Seth Dobrin | IBM Interconnect 2017
summary
IBM lays out a practical stack: hybrid data management linking legacy systems to public/private clouds, an API-driven unified governance layer built on Apache Atlas that captures metadata automatically and enforces policies, and machine learning (Watson) brought to private clouds. Open source trunks on Apache Spark power the platform. Example scale: a client has 400 terabytes a day of changes, so governance must be real-time and automated. Models trained on industry datasets stay owned and can be specialized with a company’s private data.
AI-written summary · about
chapters (20)
- 0:00Welcome and introduction
- 0:45Role of Chief Data Officer
- 1:07IBM Analytics five key areas
- 2:13Unified governance as enabler
- 3:17Data science and machine learning
- 3:58Commitment to Open Source
- 4:41Data DevOps and metadata
- 5:50Data as a first-class asset
- 6:57Progress bar of enterprise data strategies
- 8:26Inward vs outward focus of CDOs
- 9:49How to become a Chief Data Officer
- 10:52From genetics to data science
- 12:17CDO Summit agenda
- 13:20Vision for unified governance
- 14:26Governance in the big data era
- 15:29Data classification challenges
- 16:57Data ownership and model training
- 18:21Why IBM leads in data solutions
- 19:03Data at scale and innovation
- 19:45Closing remarks and Q&A
