Ray: A Distributed Execution Framework for Emerging AI Applications
summary
A new execution framework called Ray aims to replace Spark for low-latency, live-data AI workloads by combining the familiar guarantees of Hadoop/Spark with a just-in-time dataflow execution model. It supports heterogeneous task graphs across neural nets, planning, search, and simulation, offers heterogeneous task scheduling that adapts as algorithms change, and targets low-latency feedback loops for real-time decision-making. Presented from Berkeley’s Rise Lab, Ray is implemented as a distributed runtime for mixed batch and streaming AI pipelines.
AI-written summary · about
related talks
15mRay: A Distributed Compute Engine for AI
42mRay: A distributed system for emerging AI applications
22mRay, a Unified Distributed Framework for the Modern AI Stack
26mUsing Ray On Large-scale Applications at Ant Group
24mRay: A System for High-performance, Distributed ML Applications
26mRayDP: Build Large-scale End-to-end Data Analytics and AI Pipelines Using Spark and Ray
29mRay:A Distributed Computing Platform for Reinforcement Learning
26mRay: A System for High-performance, Distributed Python Applications