Snowpark is Snowflake’s developer framework for developers to code in their preferred language, and execute fast and securely within Snowflake. This session aims to provide attendees with an overview of Snowpark with its benefits for data engineering and data science workloads. The hands-on lab provides a turn-key environment to explore Snowpark for Scala by going through simple TPC-H queries examples showcasing projections, aggregations, joins, and filtering using the Snowpark Scala API. The Quickstart demonstrates the ability to push down server-side complex transformations using Scala/Java functions as Snowflake UDFs. The lab also explores additional capabilities such as the generation of synthetic data by invoking Snowflake built-in functions through the Snowpark API and the use of windowing functions that can be handy for analytics on dimensional data, or time series analysis.
Prerequisites
- Two options of deployment:
- Docker: Windows or Mac Workstation with Docker Desktop Community Edition installed.
- AWS EC2 Instance: AWS account.
- Snowflake Trial Account.
- If using Windows workstation, ssh client like putty
Snowpark Overview
Getting Started with Snowpark Lab
Siva Shanmugam
Solution Architect - Cloud Data Platform & AI, Cognizant
Naresh Shetty
Chief Architect, Cognizant
Zohar Nissare-Houssen
GSI Sales Engineer, Snowflake
Kesav Rayaprolu
Senior Sales Engineer, Snowflake