There was a massive outage of the cloud services of Amazon Web Services on Tuesday which provides back-end support to various famous sites like Pinterest, Buzzfeed and Netflix amongst others. It was totally unexpected and took hours to be restored finally. Many such sites have their back-ends stored on Amazon servers which enables them to function without having to buy their own cloud space. It is quite a common phenomenon these days to have shared cloud services which makes Amazon a definite choice. But the disruption did cause many such sites to suffer.
Though not all the sites witnessed a slowdown, there were malfunctioning and slowness reported which caused a lot of chaos. It wasn’t that Amazon itself was excluded from this mishap. Its health dashboard service was impacted badly too because of the outage and it could not update itself for the first couple of hours.
It is noted that though Amazon does not usually suffer such outages but whenever it occurs, it does last for several hours. The reason which is found to be most common for such outages is the software which could be due to manual or bug errors. Such an incident happening to AWS could be quite a considerable loss for Amazon as AWS accounts for a reasonable amount of the revenue. In the last quarter of 2016, it accounted for 8% of its revenue share.
The outage reportedly began somewhere around 12.30 eastern time and lasted till 4.50 Eastern Time making it one of the biggest outages impacting several people using the AWS servers. The most common cause for the outage is that more and more people are opting for third party service on clouds as it saves them the huge spends required on infrastructure. Also, there is no maintenance issues that get generated for them. The reason that experts site for such outages is that these web services pick different things from different services – like images could be from a different server while the other details could be from other servers. This makes the tangled servers clash leading to broken and ousted linkages.
This experts believe is the major reason for such heavy outages to happen. Since mostly these services in the united States are used by the e-commerce websites, who import price lists and images to appear attractive, the heaviness of the servers takes a toll on the overall services and makes this a software issue difficult to manage.
To avoid this, it can be done that the information be distributed across different servers like Google or Microsoft but since not all companies can afford to avail such an expensive trail of storage, it is usually providers like Amazon that are used. Companies usually prefer to sacrifice on such cascading effects considering the opportunity cost of using multiple servers to be quite high compared to a loss on such a day.
So the solution that best helps is for the business to have their own backup data on their servers as well which could help them on such rainy days and also ensure safety of their data.