Skip to main content
This is lesson 4 of 6 in this module Course 38% complete
Prev Next Skip

JDBC Performance Bottleneck

Premium Content

Sign in with your account or sign up to access this lesson.

Sign In Sign Up

The lesson explains how Spark executors connecting to a dedicated SQL pool via JDBC must all route through the control node, forcing sequential execution and eliminating parallelism, resulting in dramatically slower queries.

Previous Next
Previous Module Azure Databricks
This Module Azure Synapse Spark
Apache Spark Architecture in Synapse Analytics
Apache Spark Architecture in Synapse Analytics
Apache Spark Notebooks in Synapse Analytics
Apache Spark Notebooks in Synapse Analytics
Demo: Loading and Querying Parquet Files in Synapse Studio
Demo: Loading and Querying Parquet Files in Synapse Studio
JDBC Performance Bottleneck
JDBC Performance Bottleneck
Polybase Solution with Scala
Polybase Solution with Scala
Demo: Temp Views and Language Mixing
Demo: Temp Views and Language Mixing
Next Module Azure Synapse Dedicated SQL