What Is a Data Center?

A data center is a physical facility that houses computing infrastructure, including servers, storage systems, and networking equipment, used to run applications, process transactions, and store and distribute data at scale. Data centers serve as the operational backbone of enterprise IT, cloud services, and internet infrastructure, providing the controlled environmental conditions and redundant power and connectivity that compute-intensive workloads require. The term applies to facilities ranging from a single server room within an office building to hyperscale campuses covering hundreds of thousands of square meters.

Data centers draw from electrical engineering, mechanical engineering, network architecture, and facilities management. Their design must balance computing density, power delivery, thermal management, physical security, and connectivity in a single integrated facility, with failures in any one domain potentially interrupting service for thousands or millions of users.

Core Infrastructure Components

The infrastructure of a data center consists of compute, storage, network, power, and cooling subsystems. Servers, either physical machines or virtualized hosts, form the compute layer; storage area networks (SANs), network-attached storage (NAS), and solid-state arrays form the storage layer. Switching and routing equipment interconnects servers within the data center and connects the facility to external networks. Power systems include utility feeds, transformers, uninterruptible power supplies (UPS), and backup diesel generators, layered to deliver continuous supply through utility outages. Cooling systems, including computer room air conditioning (CRAC) units, in-row coolers, and liquid cooling loops, remove the heat generated by densely packed equipment. Cisco's overview of data center infrastructure describes how these layers are designed to operate interdependently while providing individual points of redundancy.

Types of Data Centers

Data centers fall into several categories based on ownership, scale, and service model. Enterprise data centers are built and operated by a single organization for its own applications; they offer maximum control but require substantial capital investment. Colocation facilities rent space, power, and connectivity to multiple tenants, who bring their own servers and storage. Managed hosting providers own both the facility and the equipment, leasing computing capacity on a contract basis. Cloud data centers, operated by providers such as Amazon Web Services, Microsoft Azure, and Google Cloud, offer compute and storage resources on demand over a network, with the provider managing all underlying infrastructure. IBM's data center reference distinguishes these models by the degree of customer control over hardware, software, and configuration.

Reliability and Availability

Data center reliability is classified using the Uptime Institute's four-tier rating system, from Tier I (basic site infrastructure with expected 99.671% annual availability) to Tier IV (fault-tolerant with fully redundant components and 99.995% availability). The tier designation drives decisions on component redundancy, maintenance procedures, and capital expenditure. Higher-tier facilities use N+1 or 2N redundancy for power and cooling, meaning a full backup exists for every critical system. Availability is also affected by geographic risk, including proximity to flood plains, seismic zones, and electrical grid stability. AWS's explanation of data center design describes how hyperscale operators distribute workloads across multiple availability zones in separate physical data centers to tolerate a complete facility failure without service interruption.

Applications

Data centers have applications in a wide range of sectors, including:

• Cloud computing: hosting virtual machines, object storage, and managed services
• Financial services: processing real-time transactions and algorithmic trading systems
• Healthcare: storing and retrieving electronic health records and imaging data
• Media and streaming: delivering video content to global audiences at scale
• Scientific research: running high-performance computing clusters for simulation and data analysis