Learn how to write, tune, and port SQL queries and other statements for a Big Data environment, using Impala--the massively parallel processing SQL query engine for Apache Hadoop. The best practices in this practical guide help you design database schemas that not only interoperate with other Hadoop components, and are convenient for administers to manage and monitor, but also accommodate future expansion in data size and evolution of software capabilities.
Ideal for database developers and business analysts, Getting Started with Impala includes advice from Cloudera’s development team, as well as insights from its consulting engagements with customers.
Learn how Impala integrates with a wide range of Hadoop components
Attain high performance and scalability for huge data sets on production clusters
Explore common developer tasks, such as porting code to Impala and optimizing performance
Use tutorials for working with billion-row tables, date- and time-based values, and other techniques
Learn how to transition from rigid schemas to a flexible model that evolves as needs change
Take a deep dive into joins and the roles of statistics
Chapter 1Why Impala?
Impala’s Place in the Big Data Ecosystem
Flexibility for Your Big Data Workflow
Exploratory Business Intelligence
Chapter 2Getting Up and Running with Impala
Connecting to Impala
Your First Impala Queries
Chapter 3Impala for the Database Developer
The SQL Language
Big Data Considerations
How Impala Is Like a Data Warehouse
Physical and Logical Data Layouts
Normalized and Denormalized Data
Chapter 4Common Developer Tasks for Impala
Getting Data into an Impala Table
Porting SQL Code to Impala
Using Impala from a JDBC or ODBC Application
Using Impala with a Scripting Language
Optimizing Impala Performance
Writing User-Defined Functions
Collaborating with Your Administrators
Chapter 5Tutorials and Deep Dives
Tutorial: From Unix Data File to Impala Table
Tutorial: Queries Without a Table
Tutorial: The Journey of a Billion Rows
Deep Dive: Joins and the Role of Statistics
Anti-Pattern: A Million Little Pieces
Tutorial: Across the Fourth Dimension
Tutorial: Verbose and Quiet impala-shell Output
Tutorial: When Schemas Evolve
Tutorial: Levels of Abstraction
Getting Started with Impala
Safari Books Online
| ISBN 10:
| ISBN 10:
John Russell is a software developer and technical writer, and he's currently the documentation lead for the Cloudera Impala project. He has a broad range of database and SQL experience from previous roles on industry-leading teams. For DB2, he designed and coded the very first Information Center. For Oracle Database, he documented application development subjects and designed and coded the Project Tahiti doc search engine. For MySQL, he documented the InnoDB storage engine. Originally from Newfoundland, Canada, John now resides in Berkeley, California.
The animal on the cover of Getting Started with Impala is an impala (Aepyceros melampus. This is a medium-sized antelope native to the continent of Africa, which lives in savanna and brushland habitats. They are herbivores, and spend their days in herds grazing on grass and other vegetation.Male impalas are the only sex with horns, which curve in an S shape rather like the lyre (an instrument of ancient Greece). These horns are used to fight other males during breeding season and to protect territory. Males are also noticeably larger and heavier than females: 30–36 inches tall at the shoulder and 120–160 pounds, versus the females' average of 28–33 inches and 90–120 pounds. All impalas have brown coats, black stripes on their rear legs, and a white tail with a black stripe running down its length.Impalas live in three kinds of social groups during the rainy season when mating occurs: bachelor herds of nonterritorial and young males, herds of territorial males with breeding females, and herds of females with their young. Young male impalas remain in the latter kind of group until they are about four years old, when they reach sexual maturity and establish their own territory. In the dry season, these groups often intermingle.Impalas are very fast runners, which is their main defense against natural predators such as lions, cheetahs, crocodiles, and hyenas. They can also leap nearly 10 feet in the air, which serves as a tactic to confuse or startle predators.The cover image is from Wood's Animate Creation.
Comments about oreilly Getting Started with Impala:
Impala is a recent, but very valuable addition to the Hadoop ecosystem. I must say (after reading the book) Cloudera made a big step forward in the right direction.
The rational behind bringing Impala to life is the proliferation of SQL. SQL as a language has many flavours, but in one form or another is already known to data practitioners coming to Hadoop from various platforms and DBMS. Impala implements a subset of ANSI-92 SQL specification, regardless, even the subset is powerful enough to make a developer productive. In my opinion, since SQL it is based on algebra and sets, and because HDFS (Hadoop) is just able to expose datasets Impala is the right choice for MDL and DDL even for the Big Data projects.
At 110 pages the book is not terribly long, but bear in mind Impala as a product is still under active development, as a bonus, the author has a close relationship with the product working at Cloudera, this is a big plus resulting in top professional content. John structured the book so it is basically divided into two parts: 1st and the largest is on Impala implementation and its role in data analysis and processing, the 2nd part covers most commonly used tasks, pitfalls or simply advice and techniques.
What I did not find is more on how to use it with Hive, Scoop, HBase and Pig, I will take a star out of my rating for this.
Let me reiterate, the book covers the Cloudera's Hadoop Impala distribution, if you are using a different distribution, Impala is not part of it.
Like I said, I am giving this book a 4 out of 5 stars. Good work John!
Disclaimer: the book was provided to me for free as part of O'Reilly's blogger reviewer programme.
Bottom Line Yes, I would recommend this to a friend