Making sense of the small print of cloud provider contracts

Any CIO who has been to a trade show will know that many a salesman will promise the earth when it comes to selling a service but that the reality of the offering doesn't really match those promises.

All too often the sales brochure will have phrases within them that can appear to say one thing but often mean quite another to the cloud provider keen to have something to fall back on when trouble strikes.

So when faced with the glossy marketing literature what are the key phrases to look out for? You know; the ones that may appear innocuous at the time but whose double meaning can prove problematical when dealing with the day-to-day running of a cloud service for your users.

Cloud outages are very real – just ask Amazon. While cloud providers acknowledge these events can and do happen, there will be certain assurances within the marketing that these are rare events. While no-one can and does guarantee 100 per cent uptime, there are guarantees of 99.99 per cent and sometimes 99.999 percent (the so-called “five nines”) within the prospectus. Indeed, the key phrase here is "uptime cannot be guaranteed". So while this allows a certain leeway for the providers, they still do try to offer as near as they can, but there are certain caveats for customers to be wary of.

However, cloud service providers do have some level of service to stick to. There's a problem though: while failure to meet these criteria should result in compensation for lost productivity according to any service level agreements put in place, it is often the case that the rules are stacked against the end-user.

According to Keith Bates, chairman of the Cloud Computing Centre, not all instances are covered despite the wording in the prospectus implying that they will be.

“We must question just how likely it is that a customer will ever receive a payout?” said Bates. “It is after all, no great secret that while SLAs look good for the customer on paper, they are always designed to work in favour of the service provider.”

Two scenarios can happen where the amassed downtime may or may not result in compensation. The first one is when a customer experiences continuous service downtime of 20 minutes, leaving him unable to access his email account, applications or data. Productivity is reduced; meaning business-critical work is lost. This customer is compensated as continuous downtime of 20 minutes in a single day breaks the agreed 99.99 percent continuous uptime SLA in his contract.

A second customer, however, experiences downtime of five minutes for four days in a row. On four separate occasions, he is unable to access his email account, applications or data, disrupting the productivity and damaging the business on four separate occasions. This time, even though the total was 20 minutes downtime, the customer receives no compensation as this 20 minutes was not continuous. It pays to see past the sales talk and see just how “rigorous” the SLAs say they are in the brochure.

Another phrase to look out for is “customer support available 24/7”. This may mean one thing to the CIO, but can mean something quite different to the provider. Ray Welsh, head of marketing at secure data centre provider, The Bunker, said that “many 24/7 data centres [only] have a security guard on duty overnight, who will answer the phone” while some have email only service desk contact. He added that potential customers should ask about technical support available throughout the week – for changes made to the service provided, issue resolution, etc.