December 2018 Summaries

Ronnie Chen, an engineering manager at Twitter and a technical diver, shared insights into managing failures during his talk at Chaos Conf 2018. Drawing parallels between technical diving and complex systems, Chen emphasized the importance of preparation, training, and judgment in handling failures. He recounted his experiences with technical diving, highlighting how a combination of training, real-world practice, and an understanding of complex failure patterns is crucial for survival. Chen argued that focusing solely on a single failure factor is simplistic and that understanding the interplay of multiple factors is vital. He advocated for practices such as pre-mortems, fire drills, and empowering less experienced team members to lead, as these strategies can enhance a team's overall competence and preparedness. Ultimately, Chen suggested that success in complex systems is often unrecognized because it results from quiet, incident-free operations, rather than dramatic recoveries from failures.

The experience at AWS re:Invent 2018 highlighted the growing importance of Chaos Engineering, a practice that involves purposefully creating failures to test system reliability, and the widespread adoption of containers managed by Kubernetes, which has become a standard in the tech industry. The event, attended by over 55,000 people, featured key discussions on how Chaos Engineering is increasingly recognized as essential for ensuring system robustness, with companies like Gremlin offering tools to simulate various failure scenarios. The presenter, recently appointed Director of Product at Gremlin, emphasized how the rise of managed services and complex interdependencies necessitates robust testing practices to maintain system reliability. The text also underscores the importance of being able to simulate and manage failures to ensure seamless user experiences, as more services are automated and interconnected. Additionally, the text invites readers to explore further resources on Chaos Engineering through community platforms and trials offered by Gremlin, reflecting the ongoing industry shift towards reliability as a critical feature in software development.

Kriss Rochefolle's presentation at Chaos Conf 2018 explores strategies for convincing executives to embrace Chaos Engineering, emphasizing the need for influence and emotional appeal over purely rational arguments. Drawing from his experiences at the French National Railway Company, Rochefolle suggests building familiarity with the concept among C-level executives and tailoring communication to their individual concerns and motivations. By highlighting the inevitability of failure and the potential for Chaos Engineering to enhance system resilience and minimize the impact of incidents, he aims to transform skepticism into support. Rochefolle advocates for using stories, analogies, and positive emotions to make the concept less intimidating and more appealing, ultimately seeking to inspire executives to become active sponsors rather than mere gatekeepers.