Volcano - Cloud Native Batch System for AI, BigData and HPC
CNCF [Cloud Native Computing Foundation] via YouTube
Overview
Explore a comprehensive conference talk on Volcano, the first batch computing project in CNCF, designed for AI, Big Data, and High-Performance Computing (HPC). Delve into Volcano's key features, including job-based fair-share, priority, preemption, reclaim, and queue management capabilities. Learn about its integration with computing ecosystems like spark-operator, flink-operator, kubeflow, and Cromwell. Discover the latest developments, such as co-location, elastic training, vGPU, throughput optimization, and multi-cluster scheduling for HPC users. Gain insights into Volcano's global adoption, best practices, and its role in accelerating AI training, serving, big data analysis, and improving cluster utilization. Understand the project's progress, roadmap, and its impact on cloud native computing environments.
Syllabus
Volcano – Cloud Native Batch System for AI, BigData and HPC - William (LeiBo) Wang
Taught by
CNCF [Cloud Native Computing Foundation]