Platform Engineer
William Aass Dahlen is an experienced platform engineer working for Telenor Norway where he is specialized on self-service, automation, and cloud platforms. He started his career as a systems administrator more than 10 years ago and has been working primarily with Public Cloud, Kubernetes, and Platform development.
On a bright summer’s day, without load and running pipelines, most applications thrive. Self-imploding system calls and memory overflows are horrors that mostly belong to a bygone era.
Suddenly, the weather shifts, and clouds turn dark. Your applications shiver, and requests are being dropped left and right. On-call engineers scramble, only to discover that your chosen cloud platform is having a hard time—not only with your applications but with your competitor’s too.
The root cause analysis reveals this was a fluke, a freak-of-nature event, never to happen again. Your cloud provider assures you of this and offers a 10% service credit as restitution.
However, 10% of the monthly cost is not going to cover the thousands you lost in revenue while the cloud platform was down. So, what do you do next?
While situations like these are rare, they do occur. In this talk, we will venture into the realm of planned service outage testing using chaos engineering and discuss the philosophies and practicalities of multicloud redundancy.
The talk will be in Norwegian
