Building Machine Learning and Deep Learning Models on Google Cloud Platform: A Comprehensive Guide for Beginners

© Ekaba Bisong 2019

E. . BisongBuilding Machine Learning and Deep Learning Models on Google Cloud Platformhttps://doi.org/10.1007/978-1-4842-4470-8_40

40. Google Cloud Dataflow

Ekaba Bisong¹ 

(1)

OTTAWA, ON, Canada

 

Google Cloud Dataflow provides a serverless, parallel, and distributed infrastructure for running jobs for batch and stream data processing. One of the core strengths of Dataflow is its ability to almost seamlessly handle the switch from processing of batch historical data to streaming datasets while elegantly taking into consideration the perks of streaming processing such as windowing. Dataflow is a major component of the data/ML pipeline on GCP. Typically, Dataflow is used to transform humongous datasets from a variety of sources such ...