Thinking outside the box in DevOps and SRE: avoiding common pitfalls, improving quality, and on-call experience with a bit of creative thinking

DevOps and SRE engineers are humans. As humans, we are susceptible to cognitive biases that affect our abilities to build reliable solutions and perform on-call tasks. In this session, we will look at real world examples and how outcomes can be improved via creative thinking and mitigating bias.

DevOps and SRE engineers are humans. As humans, we are susceptible to cognitive biases that affect our abilities to build reliable solutions and perform on-call tasks. In this session, we will look at real world examples and how outcomes can be improved via creative thinking and mitigating bias. You will learn: - How tunnel vision can limit our abilities to deliver value to the customer and what we can do about it - How during active incidents we have a higher propensity to fallback on our cognitive biases and how it can be improved with simplified runbooks - Potential noise effect on critical infrastructure, such as ECS, EKS, Elastic Beanstalk, k8s, etc., selection process - How lessons from behavioral economics can help to improve unit test coverage, datastore selection, and help to solve other engineering pitfalls

While there definitely will be great in-depth talks about various new technologies at DevOpsDays, the goal of this talk is to look at the human factor in selecting and using these technologies. Imaginary and real world examples will be used throughout the talk in order to demonstrate pitfalls and how outcomes can be improved. Since examples will be from the area of DevOps, SRE, and backend s/w development, basic understanding of the current technologies is suggested, but not required.