Data Center Tiers Explained
Data center tiers are established rankings for the performance of servers which house data.
The tier of the data center you utilize dictates the level of security and how much potential downtime you could experience over the course of a year.
Data centers are ranked from I to IV, with I being the worst-performing of the four and IV being the best-performing.
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.
Because of this, ensuring business continuity is a major concern for decision makers and executives. After all, the cost of downtime for SMBs is crippling, and often fatal—50% of all businesses don’t have the budget to recover from a breach of their data.
Put simply, where you keep your data matters, and with so many businesses migrating their company data to the cloud, it’s important to know the distinction between the servers that look after your information.
Today we’re going to be taking a look at how the different tiers stack up against each other and what it means for SMBs.
What Determines the 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 tier standards. 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 transformation 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.
Data Center Tiers
Tier I (Basic capacity)
A Tier I data center is the lowest-rated tier. It’s above putting a stack of servers in a closet in your office, but as far as performance goes, Tier I is at the bottom.
This is because a data center adhering to Tier I standards can only guarantee an uptime of 99.671% and lacks IT equipment that supports redundancy.
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 I 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.
They are typically best suited to very small businesses looking for an affordable option. There can be up to 1729 minutes of annual downtime.
Tier II (Redundant capacity components)
Tier II data centers offer all the capabilities of Tier I, but with added redundancy options.
Like Tier I, Tier II 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 II centers offer a marginally higher uptime of 99.741%, which translates to no more than 1361 minutes a year.
Tier I and Tier II can be thought of as similar in most respects, with Tier II being the more robust of the two.
Both types of data centers lack the performance capabilities found in Tiers III and IV, but still offer a good balance of low performance and high affordability.
Much like Tier I, Tier II data centers are commonly utilized by small businesses that want a cost-effective option for their data storage needs.
These data centers are not totally redundant, but they are more reliable and secure than Tier I alternatives.
Tier III (Comprehensive redundancy)
This is where data centers start to get a little more serious. Larger SMBs generally prefer to use at least a Tier III-rated system for the far superior redundancy protections offered.
There is a significant jump in uptime from Tier II, with Tier III offering annual uptime of 99.982%. This means that your network will experience a maximum of 95 minutes downtime per year.
The improved times owe to the requirements for more comprehensive redundancy capabilities.
Whereas Tiers I and II only require one path for power and cooling, Tier III requires an additional redundant path for backup which kicks in, in the event of failure.
N+1 redundancy means that there is an additional component for the purposes 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 III 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 I, II, and III.
These types of data centers are favored by SMBs whose IT operations need additional fail-safes over those basic protections offered by the lower tiers.
Tier IV (Fault tolerant)
Tier IV data centers are the most reliable centers—usually utilized by businesses that require constant availability.
They have an uptime of 99.995%, meaning annual downtime of no more than 26 minutes.
They also feature 2N and 2N+1, fully redundant infrastructure—the main difference between Tiers III and IV.
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 each 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 IV 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 which means that in the event of a system failure, IT operations aren’t affected in any way.
Unlike Tier III, Tier IV data centers are prepared for unplanned maintenance—businesses which use Tier IV systems will often be unaware that an outage has taken place at all.
Data Center Tiers by Uptime
- Tier I: 99.671% uptime; maximum downtime of 28.8 hours per year
- Tier II: 99.741% uptime; maximum downtime of 22.7 hours per year
- Tier III: 99.982% uptime; maximum downtime of 1.6 hours per year
- Tier IV: 99.995% uptime; maximum downtime of 0.4 hours per year
Which Data Center Is Right for You?
What data center is appropriate for your organization largely depends on your security, availability, and budget needs.
Businesses that host extensive data sets—particularly if the data pertains to customers—are prime candidates for the advanced protections offered by Tier III and Tier IV data centers.
These will often be larger or growing SMBs for whom the protections of the two lower tiers don’t provide adequate coverage.
If efficiency is your concern, then Tier III is the lowest tier that provides the services you need without having to worry about the more significant downtime that can occur with I and II.
Tier IV is recommended for businesses which value total protection, uninterrupted availability and for whom budget constraints aren’t as tight.
Interested in Learning More About Data Centers?
At Impact, we take pride in providing best-in-class solutions for our clients. Our business continuity plans utilize Tier IV data centers in nine locations around the world, allowing us to meet security, data sovereignty, and geo-redundancy requirements for all our partners.
Find out more about what Impact can do for your data center needs by getting in touch with one of our specialists.