Dear Alluxio Community,
Welcome to our newsletter! We've packed this edition with valuable insights, upcoming events, and on-demand content to keep you at the forefront of AI.
💡 Customer Spotlight
Dyna Robotics Turbocharges Foundation Model Training with Alluxio
A huge congratulations to one of our newest customers, Dyna Robotics, on successfully closing their $120M Series A to advance their work in the cutting-edge robotics market!
To improve their foundation model training performance, Dyna Robotics leverages Alluxio as a distributed caching and data access layer. With Alluxio, Dyna Robotics has accelerated model iteration cycles, which directly translates to a competitive advantage in the rapidly evolving robotics market.
Blackout Power Trading Selects Alluxio to Scale from 5,000 to 100,000+ ML Models
*For more details, check the blog
Blackout Power Trading, a private capital commodity trading fund specializing in North American power markets, leverages Alluxio's low-latency distributed caching platform to achieve multi-join double-digit millisecond latency offline feature store performance while maintaining the cost and durability benefits of Amazon S3 for persistent data storage.
📖 Good Reads
Dive into these inspiring articles and white papers from the Alluxio team and beyond
Alluxio + S3: A Tiered Architecture for Latency-Critical, Semantically-Rich Workloads
Unlock the full potential of your data with a tiered architecture that combines the cost-efficiency of Amazon S3 with the high-performance caching of Alluxio. Our latest blog post details how this powerful combination can dramatically reduce latency for your most demanding AI and analytics workloads, without the need for data migration. Benchmark results and real-world examples included!
*For more information, read the press release
Alluxio is proud to announce a remarkable second quarter, marked by the launch of Enterprise AI 3.7, which delivers groundbreaking sub-millisecond latency for AI workloads — up to 45 times faster than S3 Standard. Complemented by exceptional customer momentum with over 50% and industry-leading results in the MLPerf Storage v2.0 benchmark, these milestones further validate Alluxio’s leadership in AI infrastructure.
🗓️ Upcoming Events
Mark your calendar for our exciting upcoming events!
🎤 Featured Speakers include:
- Pratik Mishra, Senior Member of Technical Staff - Research @ AMD
- Greg Lindstrom, VP ML Trading @ Blackout Power Trading
- Bin Fan, VP of Technology @ Alluxio
Join us in person for the next AI Infra Meetup in San Francisco, CA or tune in livestreaming on Zoom! Hear from an impressive lineup of speakers as they share their insights on designing and implementing AI infrastructure at scale. Topics will cover building low-latency offline feature store for ML trading, accelerating AI using caching, storage for AI, and more!
🎥 Recent Tech Talks
Missed a session or want to review your favorite talk? Check out these on-demand videos and slides now
- Building AI Applications on Zoom
>> Watch this demo as Ojus Save, DevRel at Zoom, shows you how to build an AI agent that receives transcript data from RTMS and then decides if it has to create action items based on the transcripts that are received. - Accelerating the Data Path to the GPU for AI and Beyond
>> In this talk, Sandeep Joshi, Senior Manager at NVIDIA, shares how to accelerate the data access between GPU and storage for AI. Sandeep will dive into two options: CPU- initiated GPUDirect Storage and GPU-initiated SCADA. - Beyond S3's Basics: Architecting for AI-Native Data Access
>> Bin Fan, VP of Technology at Alluxio, introduces how Alluxio, a software layer transparently sits between application and S3 (or other object stores), provides sub-ms time to first byte (TTFB) solution, with up to 45x lower latency. - LLM Agents and Implementation Challenges
>> In this talk, Pritish Udgata from Adobe provides a comprehensive overview of implementation challenges and solutions for LLM agents.
Thanks for reading. We're excited to see you at one of the events or for the next newsletter!
.png)
