Writing Beautiful Apache Spark Code
Writing Beautiful Apache Spark Code
Processing massive datasets with ease
About the Book
This book teaches Spark fundamentals and shows you how to build production grade libraries and applications. It took years for the Spark community to develop the best practices outlined in this book. This book will fast track your Spark learning journey and put you on the path to mastery.
About the Author
Table of Contents
-
Introduction
- Typical painful workflow
- Productionalizing advanced analytics models is hard
- Why Scala?
- Who should read this book?
- Is this book for data engineers or data scientists?
- Beautiful Spark philosophy
- DataFrames vs. RDDs
- Spark streaming
- Machine learning
- The “coalesce test” for evaluating learning resources
- Will we cover the entire Spark SQL API?
- How this book is organized
- Spark programming levels
- Note about Spark versions
-
Running Spark Locally
- Starting the console
- Running Scala code in the console
- Accessing the SparkSession in the console
- Console commands
-
Databricks Community
- Creating a notebook and cluster
- Running some code
- Next steps
-
Introduction to DataFrames
- Creating DataFrames
- Adding columns
- Filtering rows
- More on schemas
- Creating DataFrames with createDataFrame()
- Next Steps
-
Working with CSV files
- Reading a CSV file into a DataFrame
- Writing a DataFrame to disk
- Reading CSV files in Databricks Notebooks
-
Just Enough Scala for Spark Programmers
- Scala function basics
- Currying functions
- object
- trait
-
package - Implicit classes
- Next steps
-
Column Methods
- A simple example
- Instantiating Column objects
- gt
- substr
- + operator
- lit
- isNull
- isNotNull
- when / otherwise
- Next steps
-
Introduction to Spark SQL functions
- High level review
- lit() function
- when() and otherwise() functions
- Writing your own SQL function
- Next steps
-
User Defined Functions (UDFs)
- Simple UDF example
- Using Column Functions
- Conclusion
-
Chaining Custom DataFrame Transformations in Spark
- Dataset Transform Method
- Transform Method with Arguments
-
Whitespace data munging with Spark
- trim(), ltrim(), and rtrim()
- singleSpace()
- removeAllWhitespace()
- Conclusion
-
Defining DataFrame Schemas with StructField and StructType
- Defining a schema to create a DataFrame
-
StructField -
Defining schemas with the
::operator -
Defining schemas with the
add()method - Common errors
-
LongType - Next steps
-
Different approaches to manually create Spark DataFrames
- toDF
- createDataFrame
- createDF
- How we’ll create DataFrames in this book
-
Dealing with null in Spark
- What is null?
- Spark uses null by default sometimes
- nullable Columns
- Native Spark code
- Scala null Conventions
- User Defined Functions
- Spark Rules for Dealing with null
-
Using JAR Files Locally
- Starting the console with a JAR file
- Adding JAR file to an existing console session
- Attaching JARs to Databricks clusters
- Review
-
Working with Spark ArrayType columns
- Scala collections
- Splitting a string into an ArrayType column
-
Directly creating an
ArrayTypecolumn -
array_contains -
explode -
collect_list - Single column array functions
- Generic single column array functions
- Multiple column array functions
- Split array column into multiple columns
- Closing thoughts
-
Working with Spark MapType Columns
- Scala maps
- Creating MapType columns
- Fetching values from maps with element_at()
- Appending MapType columns
- Creating MapType columns from two ArrayType columns
- Converting Arrays to Maps with Scala
- Merging maps with map_concat()
- Using StructType columns instead of MapType columns
- Writing MapType columns to disk
- Conclusion
-
Adding StructType columns to DataFrames
- StructType overview
- Appending StructType columns
- Using StructTypes to eliminate order dependencies
- Order dependencies can be a big problem in large Spark codebases
-
Working with dates and times
- Creating DateType columns
- year(), month(), dayofmonth()
- minute(), second()
- datediff()
- date_add()
- Next steps
-
Performing operations on multiple columns with foldLeft
- foldLeft review in Scala
- Eliminating whitespace from multiple columns
- snake_case all columns in a DataFrame
- Wrapping foldLeft operations in custom transformations
- Next steps
-
Equality Operators
- =
-
Introduction to Spark Broadcast Joins
- Conceptual overview
- Simple example
- Analyzing physical plans of joins
-
Eliminating the duplicate
citycolumn -
Diving deeper into
explain() - Next steps
-
Partitioning Data in Memory
- Intro to partitions
- coalesce
- Increasing partitions
- repartition
- Differences between coalesce and repartition
- Real World Example
-
Partitioning on Disk with partitionBy
- Memory partitioning vs. disk partitioning
- Simple example
- partitionBy with repartition(5)
- partitionBy with repartition(1)
- Partitioning datasets with a max number of files per partition
- Partitioning dataset with max rows per file
- Partitioning dataset with max rows per file pre Spark 2.2
- Small file problem
- Conclusion
-
Fast Filtering with Spark PartitionFilters and PushedFilters
- Normal DataFrame filter
-
partitionBy() - PartitionFilters
- PushedFilters
- Partitioning in memory vs. partitioning on disk
- Disk partitioning with skewed columns
- Next steps
-
Scala Text Editing
- Syntax highlighting
- Import reminders
- Import hints
- Argument type checking
- Flagging unnecessary imports
- When to use text editors and Databricks notebooks?
