In simple terms a data center implies a central repository for servers, storage, management and dissemination of information. Data center is an infrastructure facility to provide computing capabilities, storage and networking. The services offered by data centers are data storage, redundant data communications, and other security devices. Data centers offers the physical or virtual infrastructure services for companies to house computers, servers and networking systems and components for the company’s IT requirements.
The original idea of data center began as a private server room in a company connecting user computers to the server and running applications on the server for access by users. In earlier days the organization was responsible for maintenance of servers, storage, and networking components and this required many IT personnel to manage the facility. Even today large organizations maintain their in-house or internal data centers and are able to maintain service levels at a high cost.
As business and data grows constantly, in house data centers face the challenge of constant hardware and infrastructure upgrades. To overcome this challenge third party data center service providers offer IT resources for clients (companies and organizations) for use in their business operations. Many business enterprise organizations are hiring services from third party data centers, such as cloud computing. Hence, we can understand that a data center requires huge server farms, storage arrays and networks and redundant telecommunications network and components to cater to the various needs of different types of organizations. Data centers to ensure reliability in services without any disruptions from the perspective of the user.
Data center performance and benchmark standards were developed by Telecommunications Industry Association (TIA) in 2005. This is known as TIA-942 was first such standard to address data center infrastructure, particularly in the design and development process. TIA is accredited with ANSI and has over 600 members globally. TIA provides necessary enhancements for organizations involved in telecommunications, broadband, cellular and wireless services, IT, networks, satellite communications, unified communications and environment friendly technology. TIA-942 standards for data center performance and reliability is based on evaluations on general redundancy and availability of data center in its topology. The design aspects include site space and cabling infrastructure for tiered reliability along with environmental considerations.
In order to maximize efficiency and uptime in data centers, Uptime Institute Inc., a consortium of companies provided a tiered classification system for data centers. Uptime’s tiered classification is an industry standard for site infrastructure design and uptime of data center. There are four tiers (Tier 1 to Tier 4) explained by Uptime Institute related to the performance of the data center, investment and Return On Investment (ROI). Tier classifications are derived based on performance of actual site availability and by combining design topology and site location. The data center tiers and their characteristics are mentioned below to show the service availability for each tier.
S.No | Tier Requirement | Tier 1 | Tier 2 | Tier 3 | Tier 4 |
1 | Basic Redundancy | N | N+1 | N+1 | N+N |
2 | Active Paths | One | One | One | Dual |
3 | Fault Tolerant | No | No | No | Yes |
4 | Staffing | None | 1 Shift | 1 +Shifts | 24XForever |
5 | Single Point of Failure | Many | Many | Many | None |
7 | Representative failure | 6 Failure in 5 years | 1 Failure every year | 1 Failure every 2.5 years | 1 Failure in 5 years |
Tier 1: Basic site infrastructure (N). Availability: 99.671%
- Vulnerable to disruptions from planned and unplanned activity
- Single path for power and cooling
- Data center is shut down completely for performing preventive maintenance
- Annual downtime of 28.8 hours
Tier 2: Redundant Capacity Components Site Infrastructure (limited N+1). Availability: 99.741%
- Less vulnerable to disruptions from planned and unplanned activity
- Single path for power and cooling includes redundant components (N+1)
- Includes raised floor, UPS and generator
- Annual downtime of 22.0 hours
Tier 3: Concurrently Maintainable Site Infrastructure (N+1). Availability: 99.982%
- Enables planned activity (such as scheduled preventative maintenance) without disrupting computer hardware operation. Unplanned activities can still cause disruption.
- Multiple power and cooling paths (one active path), redundant components (N+1)
- Annual downtime of 1.6 hours
Tier 4: Fault Tolerant Site Infrastructure (2N+1): 99.995% availability
- Planned activity will not disrupt critical operations and can sustain at least one worst-case unplanned event with no critical load impact
- Multiple active power and cooling paths
- Annual downtime of 0.4 hours
Despite the advantages offered by each tier classification, operational sustainability plays a significant role in site availability which is ensured in tier 4 data centers. Tier 4 data centers are designed to provide continuous availability of services with reliability.