Stefan Groschupf - Hadoop Summit 2013 - theCUBE - #HadoopSummit
summary
Argues Hadoop is “a tape drive” — a sequential‑optimized file system unsuited to full SQL-era expectations — so SQL-on-Hadoop is often a mismatch; warns duct‑taping open‑source pieces costs real money (five devs at $200k becomes ~$1M/year). Claims the market is crossing the chasm as customers ask for business use cases, not platforms. Announces Datameer 3.0 Smart Analytics that exposes Clustering, Decision Trees and Column Dependencies plus recommendations, cutting integration-to-insight from 18 months to about three days and enabling a 5‑second Olympic performance gain example.
AI-written summary · about
chapters (9)
- 0:00Introduction and Hadoop Summit overview
- 1:12Crossing the chasm in big data adoption
- 2:55Data-driven sports innovation story
- 4:40Business value and sustainable metrics
- 8:32SQL on Hadoop limitations explained
- 10:38Hadoop as agility engine over traditional ETL
- 13:29DataMirror’s smart analytics platform
- 17:20Advice for entrepreneurs in analytics
- 19:46Closing remarks and thank you
