Authors: Thomas Barns & Manzoor Mohammed
Estimated read time: 2 minutes
At the heart of an SRE strategy is a desire to engineer a reliable solution 'up-front' rather than spending time on fire-fighting. However every organisation finds it difficult to break away from simply keeping the show on the road. Here are 3 practices that successful teams use to do this:
- Get past alerts and dashboards and tell the big story
- Bring new ideas into the engineering teams
- Build confidence by experimentation
1. Get past alerts and dashboards and tell the big story
If alerts and monitoring dashboards are your main contact with system data you're operating right in the fire-fighting zone. Spend time understanding relationships between long term user behaviour and capacity, then back to the user experience.
Then you're in a space where you can think proactively about how to improve your platforms and your product. What are the key things that would make a difference?
Link this big picture back to what really matters to users and SRE will maintain a stable platform and drive business performance.
2. Bring new ideas into the engineering teams
Use these to focus on building for performance. Don't be distracted by the latest tech fad, instead create efficiency by thinking outside the box. As platform cost efficiency increases system complexity is reduced. This not only helps to save costs, but slashes the time needed for traditional support tasks, leading in turn to operational efficiency.
This results in an innovation-focussed team where performance is not only a habit, but becomes an easy one.
3. Build confidence by experimentation
Know what questions you are trying to answer, don't make it up as you go along. Empower teams to try things out for themselves, help them recognise it's ok to get it wrong and to learn from that. But control your experiments carefully so that learnings are clear and actionable.
Experimenting with architectures, reducing costs, improving performance, will lead to real future-focussed change.