Bryson Koehler, The Weather Company & IBM - #IBMInterConnect 2016 - #theCUBE
summary
Claims that weather data fuels Watson and an IoT data platform with 100 terabytes a day ingested and 62 data types processed, serving forecasts for 2.2 billion locations that now update every 15 minutes. Examples: minute-scale rain alerts (7–10 minutes), rooftop personal weather stations feeding city-level accuracy, aircraft turbulence telemetry routed in real time to following planes. Reported accuracy around 78%, built cloud-scale ingestion, analytics and alerting pipelines, and ongoing work to make those capabilities self-service inside IBM.
AI-written summary · about
chapters (11)
- 0:00Welcome and introduction
- 0:42Weather Company and IBM merger
- 2:27Big data and weather forecasting
- 4:34IoT and real-world data insights
- 7:23Improving weather forecast accuracy
- 9:50IBM's IoT platform vision
- 12:35Explaining IoT and sensor data
- 16:09Predictive vs prescriptive analytics
- 17:35Lessons from IBM acquisition
- 20:00Top priorities post-acquisition
- 22:30Closing remarks
