Anyscale has released KubeRay versions 1.6 and 1.7, introducing significant enhancements for running the Ray framework on Kubernetes. These updates include a beta release of the History Server for accessing logs and events of deleted Ray clusters, improved security features like NetworkPolicy support and automated mTLS certificate management, and advancements in RayService for incremental upgrades with rollback capabilities. The releases also bring new features for RayJob and RayCluster, such as RayCronJob support and a recreate upgrade strategy, alongside improved GCS fault tolerance and enhanced observability through the Ray Dashboard. AI
IMPACT Enhances the operational capabilities for AI workloads managed by Ray on Kubernetes, improving debugging and reliability.
RANK_REASON This is a software release for a tool that facilitates running Ray on Kubernetes, not a core AI model release or research paper.
- Alibaba Cloud
- Anyscale
- Google Cloud
- Hark
- KubeRay
- Kubernetes
- Microsoft
- opensource4you
- Ray
- RayCluster
- RayJob
- RayService
- Red Hat
- Redis
- RocksDB
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →