The cloud isn’t just about servers anymore

I’ve been working with servers for a long time, ever since I was a child I’ve been experimenting with servers of various kinds at home. Fast forward to 2009, my first experience with the “cloud” was EC2 on AWS.

At the time, AWS was still in its infancy and only had a handful of services, they’d just announced their second region — Azure didn’t even exist yet (though it had been announced). At that point in time we only ran a single basic workload, partly because we didn’t have the processing power in-house to analyse the data we were working with, also as a bit of an experiment to see what the whole cloud thing was about. Until then, we ran systems and applications either from our own in-house server room or from a co-location provider in Arizona.

During 2011 we started having difficulties with the co-location provider in Arizona and made the call to move everything to AWS’ us-west-1 region. Instead of creating new EC2 instances and migrating the applications and data, we used the Import/Export service to copy what we had to AWS like-for-like. As another experiment we put a load balancer in front of two instances we’d imported for some high availability.

A couple of years later in 2013, ap-southeast-2 launched and we were happy with how AWS had performed so far and made the decision to move almost everything in-house to…