Published inLightstepHQWhat made SLOs so messy (and what we can do about it)The joys and sorrows of the SLOs that connect engineers and end-users across a distributed system.Dec 7, 20211Dec 7, 20211
Published inLightstepHQSampling, verbosity, and the case for (much) broader applications of distributed tracingOriginally posted as this twitter thread.Nov 17, 2021Nov 17, 2021
The Three “Digital Transformations” (and how to bring them together)How technology transformed Operations before it transformed Product, and how to connect these two separate practices.Sep 20, 2021Sep 20, 2021
Published inLightstepHQWhy Tracing Might Replace (Almost) All LoggingThis is a thread about why tracing will gradually replace most logging (… and explored through the lens of a relational data model).Apr 23, 20215Apr 23, 20215
Published inLightstepHQMaking Continuous Delivery “Global,” via ObservabilityThe easier part of Continuous Delivery (“CD”) is, well, “continuously delivering software.” The harder part is doing it reliably. :)Apr 7, 2021Apr 7, 2021
Published inLightstepHQThe Two Drivers of Cardinality (and what to do about them)Cardinality can be expensive but it doesn’t have to be. This is a thread about what it’s good (and bad) forMar 27, 2021Mar 27, 2021
Published inLightstepHQObservability will never replace Monitoring (… because it shouldn’t)Observability won’t replace Monitoring – it will *augment* Monitoring.Mar 27, 2021Mar 27, 2021
Published inLightstepHQDeep Stacks Versus Deep SystemsThe differences between “stacks” within a CPU and “stacks” within a distributed system (and what that means for APM and observability)Mar 27, 2021Mar 27, 2021
Published inLightstepHQAPM Is DyingHow architectural evolution is going to create deep systems (and kill APM)Mar 27, 20215Mar 27, 20215
Published inLightstepHQRegrets About Dapper and MonarchWhat we did right – and especially what we did wrong – when we designed Google’s tracing and metrics systems.Mar 27, 20211Mar 27, 20211