When it comes to data centers, there are a few different tiers or levels of classification. The first tier is the most basic and least expensive, while the fourth and fifth tiers are the most advanced and costly. This article will explore what each tier entails and what benefits businesses can expect at each level.
What is a Data Center?
Data centers offer many advantages to businesses, such as cost savings and increased productivity. A data center is simply a physical facility where computing equipment is stored for use by companies with critical applications or data that need to be accessed at any given time. Regardless, there are some key differences between traditional buildings used in this manner versus those built specifically for hosting servers.
The Growing Role of Data
The data center is a vast, ever-growing network of computers and networking equipment for processing information. They were first created in the early days of computing when technology became cheaper with time; they have evolved dramatically as it becomes more advanced now that we’re able to store so much on one small device!
The earliest versions consisted of only one big computer. Still, modern ones can contain thousands or even millions worth – all connected through different communication networks such as cell phones, making them easier to access anywhere at any time (even without an internet connection).
This shift is largely due to the exponential growth of data volumes. For example, IDC’s “Data Age” document predicts a tenfold increase in data volumes between 2018 and 2025. According to the prediction, there will be 175 zettabytes of data by 2025, which, if you try to download it at the current average Internet speed, would take 1.8 billion years.
That mass of data is now connected in multiple data centers, at the edge, in public and private clouds. But they still need to be stored somewhere. So the modern data center remains as important as ever. However, it must evolve to interoperate between major local facilities, various cloud centers and cloud providers.
As a result, data center infrastructure has grown from on-premises physical servers to virtual infrastructures that support applications and workloads in multi-cloud environments.
Dedicated firewalls are now being used to defend data centers, given the criticality of the massive amounts of consumer data that enterprises possess and the applications and server resources they use.
Components of a Data Center
The core components of any data center often include cybersecurity systems, firewalls, routers, servers, storage systems and switches.
The main common element of all data centers is servers. A data center server is a high-performance computer that has much more memory. It is much faster and has a more powerful processor. A server or group of servers may be designed to perform a single task, multiple applications, or a specific client.
Storage devices such as hard drives, solid-state drives, and robotic tape drives play an important role in any data center. Another vital component is the networking and communications equipment needed to maintain high bandwidth between servers. This includes routers, switches, network interface controllers and the endless miles of cables that help information flow through the data center.
Like other large IT systems, data centers rely on software to function. This includes operating systems and applications that run on servers, as well as clustering software such as MapReduce or Hadoop and virtualization tools to minimize the number of physical machines.
Facility Systems
Data center facilities require security systems to prevent unauthorized access to physical facilities. They are usually protected by alarms, biometric scanners, gates, security doors and security personnel. The facility will have emergency equipment such as fire alarms, sprinklers, and security measures to protect the data center from noise from servers, fans, and more.
In addition to security features, data centers rely on energy-efficient systems and equipment to keep the facility running smoothly. This includes energy equipment such as battery backups, generators, and uninterruptible power supplies. Cooling systems and air quality and temperature control equipment such as fans, chillers, blowers, sensors, plumbing and reservoirs keep servers running efficiently.
How Data Centers Operate
Data centers centralize organizations’ IT operations and equipment, providing a secure place to store, share and manage vast amounts of data. They allow businesses to view, backup and store data and protect it from man-made or natural disasters. Data centers receive, store and send data to support business-critical applications and enable data-intensive services such as:
- Artificial intelligence, big data and machine learning
- Data storage, backup, recovery and management
- E-commerce and transactions
- File sharing and email
- Real-time collaboration and productivity tools
Data center components require extensive infrastructure to support hardware and software. This includes power and cooling equipment, connections to external networks and security devices such as firewalls and intrusion protection. Application performance is guaranteed by delivery mechanisms that provide availability and fault tolerance through automatic failover and load balancing.
Types of Data Centers
There are different types of data centers that businesses can build or store their data. These include:
Data centers
One type of data center is a colocation facility, which leases space owned by the data center provider. The data center infrastructure is at the building owner’s disposal and includes bandwidth, cooling systems, networking, power and security equipment and services. The companies or service providers leasing the space are responsible for installing and managing components such as firewalls, data center servers, and storage systems.
Enterprise
One of the enterprise’s largest types of data centers is a facility built by an enterprise for its use and owned and operated by the enterprise. The facility may be located on the organization’s premises, but it is usually placed off-site in a location that provides reliable connectivity, power and security. Although constructing and equipping an enterprise data center necessitates a significant investment of money, it enables businesses to tailor it to their precise needs.
