Continuous Cluster Performance Validation @ Dynatrace
A day in the life of the cluster performance engineering team @ Dynatrace
The >2Mio LOC Dynatrace core cluster code base sees 120+ daily code commits from 40+ globally distributed engineering teams. While literally speaking different languages, our engineers express different coding styles (“old school” Java vs. new Java8 constructs like Lambdas, Streams etc.). Our two-week agile development process allows our DevOps team to push production-ready changes into our SaaS instances after every 2-week sprint, and every 4 weeks the updates go out to our managed (=On-premises) customers. This process is fully automated and pushes these changes through various stages of our deployment pipeline and orchestration engine.
We, the Cluster Performance Engineering Team, are tasked to ensure no change is impacting the clusters capability to deliver “high quality under load” (aka performance, scalability, throughput, resource consumption) as we need to ensure a resilient, robust and performing cluster. Why? Because Performance Matters!
In this session, Andreas Grabner & Thomas Steinmaurer lift the secret of how a two and a half full-time equivalent team tackles this day 2-day challenge. Join us and let us show you how “A day in the life of the cluster performance engineering team @ Dynatrace” looks like.
Thomas has been around 15+ years in relational database technology before joining Dynatrace. Being around from the beginning when we started the challenge building our new SaaS and On-Premise next-generation APM solution mainly being responsible for the Cassandra backend, he is now additionally focusing on load testing, performance and scalability aspects of the Dynatrace Cluster on a daily basis.
Here you can create the content that will be used within the module.