There was a GMail outage, as you no doubt know, last Tuesday. Apparently Google Apps was down too. Critics of cloud computing (wherein rather than having locally hosted software such as Word, Excel etc on your server, you instead use remotely hosted versions such as you can find with “Google Apps”) would argue that outages such as these question the whole basis of cloud computing. If you are a GMail user, then of course losing your email for a couple of hours is rather annoying (although with GMail you could still access old emails, sort folders and even write new mails; they just wouldn’t be sent until GMail came online), but imagine if you were at work and suddenly you couldn’t access, for example, your Word of Excel (or equivalent) – it would probably rather scupper your productivity. Therefore, critics understandably argue, why on earth would you trust your IT to remote servers over which you have no control?
A couple of obvious arguments against this come to mind, and have already been stated. Firstly, the technology is still in its relative infancy, and as time goes on you might expect reliability to improve. Secondly, we all have to contend with local outages when our servers crash from time to time, or the router goes on the blink or whatever. Just recently at my company we had problems with our email and for a few days it was touch and go whether an email sent to you arrived or not. Google promise that GMail will be available to you 99.9% of the time – and so far it has been. That’s pretty good.
But I think a third argument is simply as people use cloud computing more and more, then our attitude to its supply will become more akin to our attitude to the supply of electricity. In ye olden days, companies would have provided their own power through local generators etc., and only as the national grid became more and more reliable did business finally decide that it made more sense to get their power from a “remote source”. Now, if there is a power cut, everyone is in the same boat and your customers understand that there’s nothing you could and that these things, in a sense, are “an act of god”. Similarly, if there is a transport workers strike, as happens in London more often than one would like, then again it’s acknowledged by all, it affects all, and there is no great surprise that some time has been lost; it’s just “one of those things”.
At the moment however, if you’re a company with a cloud computing model, and you were hit by a “cloud outage” then your customers, clients and suppliers might think “eh? What are you talking about?” and they might be less than sympathetic to you having missed a deadline or whatever. However, when we’re all on the cloud, we’ll all just accept an outage as “one of those things”, and no-one will bat an eyelid.