Tuesday, February 28th was a dark day for Amazon Web Services (AWS) – literally, as Amazon’s catastrophic four-hour outage left some individuals unable even to turn on their internet-connected house lights. An estimated $310 million (£249 million) is said to have been lost by firms that rely on the cloud hosting service, with Netflix, Salesforce and Adobe amongst many of the large organisations that felt the outages’ crippling effects.
The impact was so widespread that Amazon themselves were affected, and it’s a fiasco that will no doubt leave many users and would-be users questioning whether the AWS cloud – and the cloud in general – can be trusted with the data and applications on which their business depends.
(Learn more about cloud hosting for SMEs and startups in our guide.)
So what’s the most important lesson we can take from the great AWS outage of 2017? For us, it’s that the cause of the outage wasn’t due a hardware problem or even a freak malfunction – it was down to an accidental typo in a routine operation that led to the unintentional shut down of “a larger set of servers … than intended”.
Surprising? It shouldn’t be. Time and time again, studies and statistics have proven that the number one cause of downtime is due to human error. According to the Uptime Institute, for example, 70% of the problems that data centres face are as a result of human error and poor management – not hardware failure, natural disasters or cybercrime.
The AWS outage therefore serves as a cautionary tale for service providers and users of cloud server hosting and IaaS solutions. Two key takeaways to keep in mind are:
- If your cloud provider goes down, not only will your business suffer, but your own customers will too. This can lead to a loss of clients and severe reputational damage. You’ll also be helpless to solve the problem until your cloud provider comes to a solution on their end.
- The outage highlights how even best-in-class data centres can suffer downtime as a result of human error. Amazon has some of the most sophisticated data centres on the planet, supporting a lot of large digital businesses, but a simple typo was still enough to knock a lot of their services offline.
If your business is sensitive to downtime, the AWS outage highlights the importance of choosing a cloud provider whom you totally trust to look after their infrastructure and train their staff competently – even if you’re seemingly buying a commodity service like IaaS. Take your time to find a provider with a good track record for uptime, and one that is responsive to support and outage incidents.
Learn how to navigate the cloud with our guide – it’s written especially for SMEs and startups.