InfoQ / LinkedIn Engineering
The scale is more than 250 clusters, 4,000 services, and 100,000 deployments a
week. Argo CD stalled on status issues, concurrency, and false drift. In the new
engine, each deployment runs in an isolated pod.
- Infrastructure failures: from 925 to 63, with 3.7 times more deployments.
- Failure detection: the median dropped from 16.1 to 2.45 minutes.
- Argo CD: still handles cluster add-ons.