Latest Posts

Latest Comments


What to do when your data centre provider has an extended outage


Posted by |

Organisations who take IT and data centre services from third party providers, services that may very well include business continuity, might be forgiven for scratching their heads this week. After a protracted outage suffered by Capita over the Bank Holiday weekend they are no doubt puzzling over the obvious question “How can we trust them to deliver continuity for our business when they seem unable to provide it for their own?”.

It’s certainly not something that instills a great deal of confidence, and ultimately that’s what customers of service providers like ours need above all else. Confidence that 99.99% of the time it will be business as usual; confidence that the remaining 00.01% of the time will come down to an incident dealt with so proactively it doesn’t even register; confidence that their business is safe because ours is safe; confidence that they won’t be roused from their Bank Holiday slumbers by an incident that has run away out of control to become a major outage and a massive problem.

But right now, better to take action than to waste precious time wondering ‘how come?’ So back to first principles: check your own backup provision and contingency planning. If the same type of outage happened to your provider, how would you be affected, what are your options? Given that the provider does in effect represent a single point of failure for most businesses, this is probably a good a time as any to go back to them and ask for an in-depth chat about their business continuity arrangements. Is their plan robust, mature, tested and refined regularly? And while you’re at it, why not ask about the metrics that matter, not the unchallenged figures bandied about in glossy marketing material, but the hard performance stats?

Whether you’re searching for a data centre provider or reappraising your existing provider in the light of recent events, it’s these statistics that provide the incontrovertible evidence of performance. So review and assess them, see if there are any trends, look at the response and remediation effort with previous incidents; ask yourself if they are achieving the service levels that you’d be happy to present in evidence when you are asked why your company experienced a whole day of downtime.

Be diligent. Not all service providers are the same, and certainly the disparity in investment levels and actual infrastructure quality can be huge. Scratch beneath the service and look for those rather mundane indicators of a truly fit-for-purpose data centre. Like more than one electricity feed; properly sized generators; a constant fuel supply; carrier redundancy; location, as in not on the flood plain or on the flight path or in a perceived ‘target’ zone; its security set-up; the actual state of the physical fabric and data centre innards. Go and visit if you can, and talk to existing customers.

And then more questions for the data centre provider. Do you own and manage everything yourself end-to-end or is there outsourcing and sub-contracting involved? What accreditations and certifications do you hold? When were you last audited and with what result? Who are the people you are entrusting with keeping my business safe and up and running?

Of course, how a provider handles themselves during this ‘discovery phase’ will go a long way to instilling the requisite degree of confidence – or not – in their abilities. A responsive, understanding, transparent and cooperative approach speaks volumes. A hesitant, evasive and reluctant one is akin to a damning silence.

P.S. Just in case you were wondering, over the past three years neither of Redcentric’s primary data centres in Harrogate and Reading have experienced any unscheduled downtime at all. Now that is something you can take to the board.



Post a comment

Comment submitted! Comments needs approval before being displayed.