3 ms·
I'm not particularly concerned about EC2 outage.
Despite the fact that my servers were offline for almost four days I thank my lucky stars every day for the existence of EC2.
Anyone who thinks that building something as sophisticated as EC2 can be done perfectly just doesn't understand computers.
Before cloud hosting we had to run our own servers and had far more outages and solving the problems had to be done in person running around with bits of hardware and copying data yuck. I have been able to sit back during this meltdown thinking "Amazon has a bunch of super smart people working real hard to sort out my problems, I just have to wait till they have it worked out."
This outage is not a result of laziness, incompetence or inattention to detail, and it hasn't happened for lack of attention to building reliable and robust systems. Building highly available systems on a scale never seen before must be unimaginably hard.
All it means is we can stop pretending there is some sort of non stop, never breaks perfect computer. It's just time to be more careful and redundant. If however your operational procedures are poor and you lack backups and lost data then you are probably pretty annoyed, but it's you who should be annoyed at yourself for not covering your butt.
Cloud computing isn't perfect, it's just great.
So I'm still grateful for the EC2 and I thank Amazon for it.
I just wish the Amazon EC2 servers weren't so damn slow..... :-)