Data Science in the Open Cloud Exchange model
summary
Shows an actual deployment: Open Data Hub runs on the Massachusetts Open Cloud using OpenShift and Kubernetes, with 250 cores, 1.5 TB RAM and Ceph storage, exposing an S3 API. It wires Jupyter notebooks + Spark operator to run experiments, integrates TensorFlow and considers Seldon/Kubeflow for model serving, and supports early adopters (bee‑health mapping, citation‑network analysis). Operability is the main caveat: OpenShift scale, capacity planning and multi‑tenancy cause friction but the platform is live and usable. Open Data Hub on OpenShift
AI-written summary · about
chapters (15)
- 0:00Start
- 0:28THE PROPRIETARY PUBLIC CLOUD
- 2:23WHY NOT JUST USE A PRIVATE CLOUD
- 3:47WHAT ABOUT THE OPENSTACK CLOUDS
- 4:49IS A DIFFERENT MODEL POSSIBLE?
- 5:08OPEN CLOUD EXCHANGE
- 10:03THE MASSACHUSETTS COLLABORATORS
- 10:34MASSACHUSETTS OPEN CLOUD
- 11:25THE OPEN DATA HUB IN THE MOC
- 17:26COMMON PATTERNS
- 20:16DATA SCIENCE IN THE OPEN DATA HUB
- 21:48DEMO
- 33:31LESSONS LEARNT AND WHAT'S NEXT
- 39:52OPEN DATA HUB IN CZECH REPUBLIC
- 42:47TRY IT YOURSELF
related talks
27mAbhinav Joshi & Tushar Katarki, Red Hat | KubeCon + CloudNativeCon Europe 2020 – Virtual
38mOS-Climate & SIG-SRE: Benefiting from the Operate First Community Cloud
15mAn open cloud for european researchers, staff and students
20mBuild an e2e analytics application using DataHub
35mCase Study: AI-as-a-Service on Kubernetes at Scale and In Production
21mPowering Open Data Hub with Ray
13mAn open cloud for researchers, staff and students
39mHow We Hosted 200+ Data Scientists in a K8S Cluster