If you’re an IT service provider or other B2B tech company, there’s a good chance that unplanned downtime is one of your biggest data centre nightmares. After all, it’s something that won’t just affect your business, but your customers’ business, too – landing you with complaints, missed SLAs and possibly even a black mark against your reputation.
That’s why resilience should be top-of-mind when it comes to choosing a new colocation provider. The trouble is, a lot of people don’t really know how to get an idea of a data centre’s resilience – they treat it as a box-ticking exercise, not a careful examination of factors like management and personnel.
What is data centre Tier classification?
In part, this comes down to the ubiquity of the Uptime Institute’s Tier classification scheme in the way data centres market themselves.
Introduced in the mid-1990s, the institute’s benchmarks – where Tier 1 is least resilient and Tier 4 most resilient – were developed as a means for customers to compare facilities that are mostly very different from one another in their design and upkeep. They were never meant to be interpreted as quality stamps, and treating them as such can lead to some questionable data centre sourcing decisions.
Tier 3 versus Tier 3
Case in point: most customers are advised to look for a Tier 3 facility (with Tier 4 requiring a standby mains power supply – something that very few data centres can offer). However, while two or more Tier 3 sites will have a similar design and meet the same minimum standards, they can also be very, very different from one another in terms of resilience.
UPS redundancy is one of the areas where this disparity may be most pronounced. In order to meet the requirements of Tier 3, a data centre only needs to offer one redundant UPS module – a configuration referred to as N+1. At the opposite end of the scale, it’s possible that another Tier 3 facility could provide 2(N+N) UPS redundancy (as with our Manchester data centre, which offers Tier 4-standard UPS redundancy, but can’t be described as Tier 4 due to the lack of a standby power supply from the mains).
Considering that UPS failures are the number one cause of unplanned data centre outages (accounting for a quarter of all downtime, according to a 2016 study by Emerson Network Power), it should be obvious that Tier classification alone won’t tell you all you need to know about the resilience of your colocation provider.
What else should I look for?
If you go about things the right way, you’ll probably spend a lot of time looking at a data centre’s UPS redundancy, environmental monitoring, power and cooling, and so on.
Outside of this, it’s important to remember that tech failures don’t tend to happen in a vacuum. Most unplanned downtime can actually be traced back to poor management and maintenance of the data centre environment, so it’s vital to look at factors like quality and quantity of staff, contracts and SLAs with maintenance service providers, and any quality control accreditations.
If you see any evidence of cost-cutting or poor housekeeping, this may be a sign you need to look elsewhere – Tier 3 or otherwise.
Want to find out more about how a B2B tech company should go about choosing a data centre? Click the link below to download our free whitepaper on the topic.