Oliver Ratzesberger, Teradata | Hadoop Summit 2014
summary
Argues Hadoop must drive measurable business outcomes not just visualizations and claims a production cadence of hours and days where weeks is already too slow while warning agility is a spectrum between rapid prototyping and governed systems. Describes concrete tools and practices: virtual sandboxes (<5 minute SLA), 90-day timeboxes, a crowdsourced data hub, and a QueryGrid that pushes roughly two thirds of query work down to Hadoop. Gives use cases — billion‑image processing for search and fraud, sensor data predicting locomotive failures within 12–24 hours — and says integrated, prepackaged stacks (Hadoop plus Teradata, R, SAS, Storm) make those outcomes repeatable.
AI-written summary · about
chapters (15)
- 0:00Welcome to Hadoop Summit
- 0:29Introduction to Oliver Ratzesberger
- 1:11Focus on business outcomes
- 2:36Pressure on CIOs to move faster
- 3:59Agile vs waterfall methodology
- 5:23Balancing agility and stability
- 7:07Agility push from CEO level
- 9:11Agile development at eBay
- 13:45Teradata’s relevance and execution
- 15:52Customer use cases and integration
- 17:37Big data project drivers
- 19:47Unified data architecture
- 21:52Enabling experimentation with data
- 22:34Teradata’s role in innovation
- 23:38Pre-packaged solutions for customers
