Shipping Software with an SRE Mindset
summary
Applies SRE practices to shipped on‑prem software: a productionized back-trace collector yields 300KB mini‑dumps ~35 ms after a crash so engineers get full thread stacks and locals, cutting weeks of debugging; instrumentation records every disk I/O and nanosecond latency so historical queries answer “what changed”; tracing can produce 150k+ spans per request (potentially millions/sec) and uses Jaeger for on‑demand analysis; live interrogation via BPF/dynamic tracing answers unknown unknowns without redeploys; typical 7‑day fixes shrink to ~15 minutes.
AI-written summary · about
chapters (6)
- 0:00Start
- 2:33Software Sucks.
- 3:23Why Ship Software?
- 10:14Expose Telemetry
- 16:31Logging for humans
- 17:51are solved by: Dynamic Tracing
related talks
24mShipping at Speed: A Microsoft Clarity SDK Case Study
28mTracing the Journey into Distributed Tracing
14mWeb Apps, through the Lens of SRE (Site Reliability Engineering)
27mMean Time To SRE: A DevOps story from adidas
28mThe shoemaker’s children have no shoes - why SRE teams must help themselves
30mThe Path to SRE
11mDistributed Tracing - Shipping Quickly And Confidently
54mUnraveling the Differences