Site Reliability Engineering is sometimes referred to as the practice of treating operations as if it were a software problem. It involves automating everything to achieve scale and resiliency while minimizing manual human intervention. As organizations grow and mature, systems become more complex and reaction times get slower. Expectations quickly change from "keep us up during peak hours" to "how many nines?" seemingly overnight. By the end of this talk you'll know what an SRE team looks like, what my experience has been like, and how your organization might benefit from a dedicated SRE team.

We've all heard the term DevOps, and if you've not. Please tell me what happiness feels like? DevOps is a term that was coined in 2008, and since then the idea has evolved, grown, and even flipped a few times. Let's go over what DevOps is, what it isn't, and how it can help you at your current job.

This is not just another “Use Chef, Puppet, or Ansible” talk. Tools looking for problems can lead to bigger headaches than applying sound practice to your efforts. There will always be cases when using a pure configuration management solution like the ones above may not solve all problems. Sometimes faster delivery capabilities, lack of necessary features within a plugin, or work out of sync with development teams can lead to issues ranging from annoyance to production outage. Going from manually edited application configurations on a production server to a fully automated deployment and testing solution can take many paths. If you follow agile development efforts, you can start handling configurations in a “Crawl, Walk, Run, then Fly” phase. I’ll use Apache, Tomcat, and Puppet to create some examples of problems with editing configs right in prod, and move towards some solutions to test, build, and deploy configurations just like a software engineering deployment pipeline.

Maybe you think you're using Kanban now, but are you just tracking tasks in big ugly list? Kanban's focus on Flow, Work in Process, Visualization, and Continuous Improvement requires a change of mindset. With Kanban, teams must prioritize completion of work, maximize effort, and constantly increase efficiency. Where Kanban can provide the most value is in dealing with unplanned work, competing priorities, and unknown variables at project start. Examples of these types of projects include:

• Legacy projects migrating to a cloud platform
• Budgets shifting towards newer investments
• A new merger brings on new properties
• Existing projects have shifted into an end-of-life, but not end-of-support phase

Or maybe, you're a support desk, service desk, or an on-call workforce for production issues. Kanban allows teams to learn from the work completed and pivot out of the chaos of all urgent requests all the time.