Delta Tables 101: What is a delta table? And how to build one?

Welcome to the September edition of our Azure Synapse Analytics and Microsoft MVP series!
This month, Ryan is joined by David Alzamendi, an MVP from Brisbane, Australia. David will show us the basics of delta tables, what it is, how to create a delta table using a Synapse Spark Pool, and how to consume the information using Power BI.
Azure Synapse Analytics Community site: aka.ms/Synapse...
Join the Azure Synapse Influencer program!
aka.ms/Synapse...
David's Twitter: / techtelkcorner
Ryan's Twitter: / rmajidimehr
Synapse Twitter: / azure_synapse
#azuresynapse #powerbi

[@sinadandan]

Thanks for the video, it was actually very helpful. I would be amazing if you had the notebook attached to the video's descriptions

[@TheIRox7]

How robust delta format inside synapse is? I keep seeing lots of content about it but some consulting company still explain it’s not mature enough and that « nobody » use it in production. But the combination of Synapse / Delta / PowerBI look so cool !

[@JovanPopovicJocaPC]

Delta format in Synapse Spark is the same as delta quality in OSS Apache Spark, maybe with some additional improvements customized for Synapse. Serverless had some perf issues with delta, but it is constantly improving. In serverless, the usage ratio between Parquet and Delta is roughly 2:1.

[@TheIRox7]

@@JovanPopovicJocaPC Thank's for the anwser, I'll need get a try on this. I miss so much my UPDATE SQL statement for my Lakehouse :)

[@sidsriramoffl1785]

TNice tutorials helped sooo much!

[@tinribz]

How is this any different from just loading a csv file into PBI?

[@rameshakustagi5189]

Great video, want to use dataflow or Pyspark & flatten very complex nested JSON. can you please make a video around that? Also want to know if that can be made as a generic configurable component.

[@da8233]

Thanks for the video , add notebook on video description

[@nagoorpashashaik8400]

I am able to create delta table in lake database with dbo schema using PySpark. What if we need to change dbo schema to stg schema, Is it possible using PySpark code df.write.format("delta").saveAsTable(f"{db_name}.{Delta_TableName.lower()}", mode = 'ignore')

[@endahpurboyekti7023]

on Program inside the screen.