
Jeff Magnusson: How To Create A Self-Service Data Platform For Data Scientists
Secrets of Data Analytics Leaders
00:00
Creating a Self-Service Batch Job Execution Service with Flotilla API
The chapter explores the open sourcing of Flotilla, an API designed for managing batch-oriented tasks in data science departments. Citrix's utilization of Spark and Docker on ECS for batch execution, performing data wrangling in Spark and model training in Docker containers is highlighted, showcasing how Flotilla abstracts over ECS to streamline job execution processes.
Transcript
Play full episode