Hyperscale data centers
A hyperscale data center is designed to provide the hyper-scale computing needed for cloud computing and big data storage. Hyper-scale centers enable large cloud players to provide their customers with robust, scalable applications and storage services. Typically, these centers are at least 10,000 square feet, with more than 500 cabinets and 5,000 servers running on a superfast network. The key difference between hyper-scale and enterprise data centers is the high-speed fibre network used in hyper-scale centers.
Classification of data center tiers: (tier 1, 2, 3, 4)
Data center tiers are critical when choosing a facility to house your data. A tier rating shows what a data center has to offer regarding reliability and performance, so failing to identify the right tier can lead to downtime or unnecessary costs.
What Are Data Center Tiers?
The data center tiers are a standard ranking system for data center infrastructure that indicates its dependability. Under this classification, facilities are ranked from 1 to 4, where one is the worst and 4 is the best level.
The data center receives this international rating from Uptime Institute, an independent organization that determines the facility’s level based primarily on:
Uptime Assurance.
Fault tolerance (ability to cope with planned and unplanned failures).
Cost of Service.
This unbiased tiering system objectively explains how a particular data center is performing. However, the rating is optional, so not all data centers have an assigned tier. However, most large facilities prefer to ask the Uptime Institute for a rating because the official rating helps:
Build confidence in the services.
Promote the facility’s capabilities.
Build trust and attract potential customers.
Plan for future facility upgrades and expansions.
To receive an official rating, data center staff must submit plans and drawings of the facility to the Uptime Institute (Tier Certification of Design Documents). Uptime Institute representatives visit the center on-site to inspect the work and assign a rating.
Determining the data center tier your business needs is the first step in finding the right partner. Our article on choosing a data center will help you understand the rest of the process.
What Factors Are Considered for Data Center Tier Classification
The Uptime Institute keeps the exact method of assigning levels secret, but most key indicators are in the public domain. The Uptime Institute considers the following factors when assigning a rating:
Service availability and uptime assurance.
Level of redundancy (the process of duplicating critical components and keeping them as backups and protection in case of planned or unplanned failures).
Condition of cooling and power infrastructure.
Qualifications of personnel and maintenance protocols (in particular, the possibility of simultaneous maintenance).
Cost of maintenance.
Operational sustainability and ability of the center to meet long-term business objectives.
The time it takes for the center to install a new customer.
The security level of the data center.
The neutrality of the service provider.
The tier system does not require any specific technology or design choices. The freedom to choose solutions provides flexibility in meeting tier goals, so each data center can choose the best way to meet standards and achieve the desired rating.
Learn about the benefits of a carrier-neutral data center and how you can benefit from being vendor-neutral.
Data Center Tier Ratings Explained
The Uptime Institute has certified four tiers of data centers:
Tier 1: A data center with a single power and cooling path and no redundant components. The expected uptime for this tier is 99.671% per year.
Tier 2: A data center with a single power and cooling path and some redundant components. The expected uptime at this level is 99.741% per year.
Tier 3:A data center with multiple power and cooling channels and redundant systems allows personnel to work on the installation without unplugging it. The expected uptime at this level is 99.982% per year.
Tier 4: Fully fault-tolerant data center with the redundancy of every component. The expected uptime for this tier is 99.995% per year.
The four tiers of data centers are progressive. Data centers can move up and down the rankings, and each tier includes the requirements of lower tiers.
While reliability increases as the tier increases, tier four is not always a better option than a lower-rated data center. Each tier meets different business needs, so tiers 3 and 4 (the most expensive options) are often an excessive investment.
Data Center Tiers Compared
Screen Shot
Tier 1 Data Center
Tier 1 infrastructure provides the power and cooling capacity to support a full IT workload. These facilities have a single path for power and cooling, and there is no redundancy for any critical systems. Personnel must completely shut down operations for regular maintenance or emergency repairs.
The requirements for a Tier 1 facility are as follows:
An uninterruptible power supply (UPS) protects against power surges and outages.
Dedicated space for IT systems.
Engine-generator.
Dedicated cooling equipment running after hours.
Tier 1 data centers also require systems, protocols, and equipment that keep the data center up and running after hours (nights and weekends).
Due to the lack of backup, Tier 1 data center customers can expect up to 28.8 hours of downtime annually.
