cause a domino effect with other components, as is the case with lower tiers.
Tier 4 data centers have a power outage protection of 96 hours, and this power must not be connected to any external source and must be independent. This is what’s referred to as “fault tolerance”—a capability that means that in the event of a system failure, IT operations aren’t affected in any way.
Unlike tier 3, tier 4 data centers are prepared for unplanned maintenance—businesses that use tier 4 systems will often be unaware that an outage has taken place at all.
Does a Tier 5 Data Center Exist?
While there is no formal tier 5 data center according to the Uptime Institute, there have recently been discussions around what a 5th tier would look like. Most notably, a tier 5 data center would meet all of the standards set forth by tier 4 while meeting additional sustainability standards.
The additional tier 5 standards would require the ability to run the network without water, an outside air pollutant detection system, permanently installed stored energy system monitors, securable server racks, run on local and renewable power, and potentially much more.
Formal tier 5 standards have not been published, however.
Reviewing the Data Center Tiers by Uptime
Tier 1: 99.671% uptime; maximum downtime of 28.8 hours per year
Tier 2: 99.741% uptime; maximum downtime of 22.7 hours per year
Tier 3: 99.982% uptime; maximum downtime of 1.6 hours per year
Tier 4: 99.995% uptime; maximum downtime of 0.4 hours per year
Data Center Tiers: Which Is Right for You?
The right data center tier for your organization largely depends on your current security posture, data usage, and budgetary constraints.
Businesses that host extensive data sets, particularly customer and client data, are prime candidates for the advanced protections offered by tier 3 and tier 4 data centers that the lower tiers simply don’t provide.
If efficiency is your concern, then tier 3 is the lowest data center tier that provides the services you need without having to worry about the more significant downtime that can occur with 1 and 2.
Tier 4 is recommended for businesses that value total protection, uninterrupted availability, and have more spending power.
No matter where you are in your business journey, or what data center tier is right for your organization, having a plan for your data in the digital era of today is absolutely necessary.
