What is the ideal CPU and RAM requirement for Airbyte services in a production-grade EKS environment?

The knowledge sources don’t provide per‑service CPU/RAM sizing, but they do give cluster‑level and instance‑level guidance for production EKS. From the Self‑Managed Enterprise implementation guide for production: • Run on Amazon EKS on EC2, not Fargate. • Use memory‑optimized instances (e.g. M7i / M7g families). • Use instances with at least 2 vCPUs and 8 GiB RAM. [Enterprise prereqs; 2.1 prereqs] • For a production‑ready deployment, plan for an EKS cluster with a minimum of 6 nodes. [Enterprise prereqs] For scaling workloads (syncs) themselves: • Start with a mid‑sized instance (4–8 cores) and tune based on load. [Scaling guide] • Ensure the cluster can schedule roughly

2 × <number-of-possible-concurrent-connections>

pods at once. [Scaling guide] • Connector pods’ CPU/memory can be controlled via

jobs.resources

in

values.yaml

(e.g. setting

limits

and

requests

for CPU and memory). [Scaling Airbyte after install] Because the docs do not specify “ideal” CPU/RAM per individual Airbyte service (webapp, server, worker, temporal, etc.), you’ll need to start from these cluster/instance baselines and then tune per‑service Kubernetes requests/limits based on your actual workload and monitoring. 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.