Скачать или смотреть How Bytedance Builds Large-Scale Data Processing Pipelines for Multimodal Models with Ray | RS 24

ByteDance's quest to create advanced video generation models demands processing vast amounts of high-quality video data. In this session, Xiaohong Dong, Zhibei Ma, and Liguang Xie from ByteDance unveil how they harnessed Ray to construct a robust, scalable data pipeline for their ambitious project.

The team delves into the challenges of building a data processing system capable of handling massive video datasets. They showcase how Ray's ecosystem, particularly Ray Core, Ray Data, and Ray Serve, provided crucial solutions for dynamic scaling and orchestration of heterogeneous resources. The presenters also share valuable insights on managing Ray infrastructure and best practices gleaned from their experience, offering attendees a blueprint for leveraging Ray in large-scale, multimodal AI projects.

--

Interested in more?
Watch the full Day 1 Keynote:    • Ray Summit 2024 Keynote Day 1 | Where Buil...  
Watch the full Day 2 Keynote    • Ray Summit 2024 Keynote Day 2 | Where Buil...  

--

🔗 Connect with us:
Subscribe to our YouTube channel:    / @anyscale  
Twitter: https://x.com/anyscalecompute
LinkedIn:   / joinanyscale  
Website: https://www.anyscale.com