Evaluating the capabilities and infrastructure of data centers can be a confusing experience for many colocation customers. This is especially true when it comes to redundancy, which is one of the most important aspects of data center infrastructure. Data centers use a specific terminology to describe a facility’s redundant systems, but in order to understand them, it’s good to consider why redundancy is so important in the first place.
Redundancy is a major point of emphasis for data centers because a component failure can have serious consequences. When systems fail, data and services are no longer available to the companies and customers that rely upon them. This can result in lost revenue, missed opportunities, and reputational damage that can tarnish a brand for years to come.
For data centers and other service providers, their SLA uptime guarantee stipulates service expectations in terms of how much downtime customers can expect to experience over a period of time (usually a month). Routinely falling below that baseline not only costs the provider money in the form of remuneration payments, but it can also convince their customers to find a more reliable partner elsewhere. That makes data center redundancy incredibly important to most colocation facilities.
Data center power and cooling are the primary infrastructure elements with the biggest impact on a facility’s SLA uptime. Without redundant systems to support data center cooling solutions, for instance, a malfunctioning air conditioning unit could cause servers to overheat and crash. In terms of data center power, the lack of redundant systems makes it more difficult to schedule maintenance because no part of the power infrastructure can be switched off due to data center power consumption needs.
All facilities utilize a common system to describe the degree of redundancy they’ve incorporated into their data center power and cooling systems.
The symbol “N” represents the amount of infrastructure needed to operate the facility at a full IT load. It is typically used to describe cooling units or uninterrupted power supplies (UPS), but it could apply to many other aspects of data center infrastructure. The important thing to remember is that N stands for baseline capacity. A facility with N capacity has everything it needs to operate as designed, but it has no redundancies in place to accommodate equipment failure or maintenance.
An N+1 redundancy means that a facility has the capacity needed to run a full IT load with an additional component to account for failure or maintenance. As a crude example, if four bolts were required to assemble a shelf from the hardware store, N+1 redundancy would supply five bolts. Data center N+1 redundancy standards typically require an extra unit for every four needed, so if 12 cooling units are required, an N+1 facility would have 15 units.
A 2N system is fully redundant, with a completely independent, mirrored system that can fully take over operational needs should the first system go offline for any reason. These systems are considered fault-tolerant because they can provide uninterrupted service even in the event of a significant failure to one system. 2N systems are easy to maintain because one system can be shut down for repairs or upkeep while the mirrored system continues to provide for the facility’s needs.
The highest form of data center redundancy, 2N+1 systems provide a completely paralleled backup system along with additional components to account for failure and maintenance in each system. These systems offer tremendous versatility because they have full, fault-tolerant redundancy in addition to the ability to accommodate component failure without shifting over to a backup system completely.
Selecting the appropriate amount of redundancy can be a difficult decision for an organization. While it might be tempting to default to wanting the highest level of redundancy, not every industry calls for the same uptime and availability standards. Some organizations could end up paying more for redundancy they don’t really need. For most companies, N+1 redundancy is a good baseline that balances high reliability with affordable colocation costs.
Data center redundancy is an important consideration that companies should never take for granted when evaluating potential colocation partners. Understanding how data centers assess redundancy is a good starting point for evaluating the capabilities of a specific facility. Given the risks associated with system downtime, organizations can’t afford to overlook their data center’s redundant systems.
As the Marketing Manager for vXchnge, Kaylie handles the coordination and logistics of tradeshows and events. She is responsible for social media marketing and brand promotion through various outlets. She enjoys developing new ways and events to capture the attention of the vXchnge audience.