Joel Pacheco Gonçalves
In the field of Information Technology, redundancy is defined as “the inclusion of extra components that are not strictly necessary to functioning, in case of failure in other components”
In other words, what happens in case of failure? Your infrastructure keeps up and running because you rely on a diverse and redundant data center partner, or you face a power outage because your data center partner is not built on redundant systems.
Let’s talk about power redundancy.
Power components in a Data Center
There are several power components in a Data Center and every one of them is a point of failure, which can incur significant financial and data losses for your company. There’s a quick list of the components:
- Main Power Source of the facility
- Power Backup Generator
- Uninterruptible Power Supply, commonly known as UPS
- Power Distribution Unit or PDU
- Rack Power Distribution Unit
- The built-in power supply of the equipment (server, router, switch, etc.)
A Data Center must increase the reliability of these critical power components by implementing redundancy, in order to provide a high availability environment for your peace of mind.
Read more: More power for your peace of mind
Power Redundancy Levels
The principle of redundancy in automated systems is providing a switchover of functionality to a standby or backup component in case of failure of a primary component. This is supposed to occur automatically and an operator should not be required for the switchover.
There are several types of redundancy (N, N+1, 2N, 2[N+1]) that typically refer to the number of power and cooling components that comprise the data center infrastructure systems.
Let’s make it simple by pointing out some of the core concepts of them:
- N: Meets base load requirements with no redundancy. The failure of any component will cause an outage, this level of redundancy describes a Tier 1 type of facility
- N+1: One additional unit/module/path more than the base requirement; failure on a single unit will not disrupt operations
- 2N: Two complete units/modules/paths for everyone required for operation. Failure of one entire system will not disrupt operations or affect performance for dual-corded loads. Both N+1 and 2N represent increasing levels of redundancy and describe Tier 2 and Tier 3 type of facilities, roughly mapping to the Tier 4 kind of facilities
- 2(N+1): Two complete (N+1) units/modules/paths; failure of one entire system still leaves an entire system with redundancy (N+1) for dual-corded loads
If you are wondering about what those Tiers mean, it’s nothing but a classification system from The Uptime Institute that has become an industry standard approach to site infrastructure functionality. Data Center Tiers are used to represent levels of projected availability.
Understanding the risks and impacts of a power outage will help you look forward to finding better Data Center partners for taking care of your equipment and data.
Power outages in The United States during 2014
Power outages are one of the top concerns of the Data Center industry. Whether you choose to build, buy or colocate your infrastructure, you should examine all of these power factors to determine what’s best for your business.
According to the Eaton Blackout Annual Report of 2014, electrical power outages, surges and spikes are estimated to cost more than $150 billion in annual damages to the U.S. economy.
This enormous amount of money varies not only by industry but by the scale of business operations. Eaton also concludes that, only in The United States, there were 3.634 blackouts in 2014. The country suffers more blackouts than any other developed nation in the world.
Despite the losses attached to downtime of services, if you partner with a reliable data center you have nothing to worry about. Technology has evolved to a level where this should not be a problem for you or your business.
Downtime, not for MDC
We are proud to have implemented the latest technology in standby power backup. Our generators (400kW) are fueled by Natural Gas and were set up following best practices for 2N redundancy. Definitely, downtime is not for MDC Data Centers.