Commit graph

4289 commits

Author SHA1 Message Date
Simon Mo
7a5981f244
[Serve] Feature flag and turn off placement group usage. (#15865) 2021-05-19 15:43:46 -07:00
Ian Rodney
4825f1b2a5
[client] One Driver per RayClient Server (#15923) 2021-05-19 15:40:49 -07:00
dependabot[bot]
8a9bebb5e4
[tune](deps): Bump timm from 0.3.2 to 0.4.5 in /python/requirements (#15824)
* [tune](deps): Bump timm from 0.3.2 to 0.4.5 in /python/requirements

Bumps [timm](https://github.com/rwightman/pytorch-image-models) from 0.3.2 to 0.4.5.
- [Release notes](https://github.com/rwightman/pytorch-image-models/releases)
- [Changelog](https://github.com/rwightman/pytorch-image-models/blob/master/docs/changes.md)
- [Commits](https://github.com/rwightman/pytorch-image-models/commits/v0.4.5)

Signed-off-by: dependabot[bot] <support@github.com>

* updates

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Amog Kamsetty <amogkamsetty@yahoo.com>
2021-05-19 14:29:34 -07:00
architkulkarni
c3d06697bb
[Core] Add dynamic conda env install in shim process (#15881) 2021-05-19 15:46:42 -05:00
Edward Oakes
a116875abc
[serve] Add properties + docstring + test for Deployment class (#15917) 2021-05-19 14:44:00 -05:00
Eric Liang
836c739fe5
Revert "[client] One Driver per RayClient Server (#15875)" (#15922)
This reverts commit 97d1414f23.
2021-05-19 11:58:29 -07:00
Chris K. W
df58c9c7f7
[autoscaler][aws] deprecate worker_nodes and head_node (#15584)
Co-authored-by: Ian Rodney <ian.rodney@gmail.com>
Co-authored-by: Chris Wong <cwong@anyscale.com>
Co-authored-by: Dmitri Gekhtman <dmitri.m.gekhtman@gmail.com>
2021-05-19 11:54:29 -07:00
Dmitri Gekhtman
a7a5a2b2b7
[autoscaler][kubernetes][minor][hotfix] Fix havoc-wreaking typo (#15916) 2021-05-19 13:52:26 -05:00
dependabot[bot]
c164e73c7c
[tune](deps): Bump gluoncv in /python/requirements/tune (#15845)
Bumps [gluoncv](https://github.com/dmlc/gluon-cv) from 0.9.1 to 0.10.1.post0.
- [Release notes](https://github.com/dmlc/gluon-cv/releases)
- [Commits](https://github.com/dmlc/gluon-cv/commits)

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-05-19 10:35:30 -07:00
Edward Oakes
2267befe27
[serve] Fix bug where placement group was always detached even in non-detached instances (#15885) 2021-05-19 12:22:58 -05:00
Eric Liang
2dc4198210
Increase the raylet start wait timeout to accomodate plasma preallocation (#15860)
* update

* add doc

* update

* quick fix

* no spam

* fix
2021-05-19 09:39:25 -07:00
Ian Rodney
97d1414f23
[client] One Driver per RayClient Server (#15875) 2021-05-19 09:03:09 -07:00
architkulkarni
c636bc3065
[Serve] [Core] Fix serve on Windows by disabling runtime env on Windows (#15838) 2021-05-19 10:58:40 -05:00
SangBin Cho
29e9ad01e2
Fix ray memory command broken with RAY_ADDRESS='auto'. (#15892) 2021-05-18 22:06:17 -07:00
Kathryn Zhou
891648ea9e
Bump OpenTelemetry version & bug fix (#15791) 2021-05-18 17:06:43 -07:00
dependabot[bot]
4c8813f2e8
[RLlib](deps): Bump pettingzoo in /python/requirements/rllib (#15846)
Bumps [pettingzoo](https://github.com/PettingZoo-Team/PettingZoo) from 1.8.1 to 1.8.2.
- [Release notes](https://github.com/PettingZoo-Team/PettingZoo/releases)
- [Commits](https://github.com/PettingZoo-Team/PettingZoo/compare/1.8.1...1.8.2)

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-05-18 11:25:52 +02:00
architkulkarni
194c5e3a96
[Core] Cache workers by runtime_env in worker pool (#15782)
* pass RuntimeEnv in task spec as opaque string

* lint

* set correct empty value for json: "{}" not ""

* add comment for field in proto

* fix worker pool test by checking both "" and "{}"

* add RAY_CHECK todo

* make dict empty if all values null

* remove unnecessary ser/de

* fix

* address comments

* add WorkerCacheKey with hash function

* clean up

* add naive impl., dedicated workers never killed

* put dedicated workers in idle_of_all_languages

* pipe env hash from worker.py -> Worker

* fully pipe through hash, basic cache test passing

* use int type for runtime env hash

* convert Worker env hash type from size_t to int

* fix

* add method to MockWorker to fix cpp tests

* make compatible with java streaming test

* restore old dynamic_options code to fix java test

* address comments

* add comment about sorting before hash

* add comments for private members of WorkerCacheKey
2021-05-18 00:19:27 -07:00
Yi Cheng
863532af0a
[core] API for pre-run customized functions (#15749)
* run customer setup fn

* fix

* lint

* skip on w32

* fix comment

* up

* up
2021-05-17 22:52:36 -07:00
Alex Wu
69f228d22d
[core] Record actor+job start/end times and metadata (#15803) 2021-05-17 21:38:39 -07:00
Frank Luan
0dc34566fe
Refactor raylet to allocate+write+seal one return object at a time (#15757)
* Refactor raylet to allocate+write+seal one return object at a time

