
As we welcome the spring season, we're thrilled to share several exciting developments with you!
🚀 Announcing Alluxio Enterprise AI 3.5
Our latest Alluxio AI release delivers powerful new features and capabilities to accelerate your AI workflows, including:
- New cache mode that optimizes write performance for model checkpoint files
- Three new features to optimize your cache efficiency, including TTL & priority-based cache eviction policies
- New PyTorch, PyArrow, and Ray integrations built into the Alluxio Python SDK
- Performance & security enhancements to the Alluxio S3 API
Learn more about what’s new in Alluxio AI:
- Product Blog: New Features in Alluxio Enterprise AI 3.5
- Efficiently Connected: Accelerating AI Workflows with Efficient Data Management
- Enterprise AI World: Alluxio Enterprise AI 3.5 Heralds Performance Efficiencies for AI Workloads
- Datanami: Alluxio Enhances Enterprise AI with Version 3.5 for Faster Model Training
📖 New Customer Success Stories
Case Study: RedNote Accelerates Model Training & Distribution with Alluxio

RedNote, a popular and rapidly growing e-commerce and social media platform in Asia adopted Alluxio to meet its nightly model training and distribution SLAs and lower model distribution costs. With Alluxio, RedNote efficiently trains their Search and Recommendation Models, efficiently distributes models to production for inference serving - all while lowering the cost of their multi-cloud architecture, making Alluxio a vital part of their ML infrastructure.
Case Study: Global Top 10 E-Commerce Giant Accelerates Training of Search & Recommendation AI Model with Alluxio

A publicly traded e-commerce company with over 50,000 employees and $20+ billion in annual revenue accelerates training of their search & recommendation AI models with Alluxio. Since deploying Alluxio AI Enterprise, the company’s AI/ML training workloads have become faster and more stable.
Have an Alluxio success story? We'd love to feature you! Contact us today.
🎥 Recent Tech Talks
- AI/ML Infra Meetup: Balancing Cost, Performance, and Scale - Running GPU/CPU Workloads in the Cloud - Bin Fan
- AI/ML Infra Meetup: Three Developments in AI Infra - Robert Nishihara
- AI/ML Infra Meetup: A Faster and More Cost Efficient LLM Inference Stack - Junchen Jiang
- Alluxio @ AI_dev Japan: Exploring Distributed Caching for Faster GPU Training with NVMe, GDS, and RDMA - Hope Wang & Bin Fan
- Alluxio @ PrestoCon: A Case Study in API Cost of Running Presto in the Cloud at Exabyte Scale - Tom Luckenbach
🗓️ Upcoming Events

March 6th, 4:30 PM – 8:00 PM PST: AI/ML Infra Meetup at Uber Seattle (Downtown Seattle & Virtual)
Tech Leads from Uber’s AI Platform Team, Snap’s AI Platform Team, and Alluxio will share their insights and real-world examples about LLM training, fine-tuning, & deployment, as well as designing scalable architectures, GPU optimization, building recommendation apps, and more.

March 20th, 6:00 PM – 8:00 PM PST: Accelerating AI Applications Development @ Sunnyvale, CA: In this meetup, Bin Fan (VP of Open Source @ Alluxio) will discuss a distributed caching architecture designed specifically for AI/ML workloads and will explore lessons learned building Alluxio.
Interested in speaking at one of our events? Submit your proposal here: https://forms.gle/iJX9GTMaAVQdzKc28.
Thanks for reading. We're excited to see you at one of the events or for the next newsletter in March!