Archived webinar · August 28, 2020

Databricks performance: the 2020 webinar record

This online meetup has finished. Its archive preserves the published agenda while separating historical performance messaging from current technical guidance.

Archive status

The webinar has ended, and no verified recording is attached

The source page linked to an older website URL for a recording. Because the migrated record does not provide a confirmed media file, this archive does not present that link as a working replay.

Date: August 28, 2020
Time: 11:00 am–12:00 pm
Format: Online meetup

Databricks performance webinar artwork listing the date, time, and speaker — Original artwork provides the date, time, and speaker record.

Published agenda

Diagnosis came before tuning and Delta Lake guidance

The session focused on Spark and Databricks performance practices available in 2020.

11:00

Welcome

Rodney Joyce opened the online meetup.

11:05

Performance session

Gin Jia covered diagnosing Spark cluster issues, scale-up and scale-out choices, and job-cost tradeoffs.

11:45

Questions

The listing closed with audience Q&A and a promised voucher for the selected question.

Historical performance boundary

The advertised benchmark is not a current promise

The 2020 copy promoted a Spark 3.0 demonstration framed around 10x–100x query performance. This archive records that event claim without applying it to another workload, configuration, or current Databricks release.