Apache DataFu is a collection of libraries for working with large-scale data in Hadoop. The project was inspired by the need for stable, well-tested libraries for data mining and statistics.
It consists of three libraries:
- Apache DataFu Spark: a collection of utils and user-defined functions for Apache Spark
- Apache DataFu Pig: a collection of user-defined functions for Apache Pig
- Apache DataFu Hourglass: an incremental processing framework for Apache Hadoop in MapReduce
For more information please visit the website:
If you'd like to jump in and get started, check out the corresponding guides for each library:
- Apache DataFu Spark - Getting Started
- Apache DataFu Pig - Getting Started
- Apache DataFu Hourglass - Getting Started
- Introducing Datafu Spark
- A Look at PayPal's Contributions to DataFu
- DataFu's Hourglass: Incremental Data Processing in Hadoop
- DataFu 1.0
- DataFu: The WD-40 of Big Data
- Introducing DataFu
- A Brief Tour of DataFu
- Building Data Products at LinkedIn with DataFu
- Hourglass: a Library for Incremental Processing on Hadoop (IEEE BigData 2013)
Bugs and feature requests can be filed here. For other help please see the website.
If you are starting from a source release, then you'll want to verify the release is valid and bootstrap the build environment.
To verify that the archive has the correct SHA512 checksum, the following two commands can be run. These should produce the same output.
openssl sha512 < apache-datafu-sources-x.y.z.tgz cat apache-datafu-sources-x.y.z.tgz.sha512
To verify the archive against its signature, you can run:
gpg2 --verify apache-datafu-sources-x.y.z.tgz.asc
The command above will assume you are verifying apache-datafu-sources-x.y.z.tgz
and produce "Good signature" if the archive is valid.
To build DataFu from a source release, it is first necessary to download a gradle wrapper script. This bootstrapping process requires Gradle to be installed on the source machine. Gradle is available through most package managers or directly from its website. Once you have installed Gradle and have ensured that the gradle
is available in your path, you can bootstrap the wrapper with:
gradle -b bootstrap.gradle
After the bootstrap script has completed, you should find a gradlew
script in the root of the project. The regular gradlew instructions below should then be available.
When building from a source release, the version for all generated artifacts will be of the form x.y.z
. If you were to clone the git repo and build you would find -SNAPSHOT
appended to the version. This helps to distinguish official releases from those generated from the code repository for testing purposes.
To build DataFu from a git checkout or binary release, run:
./gradlew clean assemble
Each project's jars can be found under the corresponding sub directory. For example, the datafu-pig JAR can be found under datafu-pig/build/libs
. The artifact name will be of the form datafu-pig-x.y.z.jar
if this is a source release and datafu-pig-x.y.z-SNAPSHOT.jar
if this is being built from the code repository.
This command generates the eclipse project and classpath files:
./gradlew eclipse
To load the projects in Eclipse:
- Select "File -> Import", then "Existing Projects into Workspace"
- Choose the root of the repository as the root directory
- Check "Search for nested projects"
- Click Finish
To clean up the eclipse files:
./gradlew cleanEclipse
If you would like to use Intellij, please import the base datafu directory as a Gradle project. You may run into the following issues:
- Source/Tests directories not marked - please mark them manually.
- Tests not identified - add junit manually to the module using Open Module Settings->Libraries
- When running tests you get Error scala: Output path ... is shared between - change your output path as described here
To run all the tests:
./gradlew test
To run only one module's tests - for example, only the DataFu Pig tests:
./gradlew :datafu-pig:test
To run tests for a single class, use the tests
property. For example, to run only the QuantileTests:
./gradlew :datafu-pig:test --tests QuantileTests
The tests can also be run from within Eclipse. You'll need to install the TestNG plugin for Eclipse for DataFu Pig and Hourglass. See: http://testng.org/doc/download.html.
Potential issues and workaround:
- You may run out of heap when executing tests in Eclipse. To fix this adjust your heap settings for the TestNG plugin. Go to Eclipse->Preferences. Select TestNG->Run/Debug. Add "-Xmx1G" to the JVM args.
- You may get a "broken pipe" error when running tests. If so right click on the project, open the TestNG settings, and uncheck "Use project TestNG jar".