Six deployment strategies compared: rolling, blue-green, canary and more
JeremyCMorgan · x · 2026-11-19
A comparison of six deployment strategies and their operational tradeoffs:
- Recreate: stop old, start new, accept downtime
- Rolling update: replace instances in batches
- Blue/green: run two full copies, switch traffic at once
- Canary: route a small share of users to the new version
- Shadow: mirror real traffic to the new version, discard responses
- Feature flags: deploy dark, release later
Situation-based advice: small teams on Kubernetes/ECS → tuned rolling update; bad releases cost money within minutes → blue/green; high traffic with trusted metrics → canary with automated checks; risky feature in a normal release → feature flag; rewrite to test under real load → shadow; app that can't run two versions → recreate. The author's default: rolling update with a readiness probe plus feature flags on anything risky, covering most releases with 25% extra capacity. Verified against Kubernetes 1.37, Argo Rollouts v1.10.2, Flagger v1.45.0, Gateway API v1.7.0 docs.
More from coding & agent
- New MAC Bench finds frontier agents coordinate poorly: 8-agent teams win by exploration, not teamwork — AlexGDimakis · 2026-11-19
- Dev Argues AI Needs One Interface: Claude or GPT Plus Filesystem or Sandbox — _AustinCalvert_ · 2026-11-19
- 19 Core Concepts of AI System Design: From KV Cache to Multi-Agent Systems — blaizedsouza · 2026-11-19
- 10 Free Repos That Teach AI Agent Harness Engineering: Tools, Memory, Rules — blaizedsouza · 2026-11-19
- LangSmith links coding agent spend to merged PRs to measure ROI — LangChain · 2026-11-19
- Pichai demos Gemini 3: napkin sketches become websites, images become board games — heyneighbor · 2026-11-19