-
Structuring Spark Projects
- Project name
- Package naming convention
- Typical library structure
- Applications
-
Introduction to SBT
- Sample code
- Running SBT commands
- build.sbt
-
libraryDependencies - sbt test
- sbt doc
- sbt console
- sbt package / sbt assembly
- sbt clean
- Next steps
-
Managing the SparkSession, The DataFrame Entry Point
- Accessing the SparkSession
- Example of using the SparkSession
- Creating a DataFrame
- Reading a DataFrame
- Creating a SparkSession
- Reusing the SparkSession in the test suite
- SparkContext
- Conclusion
-
Testing Spark Applications
- Hello World Example
- Testing a User Defined Function
- A Real Test
- How Testing Improves Your Codebase
- Running a Single Test File
-
Environment Specific Config in Spark Scala Projects
- Basic use case
- Environment specific code anitpattern
- Overriding config
-
Setting the
PROJECT_ENVvariable for test runs - Other implementations
- Next steps
-
Building Spark JAR Files with SBT
- JAR File Basics
- Building a Thin JAR File
- Building a Fat JAR File
- Next Steps
-
Shading Dependencies in Spark Projects with SBT
- When shading is useful
-
How to shade the
spark-dariadependency - Conclusion
-
Dependency Injection with Spark
- Code with a dependency
- Injecting a path
- Injecting an entire DataFrame
- Conclusion
-
Broadcasting Maps
- Simple example
- Refactored code
- Building Maps from data files
- Conclusion
-
Validating Spark DataFrame Schemas
- Custom Transformations Refresher
- A Custom Transformation Making a Bad Assumption
- Column Presence Validation
- Full Schema Validation
- Documenting DataFrame Assumptions is Especially Important for Chained DataFrame Transformations
- Conclusion
Authors have earned$9,675,609writing, publishing and selling on Leanpub, earning 80% royalties while saving up to 25 million pounds of CO2 and up to 46,000 trees.
Learn more about writing on Leanpub
The Leanpub 45-day 100% Happiness Guarantee
Within 45 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.
See full terms
Free Updates. DRM Free.
If you buy a Leanpub book, you get free updates for as long as the author updates the book! Many authors use Leanpub to publish their books in-progress, while they are writing them. All readers get free updates, regardless of when they bought the book or how much they paid (including free).
Most Leanpub books are available in PDF (for computers), EPUB (for phones and tablets) and MOBI (for Kindle). The formats that a book includes are shown at the top right corner of this page.
Finally, Leanpub books don't have any DRM copy-protection nonsense, so you can easily read them on any supported device.
Learn more about Leanpub's ebook formats and where to read them
Top Books
- #1
El Manual del Manager
Keyvan Akbary, Félix López, and Álvaro Salazar¿Has deseado alguna vez el haber tenido una buena introducción al rol del Engineering Manager? En este libro aprenderás lo necesario para ejercer el rol de una manera efectiva: Expectativas y Responsabilidades del Rol, 1-1s, Ayudar a Crecer, Objetivos, Planes de Carrera, Cultura, Feedback, Contratación, Cultura de Producto y mucho más.
- #2
Functional Design and Architecture
Alexander GraninSoftware Design in Functional Programming, Design Patterns and Practices, Methodologies and Application Architectures. How to build real software in Haskell with less efforts and low risks. The first complete source of knowledge.
