Making the Impossible Impossible: Improving Reliability by...
summary
Argues SLOs nudge teams toward percentage fixes but some outages are avoidable by design: using state machines to forbid invalid payment transitions, relying on garbage-collected runtimes or Rust’s ownership/borrow checker to eliminate use‑after‑free bugs (the CVE search finds 534 CVEs for use‑after‑free in 2022), and using Vitess vreplication to do online schema backfill plus coordinated cutover. These patterns remove whole classes of failure and scale in production at companies like PlanetScale and GoCardless.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 1:54The Perils of Success
- 8:45Memory Safety
- 12:39Garbage Collectors
- 13:04Ownership
- 18:59Schema Migration
related talks
21m4 BIG Software App Challenges For FinServ
14mSelective Reliability Engineering: There Is No Single Source...
45mThe Map Is Not the Territory: How SLOs Lead Us Astray, and What We Can Do about It
43mLearning to Live with Errors
22mResilience Is Not Enough
47mThe 7 Quests of Resilient Software Design
01h 02mTools for simplified large scale systems
41mOptimizing for Value - Fight Failure Demand With Bottlenecks