* Fix build

* Fix C++ and Java runtime

* Skip Windows testing

* Fix java and cpp runtime

* Fix warnings

* Fix cpp and java tests

* Fix cpp and java runtime

Co-authored-by: Stephanie Wang <swang@cs.berkeley.edu>
2021-05-17 20:06:08 -07:00
Dmitri Gekhtman
95c3d88cac
[autoscaler][kubernetes] Helm chart (#15614) 2021-05-17 16:55:10 -07:00
Dmitri Gekhtman
c2b8381015
[autoscaler][gcp] Migrate GCP config to available node types (#15805) 2021-05-17 15:45:47 -07:00
dependabot[bot]
434465e477
[tune](deps): Bump gpy from 1.9.9 to 1.10.0 in /python/requirements/tune (#15850)
Bumps [gpy](https://github.com/SheffieldML/GPy) from 1.9.9 to 1.10.0.
- [Release notes](https://github.com/SheffieldML/GPy/releases)
- [Changelog](https://github.com/SheffieldML/GPy/blob/devel/CHANGELOG.md)
- [Commits](https://github.com/SheffieldML/GPy/compare/v1.9.9...v1.10.0)

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-05-17 11:33:11 -07:00
Alex Wu
3744026897
Fix test_scheduling (#15823)
* done

* Update python/ray/tests/test_scheduling.py

Co-authored-by: Ian Rodney <ian.rodney@gmail.com>

* Update python/ray/tests/test_scheduling.py

Co-authored-by: Ian Rodney <ian.rodney@gmail.com>

* lint

Co-authored-by: Alex Wu <alex@anyscale.com>
Co-authored-by: Ian Rodney <ian.rodney@gmail.com>
2021-05-17 10:08:01 -07:00
Alex Wu
3e94114336
Namespaces (#15774) 2021-05-17 10:04:22 -07:00
Sven Mika
f25d58492d
[Testing] Dependabot for RLlib. (#15812) 2021-05-17 18:24:13 +02:00
architkulkarni
78c26ac9fe
[runtime_env] Use sys executable in shim if conda not specified (#15834) 2021-05-17 11:19:58 -05:00
lanlin
5d2ed47978
[tune] Allow to set buffer_length via tune.run (#15810) 2021-05-17 13:11:26 +01:00
Edward Oakes
cd32a92edc
[serve] Avoid exporting actor class for every replica (#15788) 2021-05-15 09:04:09 -05:00
Dmitri Gekhtman
d1b1ae0f45
[test][client][dask] Run dask tests in client mode. (#15806) 2021-05-14 17:15:59 -07:00
Edward Oakes
f6be6dbcdc
[Serve] batch slow warning for multiple replicas (#15798) 2021-05-14 13:12:32 -07:00
Ian Rodney
ec5322a463
[Client] ray.client.connect() and ray.ClientBuilder (#15706) 2021-05-14 00:08:39 -07:00
Ian Rodney
395c0ea03c
[Testing] Pin Tensorflow Version in requirements.txt (#15799) 2021-05-13 17:09:15 -07:00
Edward Oakes
28f2962bb2
[serve] Add helpful log messages when deploying (#15689) 2021-05-13 18:10:23 -05:00
Ian Rodney
42f99541d4
[Tests] Mark test_scheduling & test_memstat as Flaky (#15789) 2021-05-13 15:46:12 -07:00
Edward Oakes
6a0f087643
[serve] Randomly shuffle replicas to avoid cross-handle synchronization (#15792) 2021-05-13 17:19:27 -05:00
Richard Liaw
c624e89483
[tune] Support numpy types in TBXlogger (#15760) 2021-05-13 14:54:47 -07:00
Edward Oakes
77d713ac78
[serve] Fix shutdown logic + add test (#15790) 2021-05-13 16:43:07 -05:00
Edward Oakes
d107cca1aa
[serve] Don't deserialize backend classes in the controller (#15741) 2021-05-13 16:01:09 -05:00
Ian Rodney
859703e993
[RuntimeEnv] Log which file caused an Exception (#15772) 2021-05-13 13:48:59 -07:00
Simon Mo
838cfec122
[Tracing] Fix kwargs replacement (#15742) 2021-05-13 12:44:35 -07:00
architkulkarni
a0c1cfe034
[Core] Pass RuntimeEnv as opaque string in the task spec (#15658) 2021-05-13 10:32:00 -05:00
Simon Mo
d97ff80149
[Core] Fix global state keep initializing in busy loop (#15777) 2021-05-13 10:30:49 -05:00
Richard Liaw
ec27f9cc6b
[tune] avoid system exit error (#15687)
Signed-off-by: Richard Liaw <rliaw@berkeley.edu>
2021-05-13 00:24:30 -07:00
SangBin Cho
2fc892166f
[Test] unskip global gc (#15767) 2021-05-12 15:45:59 -07:00
Ian Rodney
6e3ec98b72
[test] Clean-Up Test-Top-Level API (#15713) 2021-05-12 13:50:17 -07:00
Dmitri Gekhtman
052d2acaee
[autoscaler][kubernetes] Restart after head failure, more consistent operator restart behavior. (#15655) 2021-05-12 11:49:11 -05:00
fyrestone
56c309416e
[Job submission] Basic job submission structure (#15103) 2021-05-12 15:08:20 +08:00
Edward Oakes
8c2ac84988
[serve] Don't filter logs by job_id when using Serve (#15691)
* dont filter logs

* modify worker
2021-05-11 15:54:44 -07:00
Eric Liang
cb59d30917
Drop profiling events if the GCS becomes backlogged (#15726) 2021-05-11 14:10:34 -07:00