YoVDO

Degrading Performance - Understanding and Solving Small Files Syndrome

Offered By: Databricks via YouTube

Tags

Apache Spark Courses Data Processing Courses Cloud Storage Courses Parquet Courses

Course Description

Overview

Explore the impact of small files on data pipeline performance in this 27-minute video from Databricks. Learn about the challenges posed by small files, particularly in Parquet format, across real-time, near-real-time, and batch processing jobs. Understand how cloud storage interacts with Apache Spark, delve into Parquet and Storage API calls, and discover why small files can degrade system performance. Gain insights into leveraging DeltaLake for a more efficient solution. Follow along with a comprehensive agenda covering Spark query lifecycle abstraction, Azure Data Lake Gen 2, file hierarchy detection, and a practical demonstration.

Syllabus

Intro
Agenda
Spark query lifecycle abstraction
Azure Data Lake Gen 2
Detecting
File Hierarchy
Demo
Demo Summary
Outro


Taught by

Databricks

Related Courses

Python for Data Science Tips, Tricks, & Techniques
LinkedIn Learning
Sound Data Engineering in Rust - From Bits to DataFrames
Databricks via YouTube
Recent Parquet Improvements in Apache Spark - Vectorized Complex Types and Column Index Support
Databricks via YouTube
Optimizing Spark SQL Jobs with Parallel and Asynchronous IO
Databricks via YouTube
The Apache Spark File Format Ecosystem - Optimizing Storage for Performance
Databricks via YouTube