For those in software development and IT operations, DevOps has likely been a term you've grown to hear and read about more and more lately. The principles and ideas of increased collaboration, transparency, empathy, and continuous improvement have spread far beyond the fast-paced world of startups and forward-thinking SaaS organizations.
This is due in large part to events such as DevOpsDays which take place all around the world. Smaller and more intimate than larger tech conferences such as Velocity and Craft, DevOpsDays offers a format that allows for more engagement and immediate takeaways for those in attendance. As a practitioner and evangelist of DevOps, I've had the pleasure of attending or speaking at 25 DevOpsDays events since 2013, as well as helping to organize two in Denver.
Portland Oregon hosted their own DevOpsDays recently (August 9th and 10th) and I was excited to be in attendance for this year's edition in the City of Roses. With a booming technology scene in the Pacific Northwest, it's no surprise that the event sold out quickly and was well attended by many notable companies such as Nike, Intel, and Columbia.
The two-day format of DevOpsDays provides a diverse method of delivering high-level DevOps resources and concepts that touch on both organizational culture and technical detail. Partially following an "un-conference" formula, DevOpsDays also include self-organizing Open Spaces, which allow those in attendance to take on the role of participant rather than strictly an attendee.
In Open Spaces, a wide range of topics is suggested and then discussed in an open format where participants can ask questions; share stories, lessons learned, and resources; and engage in casual conversations that often provide more valuable insight than the presentations and lectures that take place during the first part of the day. A major theme and focus of all DevOpsDays is diversity—not only by providing quality content that spans many topics and disciplines but also ensuring that speakers, organizers, volunteers, and attendees of all genders, race, ethnicity, and more feel welcome, heard, and included.
Portland's DevOpsDays speaker lineup included well-known figures such as Kelsey Hightower (Google Cloud), Jennifer Davis (co-author of Effective Devops), and many practitioners from organizations large and small in the Portland area. Co-author of The Phoenix Project and the forthcoming The DevOps Handbook, Gene Kim was in attendance as well, engaging with attendees both in person and through the ongoing conversation on Twitter using the hashtag #DevOpsDays. True to the spirit of all DevOpsDays events, this one was committed to providing a great deal of variety on not only topics but voices and experiences as well.
NoOps: Moving operations to the codebase
Kicking things off on day one, Kelsey Hightower (image above, and mentioned below) delivered an opening keynote that included a live demonstration on updating security certificates automatically via Kubernetes. Known for fascinating onstage demonstrations and thought-provoking anecdotes, Hightower led the audience to a deeper understanding of the term "NoOps."
The idea is to move operations-like efforts closer to the codebase. By enabling more of our teams (regardless of job title) to contribute and improve code, applications become inherently "Ops-friendly." It's this leveraging of the IT operations team's expertise that allows applications to become more resilient and scalable. As I tweeted at the time:
Understand your DevOps history
The second talk of the first day at the conference was an amazingly well researched and definitive breakdown of the history of DevOps. Nell Shamrell-Harrington of Chef (image below) gave the audience a deep dive into how we have arrived at our current understanding of how to build and maintain resilient systems at scale. Shamrell-Harrington's presentation made for an excellent verbal and visual history on not only the basic concepts of DevOps, but also the evolution that has landed us where we are today. Covering everything from early concepts of interchangeable components to the more modern principles of agile software development and everything in between, the 30-minute talk could easily serve as required viewing for anyone looking to gain greater context on the principles of DevOps and how they have evolved.
When O'Reilly Media released the book Site Reliability Engineering (SRE) earlier this year, a new discussion emerged on the differences between DevOps and SRE. David Blank-Edelman of Apcera shed light on why this conversation has come about and helped refocus our attention on the key components of SRE and its place within DevOps. Through a better understanding of concepts such as normalized reliability metrics and service-level agreements, the audience was able to take away several ideas critical to maintaining high availability and resiliency.
For me, the idea of error budgeting is one of the more fascinating and thought-provoking concepts of the SRE book (and Blank-Edelman's talk), and one that I believe all teams should spend time understanding in greater detail. As someone who spends a considerable amount of time advising on best practices for increased uptime of systems, error budgeting is a forward-thinking concept that is reshaping the way teams think about providing increased availability.
The metrics drive is on; more extroverted discussion needed
Other interesting talks included ways to implement metrics that go beyond basic monitoring and alerting on system resources. Keeping tabs on issues important to the business as a whole, rather than simply hardware, network, and software concerns, was the key takeaway from the talk by Caskey Dickson of Microsoft. JJ Asghar of Chef introduced those of us who self-identify as introverts to ideas on better managing and enjoying conferences. Simple suggestions such as taking breaks to get away from the crowd or having a card game on hand to play with others helps those who often run low on energy from a lot of social interaction or struggle to approach others they aren't familiar with. In total, we heard from eight speakers the first day before heading into the Open Space portion of the afternoon.
Burnout: Is it the biggest threat to DevOps?
Closing out the day, ThoughtWorks' Ken Mugrage delivered a poignant message on a topic that is both sensitive and extremely important within the DevOps community: burnout. Oftentimes within our fast-paced industry, we tend to push the limits of our own physical and mental abilities. Brought on by a number of contributing factors such as stress, workload, and alert fatigue, burnout has surfaced as one of the largest threats to our industry.
Mugrage introduced a number of key concepts and leading research that can help us not only identify but also address and prevent burnout much earlier. As many in attendance can attest, this topic hits close to home for a lot of people. We should all be thankful this is a topic of conversation that continues to emerge at DevOps events around the world.
Getting beyond chaos-driven development
After the usual evening drinks and networking in a beautiful location, the second day started with presentations that covered topics such as how to manage conflicting agendas and incentives between operations, development, and security; how to avoid chaos-driven product development; and how to uncover what should be monitored beyond application performance and system incidents or outages.
The closing talk on the second day included great tips on how to bring these inspiring DevOps concepts and stories back to our teams and organizations in a way that actually brings positive change. All too often, participants of DevOpsDays return to their offices full of new ideas and motivation only to have their ambitions quashed by senior management who fail to grasp the concepts and benefits.
By all accounts, the first DevOpsDays held in Portland was an immense success. If you've never been to one of these events, I highly suggest you look into attending one in your part of the world soon. You won't find a better conference not only to introduce and initiate newcomers, but also to impress those who are well seasoned in the principles of DevOps.
Image source: Alice Goldfuss/@alicegoldfuss
Keep learning
Choose the right ESM tool for your needs. Get up to speed with the our Buyer's Guide to Enterprise Service Management Tools
What will the next generation of enterprise service management tools look like? TechBeacon's Guide to Optimizing Enterprise Service Management offers the insights.
Discover more about IT Operations Monitoring with TechBeacon's Guide.
What's the best way to get your robotic process automation project off the ground? Find out how to choose the right tools—and the right project.
Ready to advance up the IT career ladder? TechBeacon's Careers Topic Center provides expert advice you need to prepare for your next move.
