Over the last year, I clicked the mid-life crisis "ship-it" button, bought a Ford Mustang, and started down the path of High Performance Driving Education (HPDE). Learning to drive fast and safely led to many reflections on parallels between HPDE and operating mission-critical systems in the cloud.

Let's have a little fun and take a journey where you will learn to be a better driver along with some tidbits on applying SRE and DevOps principles in the cloud, even in highly regulated environments like the federal government.

We'll talk about:

The First Way: Flow

- The critical importance of creating periods of intense focus when driving and developing software.

- It's a connected system; study the map, look ahead, and work together on the track so we can all go faster (at our own pace). The same goes for your product teams and dependencies that inevitably arise in large organizations.

- Going fast requires a close eye on maintenance, aka technical debt.

The Second Way: Amplify Feedback Loops

- The faster you drive or ship software, the more critical it is to have the right amount of timely feedback. What does "just right" look like when driving and operating complex software.

- It's a team sport; feedback comes from multiple contexts. Whether from flag stations on the track, instruments in your car, or dashboards and alerts from your software platforms.

The Third Way: Culture of Continual Experimentation and Learning

- HPDE is fundamentally about continuous learning. Every time you show up, pick 2-3 things you will focus on improving. What improvements are in your backlog?

- We often talk about guardrails to add safety to our systems, but that's not precisely the correct analogy. The problem with guard rails is that impacting them is usually pretty disastrous. You'll notice race tracks have dirt, grass, or tire barriers on the outside areas of fast corners to reduce the chance of impact, damage, and injuries. We need to mirror this with our software platforms through techniques like graceful degradation, circuit breakers, and the ability to quickly re-deploy.

- In HPDE, you want to be the one introducing a fault (skidding) and practicing recovery from it, not waiting for a fault to happen unexpectedly and send you sideways. Practice with your software platforms before the unexpected happens.

- At the end of every track day, drivers will consolidate their notes for what they learned, what they are still working on, and what to focus on next time. These are reviewed at the start of the next track day. Your Incident Reviews need to have the same focus.