# Ray 1.6.0 — Ray-1.6.0

- Product: Ray (https://whatsnew.fyi/product/ray)
- Vendor: Anyscale
- Date: 2021-08-23
- Version: 1.6.0
- Original notes: https://github.com/ray-project/ray/releases/tag/ray-1.6.0
- Permalink: https://whatsnew.fyi/product/ray/releases/1.6.0

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **added** — Runtime Environments are ready for general use, enabling dynamic specification of per-task, per-actor and per-job dependencies including working directory, environment variables, pip packages and conda environments
- **added** — Ray Dataset is now in alpha as an interchange format for distributed datasets powered by Arrow
- **added** — Ray Autoscaler now supports TPUs on GCP
- **added** — RLLib supports RNN/LSTM models with SAC agent (RNNSAC)
- **added** — RLLib supports ONNX model export for TensorFlow and PyTorch
- **added** — RLLib allows Policies to be added to or removed from a Trainer on-the-fly
- **added** — Tune supports dynamic trial resource allocation with ResourceChangingScheduler
- **added** — Tune supports define-by-run functions to generate search spaces with OptunaSearcher
- **changed** — pip install ray now has a significantly reduced set of dependencies
- **changed** — Ray Autoscaler supports running without NodeUpdaters
- **changed** — Ray Autoscaler provides better AWS networking configurability
- **changed** — Ray Client documentation updated for server ports and ray.init(ray://)
- **changed** — RLLib changed dependency from CV2 to Skimage
- **changed** — RLLib unified TensorFlow and PyTorch policies with respect to multi-GPU handling, improving PPO-torch performance on Atari by 33 percent on single GPU
- **changed** — Tune allows string names of searchers and schedulers to be used directly in tune.run
- **fixed** — Fixed plasma store bugs for better data processing stability
- **fixed** — Fixed placement group bug where CUDA_VISIBLE_DEVICES were not properly detected
- **fixed** — Fixed Ray Client server proxy not working with non-default Redis passwords
- **fixed** — Improved Ray stacktrace messages
- **fixed** — Improved GCS stability and scalability

##### Highlights

* [Runtime Environments](https://docs.ray.io/en/releases-1.6.0/advanced.html#runtime-environments) are ready for general use!  This feature enables you to dynamically specify per-task, per-actor and per-job dependencies, including a working directory, environment variables, pip packages and conda environments. Install it with `pip install -U 'ray[default]'`.
* Ray Dataset is now in alpha! Dataset is an interchange format for distributed datasets, powered by Arrow. You can also use it for a basic Ray native data processing experience. [Check it out here. ](https://docs.ray.io/en/releases-1.6.0/data/dataset.html)
* [Ray Lightning](https://github.com/ray-project/ray_lightning) v0.1 has been released! You can install it via `pip install ray-lightning`. Ray Lightning is a library of PyTorch Lightning plugins for distributed training using Ray. Features:
  * Enables quick and easy parallel training
  * Supports [PyTorch DDP](https://github.com/ray-project/ray_lightning#pytorch-distributed-data-parallel-plugin-on-ray), [Horovod](https://github.com/ray-project/ray_lightning#horovod-plugin-on-ray), and [Sharded DDP with Fairscale](https://github.com/ray-project/ray_lightning#model-parallel-sharded-training-on-ray)
  * Integrates with [Ray Tune for hyperparameter optimization](https://github.com/ray-project/ray_lightning#hyperparameter-tuning-with-ray-tune) and is compatible with [Ray Client](https://github.com/ray-project/ray_lightning#multi-node-training-from-your-laptop)
* `pip install ray` now has a significantly reduced set of dependencies. Features such as the dashboard, the cluster launcher, runtime environments, and observability metrics may require `pip install -U 'ray[default]'` to be enabled. Please report any issues on Github if this is an issue!

##### Ray Autoscaler

🎉 New Features:

* The Ray autoscaler now supports TPUs on GCP. Please refer to this example for spinning up a [simple TPU cluster](https://github.com/ray-project/ray/blob/releases/1.6.0/python/ray/autoscaler/gcp/tpu.yaml). (#17278)

💫Enhancements:

* Better AWS networking configurability (#17236 #17207 #14080)
* Support for running autoscaler without NodeUpdaters (#17194, #17328)

🔨 Fixes:

* Code clean up and corrections to downscaling policy (#17352)
* Docker file sync fix (#17361)  

##### Ray Client

💫Enhancements:

* Updated docs for client server ports and ray.init(ray://) (#17003, #17333)
* Better error handling for deserialization failures (#17035)

🔨 Fixes:

* Fix for server proxy not working with non-default redis passwords (#16885)

##### Ray Core

🎉 New Features:

* [Runtime Environments](https://docs.ray.io/en/releases-1.6.0/advanced.html#runtime-environments) are ready for general use!
  * Specify a working directory to upload your local files to all nodes in your cluster.
  * Specify different conda and pip dependencies for your tasks and actors and have them installed on the fly.

🔨 Fixes:

* Fix plasma store bugs for better data processing stability (#16976, #17135, #17140, #17187, #17204, #17234, #17396, #17550)
* Fix a placement group bug where CUDA_VISIBLE_DEVICES were not properly detected (#17318)
* Improved Ray stacktrace messages. (#17389)
* Improved GCS stability and scalability (#17456, #17373, #17334, #17238, #17072)

🏗 Architecture refactoring:

* Plasma store refactor for better testability and extensibility. (#17332, #17313, #17307)  

##### Ray Data Processing

Ray Dataset is now in alpha! Dataset is an interchange format for distributed datasets, powered by Arrow. You can also use it for a basic Ray native data processing experience. [Check it out here. ](https://docs.ray.io/en/releases-1.6.0/data/dataset.html)  

##### RLLib

🎉 New Features:

* Support for RNN/LSTM models with SAC (new agent: "RNNSAC"). Shoutout to ddworak94! (#16577)
* Support for ONNX model export (tf and torch). (#16805)
* Allow Policies to be added to/removed from

_[Truncated at 4000 characters — full notes: https://github.com/ray-project/ray/releases/tag/ray-1.6.0]_