Understanding Databricks & Apache Spark Performance Tuning: Lesson 02 - Spark Hardware
Description
Following up on Databricks Performance Tuning with the best place to start: allocating Spark clusters. If you don't allocate sufficient resources, nothing else will fix the problem. How many nodes? How large should the driver and workers be? Do you need GPUs or CPUs? Should you use Photon? These and many more questions will be covered in detail. Support me on Patreon https://www.patreon.com/bePatron?u=63260756 Slides https://github.com/bcafferky/shared/blob/master/DatabricksPerfTuning/lesson02_DatabricksPerfTuningHardware.pdf Referenced Video by Daniel Tomes, Databricks https://www.youtube.com/watch?v=daXEp4HmS-E&t=2111s See the Full Playlist Here: https://www.youtube.com/playlist?list=PL7_h0bRfL52qw5SkSI_P_kURU2njM1UO-
Description from YouTube. Full content on the video page.
More from Bryan Cafferky
NewsMaster Dimensional Modeling Lesson 03 - Understand the ETL Pipeline
The video explains the typical stages of a data warehouse ETL pipeline, including pre-staging (raw data), staging (cleaned data), operational data store (snapshot), and data mart (star schema). It also details the benefits of having multiple stages, such as easier debugging, data recovery, and auditability, and how this maps to the Medallion Architecture (Bronze, Silver, Gold).
TutorialsMaster Databricks 2nd Ed: Lesson 4 - Use Databricks for Free!
Databricks now offers a free edition for learning purposes, providing access to most core features within a serverless environment without requiring a credit card. This free edition has limitations, including small compute resources, no custom cluster allocation, and the absence of R or Scala language support, and is not suitable for sensitive data or production use.
TutorialsMaster Databricks 2nd Ed: Lesson 3 - Understanding Clusters
This video explains Databricks clusters, detailing their components like driver and worker nodes, configuration options such as autoscaling and Photon acceleration, and how to create and manage them within Azure. It also covers common interview questions related to cluster sizing and performance tuning, emphasizing that Databricks clusters are essentially Spark clusters enhanced with the Databricks runtime for cloud environments.
