As organizations expand their IT infrastructure, they must eventually make a decision about whether to keep their servers on-premises or colocate them with a data center facility. The decision can have serious long-term implications for a company, as deciding to continue with an on-premises solution represents a significant commitment to future capital and operating expenses. Before committing to a strategy, companies should make sure they’re making the choice that makes the most sense for their business needs.
An on-premises data center is another way of referring to the classic private data infrastructure used by companies that keep all of their data and servers in-house. In some cases, they may have a dedicated facility for their servers and computing equipment, but more often this infrastructure is located in a dedicated room in an office building. For smaller companies, this room may not be much more than a closet housing a single server or two (hence the term “data closet”).
The advantage of this arrangement is that it allows companies to have full control over their data and who has access to their systems. This is particularly valuable for organizations with valuable, proprietary assets or sensitive customer information that must be handled according to strict compliance regulations. Older companies often have legacy infrastructure with complicated hardware and network requirements, making an on-premises solution necessary for them to keep these systems up and running without re-engineering them from the ground up with modern architecture.
In a colocation arrangement, companies place their servers and network equipment in a data center environment. By renting space in a third-party facility, they gain significant advantages in terms of network connectivity, cloud computing solutions, and technical support. The data center handles all of the power and cooling requirements, which greatly simplifies the operating expenses for their customers. Increasingly, software defined data centers (SDDCs) are offering to virtualize servers, allowing companies to migrate their infrastructure while eliminating the reliance on physical hardware. This creates a great deal of flexibility for when they need to ramp up their computing or storage capabilities.
A recent study by IDG found that about two-thirds of companies already store at least some of their data in a colocation data center. Even among organizations that rely exclusively on on-premises facilities, over 70 percent have plans to migrate some data into a colocation facility at some point in the future. Interestingly, the size of a company seems to have no impact on whether or not a company pursues a data center strategy, with companies smaller and larger than 5,000 employees equally likely to colocate at least some of their operations with a third-party facility.
Backup and redundancy seem to be the greatest motivator for current colocation trends, with a little over half of all companies surveyed indicating as such. The easily expandable storage capabilities of a data center are an obvious attraction. Although massive amounts of information are produced every day by users and other network processes, advancements in data storage have all but banished the longstanding fears that data centers may soon be running out of space.
Data center statistics indicate that about 80 percent of organizations are considering using colocation facilities to support some combination of critical projects and applications. As more companies embrace the use of big data analytics, which sorts through the massive amounts of unstructured data gathered at all levels of their networks, they face escalating processing demands that are very difficult to meet with an on-premises solution. If a company only has a private data center, it can only expand its computing power by physically adding more servers to provide the extra processing punch. Not only does this require a significant capital investment, but it also increases operating costs in the short term and long term. Those servers must be powered and cooled, and even if they’re not needed in the future, the company is still stuck paying for them.
By migrating IT infrastructure into a data center, especially one offering SDDC services, companies can easily scale up their computing needs by purchasing more server capacity. They can always scale down in the future if their needs change, and they can also use the connection-rich data center environment to leverage cloud computing resources from a multitude of providers.
Regardless of their current data center model, most organizations cite uptime reliability as one of their primary concerns with their IT infrastructure. Given the high costs of downtime, it’s no wonder that reliability consistently ranked high across a number of verticals IDG surveyed. In an on-premises solution, organizations are solely responsible for maintaining their own service, which can be an all-consuming task for IT departments that could be providing business value in other areas. Colocation data centers can take these concerns off a company’s hands with high their high SLA uptime reliability. With remote hands teams at the ready to ensure that servers stay up and running when companies need them most, data centers are an increasingly attractive solution for complex network infrastructures that must maintain data availability and deliver online services. For companies considering a colocation vs cloud solution, a data center will generally offer much better SLA reliability.
While every organization faces different IT pressures, many of them are electing to make the shift to colocation data centers to take advantage of their flexibility and reliability. With advancements in server visualization and sophisticated cloud architectures like hybrid cloud deployments, it’s easier than ever for companies to utilize the resources of powerful data center facilities while also maintaining the level of control and visibility they require over their valuable assets.