Alluxio Open Source (formerly known as Tachyon) is a Distributed Caching Platform for large-scale data. It bridges the gap between computation frameworks and storage systems, enabling computation applications to connect to numerous storage systems through a common interface.
This GitHub repository is the open-source edition of Alluxio, which is purpose-built for analytics workloads. It provides caching and acceleration for structured data analytics and is widely adopted with data-intensive computation engines such as Presto, Spark, and Trino.
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.
The open-source edition of Alluxio is purpose-built for analytics workloads. It accelerates structured data analytics and scales to manage up to 100 million files. This Alluxio Open Source Edition is available for free without support and is recommended for testing, development, and small-scale production environments.
For AI and machine learning workloads, including model training, distribution and inference at scale, Alluxio Enterprise Edition provides a fundamentally different architecture with a decentralized metadata service. As a result, it scales horizontally to support tens of billions of files and delivers higher performance along with FUSE-based POSIX integration for compatibility with popular AI frameworks, including PyTorch, TensorFlow, and Ray.
To learn more about Alluxio Enterprise Edition, visit: https://www.alluxio.io/enterprise-ai. For more details about the differences between editions, visit https://www.alluxio.io/editions.
Alluxio Open Source Foundation is the owner of Alluxio project. Project operation is done by Alluxio Project Management Committee (PMC). You can check out more details on its structure and how to join Alluxio PMC here.
Please use the following to reach members of the community:
Prebuilt binaries are available to download at https://www.alluxio.io/download .
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
$ brew install alluxio
Please follow the Guide to Get Started to run a simple example with Alluxio.
To report bugs, suggest improvements, or create new feature requests, please open a Github Issue. If you are not sure whether you run into bugs or simply have general questions with respect to Alluxio, post your questions on Alluxio Slack channel.
Alluxio project provides several different client artifacts for external projects to depend on Alluxio client:
alluxio-shaded-client
is recommended generally for a project to use Alluxio client. The jar of this artifact is self-contained (including all dependencies in a shaded form to prevent dependency conflicts), and thus larger than the following two artifacts.alluxio-core-client-fs
provides Alluxio Java file system API) to access all Alluxio-specific functionalities. This artifact is included in alluxio-shaded-client
.alluxio-core-client-hdfs
provides HDFS-Compatible file system API. This artifact is included in alluxio-shaded-client
.Here are examples to declare the dependecies on alluxio-shaded-client
using Maven:
<dependency>
<groupId>org.alluxio</groupId>
<artifactId>alluxio-shaded-client</artifactId>
<version>2.6.0</version>
</dependency>
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 two new contributor tasks.