Building an Observability Platform in 389 Difficult Steps

“During the lifetime of any company or product there comes a moment when the theoretical need for observability, whether it be exception handling, (centralized) logging, metrics, or even distributed tracing becomes a concrete need. Generally this happens after your team actually needed them, but sometimes you are lucky enough to be ahead of that moment. This talk lays out a strategy for getting your team there with some specific technology comparisons in each area. This is not a vendor talk, though we will describe pros and cons of some existing commercial and OSS solutions and how you can choose between them to meet your teams’ needs.”

Dave Worth, Engineering Manager at Strava, lays out a strategy for choosing the right tech stack depending on your business and team need. Watch as he guides you through tool sets that navigate around business constraints and regulatory concerns.

What You’ll Learn

• How to determine which technologies your team needs at your level of maturity
• How to choose the right tool for your team given your constraints and regulatory concerns in a given domain such as logging, instrumentation, etc.
• Strategies for implementing your observability platform with the least possible pain

Resources

• Slides

Dave Worth’s professional life consists of being a web and backend engineer who developed specialization in observability through building reliable distributed systems at Strava, and previously DigitalOcean. In his spare time, Dave loves cycling, jiu jitsu, and searching for another great math book to only read the first 50 pages of.