Stateful Superpowers: Explore High Performa...
summary
Demonstrations show TiKV serving 1 million RPS from a 12‑node cluster (12 storage nodes, 5× replication) over local NVMe while Grafana dashboards confirm latency, and a preloaded 10 billion keys dataset proves scale. CloudNativePG brings a highly available Postgres via 90 lines of YAML that handle init, backups to object storage and automated failover. CubeFS and CSI/COZY integrations illustrate running distributed filesystem analytics and ML on Kubernetes with operators for day‑2 ops and deterministic performance.
AI-written summary · about
chapters (10)
- 0:00Welcome
- 0:43Cloud native storage matters
- 3:11Beyond just volumes
- 6:03CNCF storage projects
- 7:07Demo: cloud native PostgreSQL
- 13:49TIKV deep dive
- 18:07TIKV benchmark demo
- 21:41Real-world use cases
- 25:37Summary
- 25:57Q&A
related talks
27mFrom Zero to Hero: Scaling Postgres in Kubernetes Using the Power of...
37mData On Kubernetes, Deploying And Running PostgreSQL And Patterns For Databases In a Kubernetes Cluster.
50mUnleashing the Power of Azure for PostgreSQL
54mPostgres: From Cloud to Hybrid and On Prem Again
24mUsing a Distributed Key-Value Store
50mFrom VMs to Cloud-Native PostgreSQL in Kubernetes...
33mSponsored Session: Open Source Cloud Native Storage Solutions for Stateful Applications from ...
10m⚡ Lightning Talk: Navigating the Complexities of Stateful Workloads in Kubernetes