An Adventure in Multi Cloud with Kubernetes and Distributed Data
summary
Deploys a Yugabyte across AWS, Google, DigitalOcean cluster on Kubernetes via CAST AI, using Helm charts, cloud storage classes for PVs, node labels for affinity, and an nginx ingress with Let's Encrypt. The demo brings the cluster up in minutes, wires a Redis-compatible table for a containerized e‑commerce app, kills a leader node and shows the cart remains consistent after failover. Networking uses a Cilium virtual multi-cloud private network and benchmarks with tpcc 128→126 tpm (negligible loss), with write latencies rising from ~51ms to ~121ms across clouds.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 2:30Getting started with Kubernetes
- 5:22Demo cluster setup
- 11:58DNS name
- 12:48Deployment file
- 15:23DNS configuration
- 17:10Deployment
- 18:39Persistence
- 20:19Latency
- 21:35High Availability
- 23:56Benchmarks
related talks
41mAn Adventure in Multi-Cloud with Kubernetes
35mKubernetes journey
35mKubernetes Persistent Data Challenges – AZ, Region and Multi-Clou...
33mKubernetes-Driven Multi-Cloud Deployments: Maximizing Scalability and Flexibility
27mKubernetes Everywhere: Lessons Learned From Going Multi-Cloud
36mThe Architecture of a Multi-Cloud Environment with Kubernetes
37mDemo: Take Control of your Multi-Cloud PaaS and Kubernetes
35mHybrid Cloud Powered by Kubernetes [I]