Lynn reviews a very complex dynamic ML streaming architecture using Hadoop Spark on AWS for just-in-time market pricing and advertising.
- [Instructor] As we're continuing our tour…of Hadoop architectures,…I wanted to end on an aspirational note.…A couple of things, first of all,…I'm obviously a big fan of Spark,…you can tell from the focus of this course.…And I see lots of customer demand,…and I see lots of development,…and exciting stuff happening in the Spark ecosystem,…and Spark is a key part of the majority…of the Modern Hadoop Pipelines that I see.…So the business situation is the company needs…to do dynamic price prediction of airline ticket prices…which in and of itself doesn't sound super exciting…but how they solved it really is quite intriguing,…using a Modern Hadoop architecture.…
What they did is, it's a variant of…what we actually looked at earlier in this course…where we ran a notebook…and we looked at using Machine Learning on Spark.…The results were kind of underwhelming.…Do you remember back there,…where it was like 29% prediction correct?…And then we used, basically, an implementation of a matrix…so that we could examine all different possibilities…
Author
Released
7/5/2017- Relate which file system is typically used with Hadoop.
- Explain the differences between Apache and commercial Hadoop distributions.
- Cite how to set up IDE - VS Code + Python extension.
- Relate the value of Databricks community edition.
- Compare YARN vs. Standalone.
- Review various streaming options.
- Recall how to select your programming language.
- Describe the Databricks environment.
Skill Level Intermediate
Duration
Views
Related Courses
-
Apache Spark Essential Training
with Ben Sullins1h 27m Intermediate
-
Introduction
-
Welcome53s
-
-
1. Hadoop Core Fundamentals
-
Modern Hadoop1m 53s
-
Hadoop libraries1m 23s
-
Run Hadoop job on GCP1m 52s
-
Databricks on AWS2m 32s
-
-
2. Setting Up a Hadoop Dev Environment
-
Load data into tables1m 51s
-
3. Hadoop Batch Processing
-
Processing options1m 2s
-
Resource coordinators1m 30s
-
Compare YARN vs. Standalone1m 30s
-
-
4. Fast Hadoop Options
-
Big data streaming1m 57s
-
Streaming options1m 10s
-
Apache Spark basics1m 46s
-
Spark use cases1m 2s
-
5. Spark Basics
-
Apache Spark libraries3m 24s
-
Spark shell1m 53s
-
-
6. Using Spark
-
Tour the notebook5m 29s
-
Import and export notebooks2m 56s
-
Calculate pi on Spark8m 19s
-
Import data2m 50s
-
Transformations and actions4m 43s
-
Caching and the DAG6m 49s
-
7. Spark Libraries
-
Spark SQL8m 34s
-
SparkR6m 11s
-
Spark ML: Preparing data4m 21s
-
Spark ML: Building the model3m 50s
-
MXNet or TensorFlow2m 30s
-
Spark with GraphX2m 12s
-
-
8. Spark Streaming
-
Spark streaming4m 21s
-
9. Hadoop Streaming
-
Pub/Sub on GCP3m 59s
-
Apache Kafka1m 26s
-
Kafka architecture1m 6s
-
Apache Storm1m 30s
-
Storm architecture1m 36s
-
-
10. Modern Hadoop Architectures
-
Conclusion
-
Next steps26s
-
- Mark as unwatched
- Mark all as unwatched
Are you sure you want to mark all the videos in this course as unwatched?
This will not affect your course history, your reports, or your certificates of completion for this course.
CancelTake notes with your new membership!
Type in the entry box, then click Enter to save your note.
1:30Press on any video thumbnail to jump immediately to the timecode shown.
Notes are saved with you account but can also be exported as plain text, MS Word, PDF, Google Doc, or Evernote.
Share this video
Embed this video
Video: Spark Streaming architecture for dynamic prediction