How serverless systems fail: A walkthrough of serverless failures, catching them, and avoiding them

Every change in technology brings new and unexpected ways for systems to fail. We’ll go through the ways AWS serverless systems fail when you don’t own the infrastructure and it automatically scales for you.

YOU own your availability but serverless architecture means you own less of your infrastructure. How do build reliable systems when you’ve chosen to outsource nearly all your system’s infrastructure? Serverless presents familiar systems reliability problems but in new and unique ways.

We’ll cover areas and how they can be addressed.

• How do you handle systems that can automatically scale almost without limits?
• In what ways do serverless systems break and how do you architect in fault tolerance?
• Where do you go to find out what is wrong?
• The importance of price as a new failure dimension.
• And finally how do we as ops people remain in tune with our dev teams to ensure they’re building reliable systems?

The listener should come away with a better understanding of the challenges their organization will face with systems reliability if they choose to adopt serverless architecture, both at a technical level and organizational level, and how they can mitigate or avoid these issues.

I’m an engineer for an early stage statrup that has chosen to go the serverless route for building its product. In building our product we have routinely found new ways for our product to fail as an AWS serverless system.