Defining the Data Center
Glossary of Top Industry Terms
Data Center Glossary
Structured and unstructured data sets that are too large and complex for traditional processing methods to deal with. Challenges include the capture, storage, analysis, sharing and protection of such data. Cabinet A metal framed chassis that holds, secures and organizes a vertical stack of network and server hardware, including routers, switches, access points, storage devices and modems. Also known as a rack.
A carrier hotel is a data center where technology infrastructure connects to a range of telecoms and network service providers. Businesses rent floor space for their servers, storage devices and other IT hardware, while the carrier hotel provides the power, bandwidth, cooling and security. Also known as a colocation center.
Carrier-Neutral Data Center
A carrier-neutral data center facilitates interconnection between numerous telecoms carriers and colocation providers, thus enabling customers to switch providers without physically moving to a new site. Also known as a network-neutral data center.
Clean and Renewable Energy
Clean or renewable energy refers to power generated from sustainable and environmentally friendly sources such as solar, wind, water and geothermal, with minimum pollution or carbon footprint.
Cloud computing is the delivery of software, storage, and other computing services via the Internet (the cloud), rather than being deployed on local hardware. Cloud services are typically charged on a monthly usage basis.
In a cloud-hosted solution, customers rent virtual server space instead of physical servers. The virtual partitions draw resources from an array of underlying physical servers installed in a data center. The physical server itself may be shared with other applications or customers. This is generally considered less secure than a colocation or a VPS.
A private, direct connection to the cloud from within a data center. The connections are usually to major cloud providers such as Amazon Web Services (AWS), Microsoft Azure, Google Cloud, IBM Softlayer, VMWare, Rackspace, Alibaba Cloud and Oracle Cloud. Find Cloud On-Ramps on Cloudscene
A cloud server is a remote virtual server hosted by a cloud computing service. Cloud servers offer similar functionality to dedicated physical servers, but can be more reliable, scalable, flexible and cost-effective.
Cloud Service Provider
A cloud service provider is an organization that offers cloud-computing services such as IaaS, PaaS or SaaS. These services are typically offered to customers under an on-demand, pay-as-you-go model. Find Service Providers on Cloudscene
A colocation center (colo) is a data center where technology infrastructure connects to a range of telecoms and network service providers. Businesses rent floor space for their servers, storage devices and other IT hardware. Also known as a carrier hotel. Find Data Centers on Cloudscene
A cross-connect is a hardware connection between separate racks/cabinets/infrastructure provided by a data center. For example, a link between a network and an Internet service provider.
Dark fiber is unused optical fiber. Companies that lay cables can gain economies of scale by installing more than is immediately needed, with the excess fiber remaining dark (unlit) until purchased by future customers.
A data center is a building that hosts servers, storage devices, network equipment and other IT infrastructure. These facilities typically include backup power supplies, redundant communications links, cooling systems, fire suppression and security protection. Find Data Centers on Cloudscene
Data Center Operator
A data center operator is a company that runs and manages a facility enabling interconnection between businesses and their customers, partners, networks and IT equipment. Find Data Center Operators on Cloudscene
Data Center Security
Data center security refers to practices that protect data centers from a range of attacks and threats, both physical and digital. Security measures may include biometric authentication, mantraps, armed personnel, firewalls and anti-malware systems. Read about the world’s most secure data centers
A dedicated server is a form of Internet hosting where the customer leases an entire server not shared with other customers. This gives the organization complete control over the server’s operating system and hardware. Also known as a dedicated hosting service.
Disaster Recovery (DR)
Disaster Recovery is a strategy that mitigates the impact of negative events (such as fire, flood or power failure) by enabling an organization to quickly resume operations following the event. Examples include switching mission-critical functions to a backup location.
Express Routing uses a private connection between a data center and its customer’s infrastructure, avoiding the public Internet and thereby offering higher reliability and lower latency than regular connections.
With geothermal cooling, a closed-loop coolant-filled piping system runs under the ground around the data center, using the steady underground temperatures to help cool the facility.
Green Data Center
A green data center is one where the power, cooling, lighting, construction and maintenance are designed for maximum energy efficiency, minimum carbon footprint and reduced environmental impact. Read about modern data centers “going green”
Heating, Ventilation and Air Conditioning (HVAC)
HVAC refers to the industry standard technology that provides heating, cooling and air quality services to buildings and vehicles.
Hybrid cloud combines and integrates on-premises, private cloud and public cloud services, giving businesses the flexibility to switch workloads to the most efficient platform as needed.
Infrastructure as a Service (IaaS)
IaaS is a cloud-computing model that delivers virtualized computing resources over the Internet, with the hardware (servers, storage, network devices etc.) being provided and managed by an external service provider.
International Business Exchange (IBX)
IBX is a registered trademark of Equinix, and refers to Equinix’s network of colocation data centers designed to protect customers’ mission-critical data, with high reliability, redundancy, power density and security. Find Equinix data centers on Cloudscene
Interconnection Data center
interconnection refers to the networking of multiple separate data centers and/or Internet service providers to achieve business or technology objectives. Interconnection allows facilities to share workloads for the most efficient use of resources.
Intra-campus cross-connect uses fiber optic cables to interconnect clusters of nearby data centers, enabling high-speed data transfer between sites and allowing customers to leverage the combined power of the campus.
Internet Exchange (IX)
An Internet exchange is a physical facility through which Internet service providers transfer traffic between their networks using a direct connection, rather than through a third party network. The advantages are reduced cost, lower latency and more efficient bandwidth utilization.
Internet of Things (IoT)
The IoT refers to the global network of smart devices, vehicles, buildings and other objects embedded with intelligent software and sensors that enable these items to communicate and collect data.
