Prescriptively Benchmarking Kubernetes System and Application Using K-Bench

on YouTube ↗25m

" summary

K-Bench is a configurable framework that provisions Kubernetes resources from YAML, runs 1000-pod startup test style workloads and reports millisecond-granularity 50/90/99 API and operation latencies by combining novel client+server timing (callback and event tracking) to measure end-to-end convergence. It exposes a container interface to run fio, iperf3, redis and other benchmarks, ships blueprints for density, bandwidth and IO tests, pushes diagnostics to Telegraf→Wavefront/Grafana, and identifies surprising results like a virtualized cluster outperforming bare metal.

AI-written summary · about

" chapters (13)

1. 0:00Introduction
2. 0:50Performance Aspects
3. 3:12KBench Overview
4. 4:34KBench Features
5. 7:51Basic Terminology
6. 10:30Design Framework
7. 16:00Metrics
8. 17:54Benchmarking with KBench
9. 20:02Preintegrated workloads
10. 21:29Deep infrastructure diagnostics
11. 22:31Use cases
12. 23:38DP Internode
13. 24:16Conclusion

" related talks

• 20mKubeVirt Benchmar...Bhumitra Nagar, Dhruv Bhatnagar · Data on Kubernetes Day Europe 2026
• 44mBenchmark testingDustin Black · DevConf.CZ 2019
• 06mIGNITE: How to benchmark … poorlyPhilipp Krenn · OSMC 2022
• 05mCeph BenchOps: Following DevOps Practices for Benchmarking CephIvo Jimenez · Cephalocon Barcelona 2019
• 16mWhy Kubernetes Applications Require a New Approach to Testing using TestkubeBruno Leonel André Lopes · CloudOpen @ OSS LATAM 2022
• 16mWhy Kubernetes Applications Require a New Approach to Testing using TestkubeBruno Leonel André Lopes · Open Source Summit Latin America 2022
• 31mLoad Testing Kubernetes: How to Optimize Your Cluster Resource Allocation in Production [I]Harrison Harnisch · CloudNativeCon + KubeCon Europe 2017
• 24mWhy Kubernetes Applications Require a New Approach to Testing using TestkubeBruno Leonel André Lopes · Continuous Delivery Summit 2022