What's up with Spark(1/5) - Spark Architecture
Description
In this episode of What's up with___? Andrew Moll meets with Alejandro Guerrero Gonzalez and Joel Zambrano, Engineers on the HDInsight team, and learns all about the interworkings of Apache Spark. The team relies on your knowledge of traditional Hadoop, giving an overview of the major changes when developing on Spark. The video takes you through what is an RDD, how the Spark master and driver work together, and how we interact with the cluster once it's provisioned. You'll want to start here if you are just getting your feet wet with Spark! The spark documentation can be found here and HDInsight specific docs are here!
Share
Download
Download this episode
- MP3 (18.6 MB)
- Low Quality MP4 (49.7 MB)
- Mid Quality MP4 (205.2 MB)
- High Quality MP4 (317.9 MB)
More episodes in this series
What's up with Spark(3/5) - Jupyter and Zepplein Notebooks

Related episodes
What's up with Spark(4/5) - Spark Machine Learning

What's up with Spark(2/5) - SparkSQL
Create Spark Applications with the Azure Toolkit for IntelliJ

Debug HDInsight Spark Applications with Azure Toolkit for IntelliJ

Parsing Akamai logs using Azure HD Insight Spark Cluster.

What's up with Spark(5/5) - Spark Streaming

What's up with Spark(3/5) - Jupyter and Zepplein Notebooks

Introducing ML Services 9.3 in Azure HDInsight

4. Game Analytics with Azure

Live from Build 2016: A Little Chat About Big Data with Matt Winkler

The Discussion
-
AdrianPoplavsky "m&r is batch and spark is interactive". That was a pretty good kickoff summary.