Internet Protocol Transit (IP Transit)
Internet Protocol Transit is a service that enables traffic to traverse an ISP’s network and connect to the wider Internet.
latency is the delay that occurs in transmitting data over a network. High latency can adversely impact the function of many applications.
Managed hosting is a provisioning model whereby the hardware is owned by a service provider and leased to a single client, who also pays for ongoing management and maintenance of the infrastructure.
A meet-me room is an area within a colocation data center where telecoms companies can physically interconnect circuits and exchange data without incurring local loop charges.
Metro Ethernet refers to an Ethernet network deployed across a metropolitan area and used to connect businesses to the Internet, to a larger network, or to other offices.
A network is a group of computer systems linked together. Types of networks include Local Area Network (LAN), Wide Area Network (WAN), Wireless Local Area Network (WLAN), Storage Area Network (SAN) and Metro Area Network (MAN). Networks may be further categorized based on topology, protocol and architecture.
A network of interconnected data centers enabling internet traffic exchange between networks for service providers, internet service provider (ISPs) and content delivery networks (CDNs).
N+x redundancy is a resiliency strategy that guarantees availability in the event of component failure. If a component, N, has at least one independent backup, then this is known as N+1 redundancy. Two backups would be N+2 and so forth.
Network Service Provider (NSP)
A network service provider is a business that offers direct Internet backbone access to ISPs. Examples of NSPs include telecoms companies, data carriers and wireless communications providers. Find Service Providers on Cloudscene
Platform as a Service (PaaS)
PaaS is a cloud-computing model that delivers an environment that enables developers to build apps and services over the Internet, meaning the developers do not need to deploy local hardware and software.
Private Branch Exchange (PBX)
A PBX is a telephony system within an organization that switches internal calls between users and allows users to make and receive external calls over shared phone lines.
Payment Card Industry Data Security Standards (PCI DSS)
A proprietary security standard for the handling of major branded credit cards, administered by the Payment Card Industry Security Standards Council, and designed to reduce fraud and improve the control of personal card data.
A physical server is a hardware-based computing device that provides data to other computers. The term is used to differentiate from software-based virtual servers.
Point of Presence (PoP)
A PoP is an interface or demarcation point between communicating entities such as telecoms carriers or Internet service providers. In a colocation data center, the PoPs will often be situated in a Meet-Me Room.
Power redundancy involves the duplication of critical power systems so that power will always remain available in the event of failure of one component. High availability power supplies are often configured as N+1, N+2 and 2N.
A private cloud resides on an organization’s intranet or hosted data center where all hardware, management, maintenance, security and updates are the responsibility of the organization.
A public cloud resides in a service provider’s data center and the provider is responsible for all management and maintenance. This reduces costs and improves scalability, however security may be less robust than a private cloud solution.
Power Usage Effectiveness (PUE)
PUE is an energy efficiency metric developed by a consortium known as ‘The Green Grid’. It measures the ratio of total power consumed by a data center relative to the power used to run its IT equipment. Read about modern data centers “going green”
A metal framed chassis that holds, secures and organizes a vertical stack of network and server hardware, including routers, switches, access points, storage devices and modems. Also known as a cabinet.
A standard unit of measure used to describe the height of a server mounted on a 19-inch or 23-inch rack. The height of one rack unit is 44.45mm. The common written form is expressed with a “U” eg. 1U = 1 rack unit.
Remote hands is a service provided by colocation data centers that enables customers to outsource basic IT maintenance tasks to technicians employed by the data center, thereby allowing customers to focus on their own core business. Remote hands’ technicians assist with simple tasks like running cables, checking ports, observing indicators and rebooting servers.
Service providers offer a range of services to other businesses. Services may include networking, communications, legal advice, consulting, data storage, management, maintenance and more. Also refer to Cloud Service Provider. Find Service Providers on Cloudscene
Smart hands is similar to a remote hands service but involves more complex tasks such as installing equipment, configuring firewalls, circuit testing and troubleshooting. While remote hands’ services are often offered at no additional cost, smart hands services are normally billed by the hour.
Software as a Service (SaaS)
SaaS is a cloud-based software distribution model that enables customers to access applications hosted over the Internet (typically via a web browser) and licensed on a subscription basis.
A subsea cable is a cable laid on the ocean floor between land-based stations and used to transmit telecommunication signals between the world’s continents. Modern cables are around 25 mm in diameter with an optical fiber core. Read about Google’s latest investment in an APAC subsea cable
Data centers offer space options in various sizes, from small cabinets or racks, to multi-rack cages and private suites. A suite is designed for customers seeking a larger enclosed floor space with customer-specific security controls.
Uninterruptible Power Supply (UPS)
A UPS is a device that provides emergency power when the primary power source fails, allowing equipment to continue to operate for a limited time. It can also provide protection from power surges.
Uptime is a measure of the time that a piece of equipment is active and operational. It is normally expressed as a percentage – for example, five nines reliability refers to a device that is operational 99.999% of the time.
Virtualization refers to the creation of a logical or virtual version of a computing resource, such as a server, network, operating system or storage device. Virtualization makes more efficient use of IT hardware, allowing for the provision of more flexible, scalable and lower cost services. Virtual Private Server (VPS) A VPS is a virtual machine that resides on a physical machine within a data center. The VPS runs its own virtual operating system and is offered as a service to customers who may install almost any software that is compatible with the OS.
Wide Area Network (WAN)
A WAN is a computer or telecommunications network in which the interconnected computers extend over a wide geographical distance spanning regions, countries or continents.