Understanding Databricks/Spark
Performance Tuning
Lesson 01: The Spark Architecture & Bottlenecks
by Bryan Cafferky from my YouTube channel
Where We're Going?
• Databricks/Apache Spark Architecture
• Performance Bottlenecks
• Tuning Options Vary By Platform
• Wrap Up
Spark Cluster
External Storage
Worker Nodes
Core Core Core Core Core Core Core Core Core Core Core Core Core Core Core Core Core Core
Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk Disk
Do Count Do Count Do Count Do Count Do Count Do Count Do Count Do Count Do Count
Send Back Result Partition Data by City
and Copy One City to Each Node Executor
Constraints Distributing the Data Over the Cluster
Driver Node
Core Core
➢ Hardware/Resources Disk Disk
➢ Software (Spark/DBR)
➢ Environment Configurations SELECT City, Count(*) FROM PhoneBook
Group By City
➢ Your Code/Application Order By City
➢ Data Source & Format
➢ Data Distribution
Phone Book
Options Vary By Platform
Delta Lake vs. Non Delta Lake
Open-Source Spark
Databricks
Wrapping Up
• Databricks/Apache Spark Architecture
• Performance
ThankBottleneck
You!
• Tuning Options Vary By Platform