Tier 2 data center
Tier 2 infrastructure has all the characteristics of Tier 1 data centers but additional backup capabilities. These data centers provide better fault protection with:
Additional generators.
Energy storage units.
Coolers.
Raised floors.
UPS modules.
Pumps.
Heat dissipation equipment.
Fuel tanks and cells.
Additional cooling units.
Like Tier 1 centers, Tier 2 centers rely on a single distribution channel for power and cooling, so these facilities are still vulnerable to unexpected failures. Uptime is better than lower-rated data centers, so Tier 2 customers have up to 22 hours of downtime per year.
Tier 3 Data Center
A Tier 3 data center is a concurrently serviced facility with multiple power and cooling distribution paths. Unlike Tier 1 and Tier 2, a Tier 3 facility does not require a complete shutdown during maintenance or equipment replacement.
A Tier 3 facility requires all of the components present in a Tier 2 data center, but these facilities should also have N+1 availability:
- “N” denotes the necessary capacity to support the full IT load.
- “+1” means an additional component for backup.
N+1 redundancy ensures that the additional component will be operational if the primary component fails or personnel remove the part for routine maintenance.
Tier 3 data centers require a backup solution that can maintain operations in the event of a local or region-wide power outage. The facility must ensure that the equipment can operate for at least 72 hours after the outage.
Tier 3 facilities have a significant jump in availability compared to lower ratings. Customers using Tier 3 data centers experience up to 1.6 hours of downtime annually.
Tier 4 Data Center
The Tiers 4 data centers add resilience features to the list of Tier 3 standards. These data centers are made up of a variety of physically separated systems that function as backups and distribution paths. In addition to all of the Tier 3 conditions, a Tier 4 facility must provide:
All components support two generators, two UPS systems, and two cooling systems.
Each distribution path is independent so that failure of one of them will not cause a domino effect on the other components.
Operation continues for a minimum of 96 hours after a local or regional power outage.
The power supply is not connected to any external source.
There is no connection to any external power source. For a Tier 4 data center, the gap between duplicate components is critical. Physical separation prevents one event from taking down both systems.
Tier 4 data centers have redundancy of 2N or 2N+1:
A 2N (or N+N) redundancy means that the center has a fully mirrored, independent standby system. If something happens to the main component, an identical backup starts running to ensure uninterrupted operation.
If the secondary system fails, the 2N+1 model provides twice the operating capacity (2N) and an additional backup component (+1).
A Tier 4 facility can provide customers with no more than 26.3 minutes of downtime annually. Tier 4 service level agreements (SLAs) do not guarantee 100% uptime because there is a small chance that a component could run into trouble while maintaining its backup counterparts.
Looking for a more flexible hosting option? A virtual private data center provides the pros of a local data center without the high upfront scalability costs. Our introductory article on virtual data centers (VDCs) discusses software-defined systems.
Which Data Center Tier Is Right for You?
While higher tiers offer more reliable services, tier 3 or 4 is not always the right choice. Decision-makers should choose the data center tier that best meets their business needs.
Certain types of companies tend to gravitate toward certain tiers. Below is a breakdown of the common customers of each tier:
Tier 1:
These data centers are best suited for small businesses and start-ups looking for the most affordable hosting option. Smaller companies without complex IT requirements or 24/7 operations can incur frequent downtime.
Tier 2:
These centers are the best option for small and medium-sized businesses that need a cost-effective and more reliable option than a Tier 1 partner. Small and medium-sized companies typically use Tier 2 facilities to host non-critical data backups or databases.
Tier 3:
These data centers are ideal for large companies with IT operations that need additional failover protection. Companies that host vast amounts of data (especially customer data) are prime candidates for this tier.
Tier 4:
These data centers are suitable for businesses without budgetary constraints that require uninterrupted availability. Governmental organizations and large enterprises with mission-critical servers and high customer or business requirements are typical users of Tier 4 centers.
Typically, the cost and availability of a tier are crucial factors to consider. It’s a waste of money to pay for a Tier 3 data center when a less expensive one would do the job. Similarly, being located in a Tier 2 data center when you require higher uptime can impact your revenue, productivity, customer satisfaction, and reputation.
Choose the Right Tier and Avoid Downtime Issues (or Unnecessary Overpaying)
Understanding the differences between storage facilities that house your data is critical. Use this article to educate yourself about the many available storage alternatives and make an informed decision to prevent your company from overpaying and experiencing downtime difficulties.