Reliability is the subject of a debate that is difficult to settle – and this article goes on to add that “in a virtualised environment the loss of the network connection can cut access to multiple servers.”

Well, of course it can because the network is a single point of failure.

And, do you know what?

The server that the virtualised environment is running on is also a single point of failure because there is only one motherboard, one set of processors and one instance of the memory.

That’s why you need to double up on everything – it’s called fault tolerance and is sadly lacking in the majority of cloud providers.