Book Image

Spark Cookbook

By : Rishi Yadav
Book Image

Spark Cookbook

By: Rishi Yadav

Overview of this book

Related Content you might be interested in

Current Title:

Small book image
Spark Cookbook
Rishi Yadav
Jul, 2015 | 226 pages
No titles found
Table of Contents (19 chapters)
Spark Cookbook
Spark Cookbook
Credits
Credits
About the Author
About the Author
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Getting Started with Apache Spark
Getting Started with Apache Spark
Introduction
Installing Spark from binaries
Building the Spark source code with Maven
Launching Spark on Amazon EC2
Deploying on a cluster in standalone mode
Deploying on a cluster with Mesos
Deploying on a cluster with YARN
Using Tachyon as an off-heap storage layer
2
Developing Applications with Spark
Developing Applications with Spark
Introduction
Exploring the Spark shell
Developing Spark applications in Eclipse with Maven
Developing Spark applications in Eclipse with SBT
Developing a Spark application in IntelliJ IDEA with Maven
Developing a Spark application in IntelliJ IDEA with SBT
3
External Data Sources
External Data Sources
Introduction
Loading data from the local filesystem
Loading data from HDFS
Loading data from HDFS using a custom InputFormat
Loading data from Amazon S3
Loading data from Apache Cassandra
Loading data from relational databases
4
Spark SQL
Spark SQL
Introduction
Understanding the Catalyst optimizer
Creating HiveContext
Inferring schema using case classes
Programmatically specifying the schema
Loading and saving data using the Parquet format
Loading and saving data using the JSON format
Loading and saving data from relational databases
Loading and saving data from an arbitrary source
5
Spark Streaming
Spark Streaming
Introduction
Word count using Streaming
Streaming Twitter data
Streaming using Kafka
6
Getting Started with Machine Learning Using MLlib
Getting Started with Machine Learning Using MLlib
Introduction
Creating vectors
Creating a labeled point
Creating matrices
Calculating summary statistics
Calculating correlation
Doing hypothesis testing
Creating machine learning pipelines using ML
7
Supervised Learning with MLlib – Regression
Supervised Learning with MLlib – Regression
Introduction
Using linear regression
Understanding cost function
Doing linear regression with lasso
Doing ridge regression
8
Supervised Learning with MLlib – Classification
Supervised Learning with MLlib – Classification
Introduction
Doing classification using logistic regression
Doing binary classification using SVM
Doing classification using decision trees
Doing classification using Random Forests
Doing classification using Gradient Boosted Trees
Doing classification with Naïve Bayes
9
Unsupervised Learning with MLlib
Unsupervised Learning with MLlib
Introduction
Clustering using k-means
Dimensionality reduction with principal component analysis
Dimensionality reduction with singular value decomposition
10
Recommender Systems
Recommender Systems
Introduction
Collaborative filtering using explicit feedback
Collaborative filtering using implicit feedback
11
Graph Processing Using GraphX
Graph Processing Using GraphX
Introduction
Fundamental operations on graphs
Using PageRank
Finding connected components
Performing neighborhood aggregation
12
Optimizations and Performance Tuning
Optimizations and Performance Tuning
Introduction
Optimizing memory
Using compression to improve performance
Using serialization to improve performance
Optimizing garbage collection
Optimizing the level of parallelism
Understanding the future of optimization – project Tungsten
Index
Index

Understanding the future of optimization – project Tungsten

Project Tungsten, starting with Spark Version 1.4, is the initiative to bring Spark closer to bare metal. The goal of this project is to substantially improve the memory and CPU efficiency of the Spark applications and push the limits of underlying hardware.

In distributed systems, conventional wisdom has been to always optimize network I/O as that has been the most scarce and bottlenecked resource. This trend has changed in the last few years. Network bandwidth, in the last 5 years, has changed from 1 gigabit per second to 10 gigabit per second.

On similar lines, the disk bandwidth has increased from 50 MB/s to 500 MB/s and SSDs are being deployed more and more. CPU clock speed, on the other hand, was ~3 GHz 5 years back and is still the same. This has unseated the network and made CPU the new bottleneck in distributed processing.

Note

Another trend that has put more load on CPU performance is the new compressed data formats such as...

Previous Section
End of Chapter 12
Next Chapter