A collaboration of Alibaba, Alluxio, and Nanjing University in tackling the problems of Deep Learning model training in the cloud. Our goal was to reduce the cost and complexity of data access for Deep Learning training in a hybrid environment, which resulted in over 40% reduction in training time and cost.
welcome to the alluxio community
Founder Haoyuan Li was a Ph.D. student at UC Berkeley AMPLab when he built the beginnings of Alluxio (originally called Tachyon) with the mission to orchestrate data for all data driven applications. Today, in addition to running critical workloads for thousands of users across the world, Alluxio has a vibrant community that has made countless contributions to the open source project.
Welcome to Alluxio Community! We would invite you to read about, try out, use, and contribute to Alluxio, as well as to share your experience, feedback, suggestions and ideas!
Slack
Ask questions, get answers.
GitHub
Become a contributor.
MAILING LIST
Join the Google group.
Join our channel.
For many latency-sensitive SQL workloads, Presto is often bound by retrieving distant data. In this talk, Rohit Jain, James Sun from Facebook and Bin … Continued
This article shares a collaborative initiative between two open-source software, Intel Analytics Zoo and Alluxio, in accelerating deep learning on Apache Spark in the cloud environment. We propose a new architecture for training deep learning models in a hybrid cloud environment and demonstrate its performance benefits through standardized benchmarks
Join an Alluxio community event Near You
Don’t see an event in your area? Want to start a local Alluxio meetup? Drop us a note!
ACADEMIC PAPERS
join the community
4,000+
Stars
1,000+
Contributors
1 Million+
Downloads
apache 2.0 licensed
Alluxio Contributors
The Alluxio Open Source Contributors and Project Management Committee members come from a diverse and experienced background. The project members includes committers with decades of experience from Tencent, Google, Palantir, UC Berkeley, Carnegie Mellon, IBM, Intel and JD.com.
