Modern companies use computers in almost all aspects of doing business—communication, information storage, accounting, and day-to-day business functions. A data center is a centralized physical facility where corporate computers, network, storage, and other IT equipment that support business operations live. The computers in a data center contain or facilitate business-critical applications, services, and data.
Data centers come in all sizes—they may fill a closet, a dedicated room, or a warehouse. Some businesses with a large amount of IT equipment in their data centers may require more than one data center facility. Companies can also choose to rent server space and have someone else maintain their data center.
A data center might extend outside of a physical facility by using a private or public cloud to augment its operations or storage. A virtualized data center can use servers in remote locations when needed to run larger workloads.
Because data centers contain so much expensive IT equipment, they have special requirements for security and power.
Data centers can have different levels of security, latency, and resiliency—these levels are known as tiers. Tiered data centers allow for hierarchical data storage, where the files and data for applications that users access most frequently, or that require very high performance, can live in one tier, while data that is not accessed as frequently can live in a different tier. In public cloud-based storage-as-a-service models, tiers with low latency are generally more expensive than tiers with higher latency.
The tier with the lowest latency and the fastest access typically lives closest to the users, where users can easily interact with it on a daily basis. The fastest tier is known as Tier Zero. Businesses such as financial services or medical research that use high-performance computing for big data analytics are frequent users of Tier Zero storage. Lower performing storage tiers can act as back-up for the primary tier, or as storage for data that is less frequently accessed. Archived data can live in the lowest and least expensive storage tier, where high latency is less of an issue.
Physical data centers are limited by space and their reliance on hardware and equipment. Now, servers can be virtualized, where processing power and memory are separated from the hardware in a virtual machine. With data center virtualization, administrators can create a virtual data center infrastructure using remote servers to share the workload and storage. A virtual data center requires less equipment, less power, and less room than a traditional physical data center. A virtual data center can also access or “burst into” either a public or private cloud when more storage or processing resources are required. Virtual data centers are also known as software-defined data centers because all of the services they provide, including networking and storage, are delivered through software rather than hardware.
Cloud service providers can provide virtualized data centers as a service, taking over the management and delivery of a company’s core operations, networking, and storage. These providers share virtual networks, storage, and servers among several different companies, shifting storage space and workload processing among servers as needed. An organization can also use part of its own infrastructure to create a private cloud for the same purpose or use a combination of private and public clouds, which is known as a hybrid cloud environment. A virtual data center can co-exist with an existing physical data center, running in parallel, or businesses may use modern data center equipment to run a fully virtual data center. A hyperconverged infrastructure (HCI) integrates compute, storage, and networking equipment with virtualization software and optimizes it to run as a single system. With an HCI, all of the data center functions run on software that is closely integrated with the hardware. This software-driven approach enables the automation of storage tasks and operations that previously required server hardware to be manually adjusted.
Modern businesses are figuring out how to take advantage of new IoT technology that allows them to learn more about their customers and better engage them. The data collection and analysis that supports this type of research and provides an optimized customer experience requires some heavy lifting from the data center. Virtualization, HCI, and the cloud are transforming data centers, allowing them to become flexible and responsive to fluctuations in workloads in real time, and to manage larger amounts of data. Software-defined data centers can be much more cost efficient than physical data centers to build and maintain. Using a virtual data center infrastructure, especially when combined with a private or public cloud, allows businesses to save money on physical infrastructure, space, and power. Virtualized data centers also allow companies to have more flexibility in choosing their hardware since cloud-based infrastructure-as-a-service (IaaS) platforms run on a variety of different hardware. And with tiered storage plans, the cost of public cloud services is becoming more affordable. As the Internet of Things expands, and the amount of data that is generated on a daily basis increases exponentially, the scalability and processing power of virtual data centers will become more and more critical.