Scalability & Performance Overview
Purpose: For platform engineers and architects, explains openCenter's approach to scalability — how we test limits, set performance targets, and provide tuning guidance for production deployments.
Philosophy
openCenter is designed to run production workloads at enterprise scale. Every release is validated against defined cluster-size profiles, and we publish tested limits so operators can plan capacity with confidence rather than guesswork.
What This Section Covers
| Page | Focus |
|---|---|
| Cluster Limits | Tested maximums for nodes, pods, services, and GitOps reconciliations |
| Performance Tuning | Component-level tuning for etcd, API server, kubelet, and FluxCD |
| Benchmarking | Methodology, tooling, and published benchmark results |
| Resource Optimization | Right-sizing, autoscaling patterns, and quota strategies |
| Network Performance | CNI benchmarks, MTU tuning, and eBPF acceleration |
| Storage Performance | I/O benchmarks, CSI tuning, and provisioning latency |
| Observability at Scale | Scaling the monitoring stack without drowning in cardinality |
Scale Profiles
openCenter validates against three reference profiles:
| Profile | Nodes | Pods | Services | Use Case |
|---|---|---|---|---|
| Small | 3–10 | ≤500 | ≤100 | Development, PoC, edge sites |
| Medium | 11–50 | ≤5,000 | ≤500 | Single-team production, departmental |
| Large | 51–200 | ≤25,000 | ≤2,000 | Multi-team enterprise, shared platform |
Related
- Capacity Sizing — resource requirements for initial deployment
- Reference Topologies — deployment patterns by use case