Scalding is a Scala library that makes it easy to write MapReduce jobs in Hadoop. Instead of forcing you to write raw map and reduce functions, Scalding allows you to write code that looks like natural Scala. It's similar to other MapReduce platforms like Pig, but offers a more powerful level of abstraction due to its built-in integration with Scala and the JVM.
Scalding is built on top of Cascading, a Java library that abstracts away much of the complexity of Hadoop.
Current version: 0.6.0
Hadoop is a distributed system for counting words. Here is how it's done in Scalding.
package com.twitter.scalding.examples
import com.twitter.scalding._
class WordCountJob(args : Args) extends Job(args) {
TextLine( args("input") )
.flatMap('line -> 'word) { line : String => tokenize(line) }
.groupBy('word) { _.size }
.write( Tsv( args("output") ) )
// Split a piece of text into individual words.
def tokenize(text : String) : Array[String] = {
// Lowercase each word and remove punctuation.
text.toLowerCase.replaceAll("[^a-zA-Z0-9\\s]", "").split("\\s+")
}
}
Notice that the tokenize
function, which is standard Scala, integrates naturally with the rest of the MapReduce job. This is a very powerful feature of Scalding. (Compare it to the use of UDFs in Pig.)
You can find more example code under examples/. If you're interested in comparing Scalding to other languages, see the Rosetta Code page, which contains several MapReduce tasks translated from other frameworks (e.g., Pig and Hadoop Streaming) into Scalding.
- Check out the Getting Started page on the wiki.
- Next, go through the runnable tutorials provided in the source.
- The API Reference contains general documentation, as well as many example Scalding snippets.
- The Scalding Wiki contains more useful information.
- Install sbt 0.11.3 (sorry, but the assembly plugin is sbt version dependent).
sbt update
(takes 2 minutes or more)sbt test
sbt assembly
(needed to make the jar used by the scald.rb script)
We use Travis CI to verify the build:
The current version is 0.6.0 and is available from maven central: org="com.twitter", artifact="scalding_2.8.1" or artifact="scalding_2.9.1".
Currently we are using the cascading-user mailing list for discussions: http://groups.google.com/group/cascading-user
In the remote possibility that there exist bugs in this code, please report them to: https://github.com/twitter/scalding/issues
Follow @Scalding on Twitter for updates.
- Avi Bryant http://twitter.com/avibryant
- Oscar Boykin http://twitter.com/posco
- Argyris Zymnis http://twitter.com/argyris
Thanks for assistance and contributions:
- Chris Wensel http://twitter.com/cwensel
- Ning Liang http://twitter.com/ningliang
- Dmitriy Ryaboy http://twitter.com/squarecog
- Dong Wang http://twitter.com/dongwang218
- Edwin Chen http://twitter.com/edchedch
- Sam Ritchie http://twitter.com/sritchie09
Copyright 2012 Twitter, Inc.
Licensed under the Apache License, Version 2.0: http://www.apache.org/licenses/LICENSE-2.0