Skip to content

Alluxio, data orchestration for analytics and machine learning in the cloud

License

Notifications You must be signed in to change notification settings

Eugene-Mark/alluxio

 
 

Repository files navigation

logo

Slack Release Docker Pulls Documentation Twitter Follow License

What is Alluxio

Alluxio (formerly known as Tachyon) is a virtual distributed storage system. It bridges the gap between computation frameworks and storage systems, enabling computation applications to connect to numerous storage systems through a common interface. Read more about Alluxio Overview.

The Alluxio project originated from a research project called Tachyon at AMPLab, UC Berkeley, which was the data layer of the Berkeley Data Analytics Stack (BDAS). For more details, please refer to Haoyuan Li's PhD dissertation Alluxio: A Virtual Distributed File System.

Who Uses Alluxio

Alluxio is used in production to manage Petabytes of data in many leading companies, with the largest deployment exceeding 3,000 nodes. You can find more use cases at Powered by Alluxio or visit our first community conference (Data Orchestration Summit) to learn from other community members!

Who Owns and Manages Alluxio Project

Alluxio Open Source Foundation is the owner of Alluxio project. Project operation is done by Alluxio Project Management Committee (PMC). You can checkout more details in its structure and how to join Alluxio PMC here.

Community and Events

Please use the following to reach members of the community:

Download Alluxio

Binary download

Prebuilt binaries are available to download at https://www.alluxio.io/download .

Docker

Download and start an Alluxio master and a worker. More details can be found in documentation.

# Create a network for connecting Alluxio containers
$ docker network create alluxio_nw
# Create a volume for storing ufs data
$ docker volume create ufs
# Launch the Alluxio master
$ docker run -d --net=alluxio_nw \
    -p 19999:19999 \
    --name=alluxio-master \
    -v ufs:/opt/alluxio/underFSStorage \
    alluxio/alluxio master
# Launch the Alluxio worker
$ export ALLUXIO_WORKER_RAMDISK_SIZE=1G
$ docker run -d --net=alluxio_nw \
    --shm-size=${ALLUXIO_WORKER_RAMDISK_SIZE} \
    --name=alluxio-worker \
    -v ufs:/opt/alluxio/underFSStorage \
    -e ALLUXIO_JAVA_OPTS="-Dalluxio.worker.ramdisk.size=${ALLUXIO_WORKER_RAMDISK_SIZE} -Dalluxio.master.hostname=alluxio-master" \
    alluxio/alluxio worker

MacOS Homebrew

$ brew install alluxio

Quick Start

Please follow the Guide to Get Started to run a simple example with Alluxio.

Report a Bug

To report bugs, suggest improvements, or create new feature requests, please open a Github Issue. Our previous Alluxio JIRA system has been deprecated since December 2018.

Depend on Alluxio

Alluxio project provides several different client artifacts for external projects to depend on Alluxio client:

  • Artifact alluxio-core-client-fs provides a client for Alluxio Java file system API) to access all Alluxio-specific functionalities.
  • Artifact alluxio-core-client-hdfs provides a client for HDFS-Compatible file system API.
  • Artifact alluxio-shaded-client, available in Alluxio 2.0.1 or later, includes both above clients and all their dependencies but in a shaded form to prevent conflicts. This client jar is self-contained but the size is also much larger than the other two. This artifact is recommended generally for a project to use Alluxio client.

Here are examples to declare the dependecies on alluxio-shaded-client using Maven or SBT:

  • Apache Maven

    <dependency>
      <groupId>org.alluxio</groupId>
      <artifactId>alluxio-shaded-client</artifactId>
      <version>2.3.0</version>
    </dependency>
  • SBT

    libraryDependencies += "org.alluxio" % "alluxio-shaded-client" % "2.3.0"
    

Contributing

Contributions via GitHub pull requests are gladly accepted from their original author. Along with any pull requests, please state that the contribution is your original work and that you license the work to the project under the project's open source license. Whether or not you state this explicitly, by submitting any copyrighted material via pull request, email, or other means you agree to license the material under the project's open source license and warrant that you have the legal authority to do so. For a more detailed step-by-step guide, please read how to contribute to Alluxio. For new contributor, please take 2 new contributor tasks.

Useful Links

About

Alluxio, data orchestration for analytics and machine learning in the cloud

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Java 93.2%
  • TypeScript 2.2%
  • Go 1.9%
  • Shell 1.9%
  • C++ 0.2%
  • Mustache 0.1%
  • Other 0.5%