Ray version 1.13 is here!
The highlights in this release include:
- Scalable shuffle in Ray Datasets (now in alpha), improving support for shuffling terabyte-scale and larger datasets
- Performance fixes and other enhancements to the Ray Serve Deployment Graph API
- Enhancements to KubeRay, including improved ​​stability of autoscaling support (alpha)
- Usage stats data collection, now on by default (guarded by an opt-out prompt), helping the open source Ray engineering team better understand how to improve our libraries and core APIs
And there’s more. Head over to the release blog for the deep dive.