- #3
Ansible for Kubernetes
Jeff GeerlingAnsible is a powerful infrastructure automation tool. Kubernetes is a powerful application deployment platform. Learn how to use these tools to automate massively-scalable, highly-available infrastructure.
- #4
Ansible for DevOps
Jeff GeerlingAnsible is a simple, but powerful, server and configuration management tool. Learn to use Ansible effectively, whether you manage one server—or thousands.
- #5
Practical FP in Scala: A hands-on approach
Gabriel VolpeA practical book aimed for those familiar with functional programming in Scala who are yet not confident about architecting an application from scratch.
Together, we will develop a purely functional application using the best libraries in the Cats ecosystem, while learning about design patterns and best practices.
- #6
Tame your Work Flow
Steve Tendon and Daniel DoironDo you need a high performance enterprise governance approach improving management, execution and delivery while dealing with multiple projects/products, events, stakeholders and teams? Giving you better bottom line results, faster time to market, less work, better predictability, happier employees, and delighted clients? Then learn about TameFlow!
- #7
C++ Best Practices
Jason TurnerLevel up your C++, get the tools working for you, eliminate common problems, and move on to more exciting things!
- #8
Cloud Strategy
Gregor Hohpe“Strategy is the difference between making a wish and making it come true.” A successful migration to the cloud shouldn’t be driven by wishes, but guided by a sound strategy, frameworks, and decision models. This book tells you how—without becoming superficial nor getting lost in technology and product details.
- #9
Machine Learning Engineering
Andriy Burkov"If you intend to use machine learning to solve business problems at scale, I'm delighted you got your hands on this book."
—Cassie Kozyrkov, Chief Decision Scientist at Google
"Foundational work about the reality of building machine learning models in production."
—Karolis Urbonas, Head of Machine Learning and Science at Amazon
- #10
Composing Software
Eric ElliottAll software design is composition: the act of breaking complex problems down into smaller problems and composing those solutions. Most developers have a limited understanding of compositional techniques. It's time for that to change.
Top Bundles
- #1
Quality Software
11 Books
The Quality Software Bundle is for managers, would-be managers, and any of us who find themselves being managed and confused. This comprehensive bundle covers the entire span of software development approaches, from hacking through waterfall, cascade, prototyping, Iterative enhancement, reusable code, off-the-shelf, to Agile teams. The bundle... - #2
The Node.js Bundle
3 Books
This bundle combines three bestselling Leanpub Node.js books into a package that gives you everything you need to get started with developing Node.js applications at an unbeatable price. - #3
The Tester's Library
8 Books
The Tester's Library consists of eight five-star books that every software tester should read and re-read. As bound books, this collection would cost over $200. Even as e-books, their price would exceed $80, but in this bundle, their cost is only $49.99. Here are the books, and why they should be in your library: Perfect Software and Other... - #4
agile
11 Books
In this bundle, you will find 10 different agile books. They are about different aspects of being agile. - finding a job - doing coding dojo's - Retrospectives - Personal kanban - a non-typical coaching book and even a book that gives you an insight in the lives of some agile people. - #5
WTFlop 6M + HU - Beta Bundle
6 Books
- #6
Marionette.js A to Z
3 Books
Marionette.js has become a popular Backbone-based javascript library, because it helps you avoid boiler-plate code and focus on what makes your app special. But how do you get started?Whether you want to finally get started with dynamic client-side programming (like one reader did), or are an experienced front-end dev who has inherited a... - #7
Complete Scala Bundle
3 Books
Scala is a general-purpose programming language and it's getting extremely popular these days. Some say that learning Scala could be a challenging task. My experience, however, suggests that this is actually a myth that has very little to do with reality. With the right approach, learning Scala can be easy, fun and rewarding.The first book from... - #8
Build A Better Backbone App
3 Books
The best way to learn new development skills is through experience, but that takes time you don't have.Get the best of both worlds with this bundle: you'll learn how to produce modern web applications by learning from experienced developers like Derick Bailey and David Sulc. BackboneJS is one of the favorite tools on the web today, but it... - #9
People Skills—Soft but Difficult
7 Books
Perhaps you've been told that "lack of people skills" has been holding you back. No wonder: you may have had hundreds of hours of technical training, but little or no "people skills" guidance.You've heard it said that people skills are "soft," whereas technical skills are "hard." For you, though, technical skills are "easy," but people skills... - #10
SurviveJS - Webpack + React
2 Books
Get both SurviveJS - Webpack and SurviveJS - React for a single price!
