Data foundation for analytics and AI
summary
Cloud data stack that ties Amazon S3, S3 tables, and Apache Iceberg into SageMaker Unified Studio, claiming zero ETL for up to 15 sources and an S3-native lakehouse with 3x faster queries and 10x transaction rates. Vendors report rapid adoption (1,000+ customers creating 5,000+ tables each, tens of millions of weekly requests). Examples: HyperPod cuts training time ~40%, FanDuel cuts analytics cost ~80%, BMW runs 10M messages/hr with 99.9% uptime; P50 query latency ~8s for zero-ETL flows.
AI-written summary · about
chapters (12)
- 0:00Introduction to data foundation
- 1:05Why data foundations matter now
- 3:33Customer success with data foundations
- 5:57AWS pillars for data foundations
- 7:19SageMaker Unified Studio overview
- 13:17Lakehouse architecture explained
- 16:49Zero ETL and data integration
- 20:21Future-proofing data investments
- 23:30AI-ready data governance
- 26:43Unstructured data for generative AI
- 27:45Security and trust in data platforms
- 28:07Next steps and resources
related talks
47mBuild cloud data management across analytics, ML & gen AI applications
28mTurn Data Into Insights
25mHow to build a platform for AI and analytics based on Apache Iceberg
47mDataOps for AI-Ready Platforms
19mBuilding modern data lakes on Amazon S3
43mData warehousing reinvented for today's needs
52mPreparing data for reporting, ML, and AI on AWS
29mGrowing insights through hybrid data architectures