Recorded at SpringOne Platform 2016.
Speakers: Katie Mooney; Director, Marketing Operations, zData; Cahlen Humphreys; Big Data Solutions Architect, zData & Dillon Woods; Chief Technology Officer, zData
The goal of this session is to teach companies how they can use real-time data to make predictions with Spring Cloud Data Flow, Spark 2.0 and Spark ML. It will cover how to train your model using Spark ML in Spark 2.0. Once trained we will show you how to make real-time predictions using the model and Spring Cloud Data Flow.
In this session we will point out key differences in Spring XD that have been resolved in Spring Cloud Data Flow. We will highlight why we feel that Spring Cloud Data Flow has a much more promising future with it’s real time analytics dashboard written in Spring. This session will also highlight Cloud Foundry and how it is used to quickly deploy and integrate new stream features.
What we cover:
-Best practices for streaming data
-Bus services for data point transport
-Stream scaling and development
-Feature cleansing, normalization, and transformation techniques
-Training and Validating models in Spark and Spark ML
-Monitoring your streaming applications in Cloud Foundry