Keynote: Building a Fusion Engine with Ray
summary
Ant Group builds a fusion engine on Ray to unify AI, graph, streaming and OLAP workloads and claims production success at scale. The stack runs on Ray with a Java API, a fault-tolerant GCS and a worker-to-worker transport; deployed frameworks include a dynamic real-time graph engine, a complex online decision engine and online ML. Systems serve Double 11 traffic for 1.3 billion users, handle real-time anti-fraud and report six hot queries within one second; Ant contributes ~22% of Ray’s codebase.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:50Computing Evolution
- 4:14Challenges
- 5:17Fusion Engine
- 6:06Field Paradigms
- 7:45Overall Architecture
- 8:19Features
- 9:17Outro
related talks
14mFusion Computing at Ant Group
15mRay: A Distributed Compute Engine for AI
11mKeynote Q&A
03mRay: A Distributed Execution Framework for Emerging AI Applications
23mImproving Ray for Large-scale Applications
26mData Processing on Ray
28mScaling Multimodal Data and Reinforcement Learning with Ray
31mBuilding a Virtual Biology Platform on Ray