A Quantitative Study on Argo Scalability
summary
Measured Argo CD scalability shows 10k applications in <40 minutes when using Argo CD with a wrapper plugin for per-app customization, Cluster Loader 2 to inject load, and Cluster Loader 2 and Prometheus/Grafana for metrics. Experiments expose that CPU becomes the bottleneck (single t2.medium replica spikes; 3 replicas don’t linearly reduce sync time) and that the default resync period = 3 minutes strongly affects overall sync delay; options are reduce customization cost, add compute, or accept longer delays.
AI-written summary · about
chapters (6)
- 0:00Start
- 1:51Research Question
- 3:56Hurdle 1 - CPU Limitation
- 5:19Hurdle 2 - Default Resync Period
- 7:39KCP-Edge Community
- 13:11Experimental Framework
related talks
23mArgo at Scale: Navigating Complex Multi-Dimensional Deployments Across H...
18mUnveil the Secret Ingredients for Argo CD in the Enterprise-Scale
29mMastering Efficiency in Argo CD: Scaling Smarter, Not Costlier
10mKeynote: Pushing the Limits: Scaling ArgoCD’s Repo-Server To Handle 30...
29mUnleash ArgoCD Observability Superpowers
36mHow Adobe Planned For Scale With Argo CD, Cluster API, And VCluster
34mInfrastructure at Scale Using ArgoCD
27mWhy Is It Taking so Long?! Shining the Light on Application..