The proverb above - Everything fails, all the time - is a quip often cited to surveyors who are just getting started with building on AWS.

• Truth be told, I didn’t really grasp its significance at first; not yet understanding the cloud’s ecosystem, I interpreted it as just another recursive and cryptic mantra.

• Now, almost parodying Neo’s encounter with The Architect in Matrix Reloaded, this mantra punctuates my purpose in the cloud.

You see, Vogels was a huge driving force behind the reckoning of highly-available systems.

• Instead of rolling a boulder up a hill for all eternity, he deftly reminds us to instead plan for failure. In cases like:

• - Hard drives failing, so leverage S3 data-stores on multiple hard drives to prevent data loss.

• - Computing hardware can fail, so virtual machines can be automatically restarted elsewhere.

• - Even entire data centres can fail, so there are multiple facilities per region designed to be used in parallel to pick up the slack.

For our purposes, outages in IT can cause losses in trust and money - the bread & butter of business.

• This is my angle as a solutions architect: knowing how to prepare against losses by implementing the right tools and architecture.

• Considering the exponential growth of cloud and new technologies, it’s easy to see that designing for failure is a fundamental principle on AWS.

• And architecting reliable systems is precisely what I’m trained to do.