Multi-Cloud Strategy in Practice
summary
Defines multi-single cloud vs true multi cloud and claims Target runs identical containers across public cloud, private cloud, data centers and every store Kubernetes cluster so services and data can be cooperative and movable. Requires two concrete capabilities: CI/CD (they use Spinnaker and Unimatrix pipelines) and bi-directional data movement (they build Autobahn to stream change events). Shows staged store rollout—deploy to 5 then 10 then 50 stores—prefers availability over consistency and outlines PIE, Maze and CDN FaaS as next steps.
AI-written summary · about
chapters (19)
- 0:00Introduction and background
- 3:00Defining multi cloud
- 5:19Hybrid cloud explained
- 7:20Cloud deployment models
- 9:43True multi cloud definition
- 11:20Is multi cloud right for me
- 13:23Resiliency and latency needs
- 14:46Cost considerations
- 16:12Architectural flexibility
- 17:38Omni-channel retail focus
- 18:02CI/CD for multi cloud
- 19:51Spinnaker deployment demo
- 22:10Data movement essentials
- 23:52Target Retail Platform data flow
- 26:20Autobahn data movement
- 27:56Target Cloud Continuum
- 29:12Future with PIE and functions
- 30:30CDN as runtime environment
- 31:03Closing remarks
related talks
38m50 Shades of Multi-Cloud Infrastructure
20mShould we multi-cloud?
43mGetting to a Multi-Cloud Deployment - A Case Study of
37mImplementing Multi-Cloud Compatibility: What, How, and Why
45mMultiCloud
49mApplications architecture spanning multiple clouds
15mJourney to Hybrid Multi-Cloud
43mUsing Kubernetes, Spinnaker and Istio to Manage a Multi-cloud Environment