Delta Live Tables

Question 1

What is Delta Live Tables?

Answer 1

A managed data pipeline tool exclusive to DB
1/ETL Tool for data warehousing
2/Managed structured streaming for Spark

Question 2

What problem does Delta Live Tables solve?

Answer 2

1/Write pipelines in simple, declarative code
2/Easily implement and monitor data quality constraints
3/Easily track and view data lineage

Question 3

Why do customers care about Delta live tables?

Answer 3

Helps accelerate data pipeline development, reduces burden of performance tuning and maintenance

Question 4

When should you position Delta live tables with a customer?

Answer 4

If you see any data pipeline/ETL work in databricks,batch, or stream

Question 5

How do Delta Live Tables work?

Answer 5

1/Customers write DLT-specific SQL/Python code in a notebook
2/Once written, switch to workflows, click DLT and set up cluster to run the notebook
3/Once run DLT populates the UI with a data lineage diagram and writes out logs that are used for monitoring

Question 6

What are key features of Delta Live Tables?

Answer 6

1/Automatic table optimization
2/Enhanced autoscaling for spiky workloads
3/Easily create test pipelines

Question 7

True or false : DLT is only for streaming

Answer 7

False : Its continuous or triggered mode

Question 8

True or False : DLT is low-code data engineering and not for advanced Spark users

Answer 8

False : DLT improves efficiency, making pipeline code simple, can handle optimization, testing, error-handling, monitoring, and documentation

Question 9

What to look for when pitching DLT

Answer 9

1/Pipelines with unpredictable workloads
2/Complex pipelines with many downstream tables
3/Mention of data lineage and/or quality

Question 10

Competitors?

Answer 10

DBT, and any structured streaming

Question 11

Cost

Answer 11

It costs more than others but better performant. Starts at $0.20 DBU for Core