Commit graph

4529 commits

Author SHA1 Message Date
Dmitri Gekhtman
0cbeefdadf
[autoscaler] GPU=0 resource tweak (#16761)
* tweak

* use >, some fields don't admit an order
2021-06-30 16:51:25 -07:00
Ian Rodney
1f14f366fd
[Client] Make Client_Hook per-thread (#16731) 2021-06-30 11:48:55 -07:00
architkulkarni
3962ce286b
[Core] Bump override_environment_variables removal date from Ray 1.5 to 1.6 (#16755) 2021-06-30 09:31:08 -05:00
architkulkarni
077c04b06c
[Serve] Fix fastapi tutorial and update doc with deprecation warning (#16759) 2021-06-30 09:30:49 -05:00
SongGuyang
1d5c72ebde
[C++ API] rename 'ray_redis_address' to 'ray_address' (#16772) 2021-06-30 20:50:28 +08:00
Kai Fricke
4b6f8a1ece
[cli] use shutil.move in ray cluster-dump and allow passing of tempfile (#16738) 2021-06-30 07:49:03 +01:00
Amog Kamsetty
69507f53db
[Horovod] Add Horovod example (#16742)
* wip

* updates

* updates

* update

* formatting

* updates

* updates

* update

* fix

* add timeout
2021-06-29 19:15:15 -07:00
Simon Mo
2ac8a197db
[Serve] Copy FastAPI ResponseModel field (#16760) 2021-06-29 16:28:08 -07:00
Patrick Ames
cf8785b0e1
[docs] Note that ordering of objects returned is preserved for ray.get. (#16763) 2021-06-29 16:17:16 -07:00
Richard Liaw
bcb73ed58b
finished impl (#16753) 2021-06-29 14:37:42 -07:00
Amog Kamsetty
56068f8f81
Skip test_component_failures_2 on Windows (#16745) 2021-06-29 14:06:09 -07:00
Ian Rodney
b8f950775e
[Client] Keep client_mode for dumps_from_client (#16732) 2021-06-29 13:30:10 -07:00
Amog Kamsetty
c0560dadef
[Docker] Pin Tensorflow (#16741) 2021-06-29 11:14:46 -07:00
Dmitri Gekhtman
257d072d13
[kubernetes][release] K8s release test instructions (#16662) 2021-06-29 10:57:35 -07:00
chenk008
c318293d9f
[Core] start worker in container (#16671) 2021-06-29 10:12:47 -07:00
Ian Rodney
b3532cc2d1
[Client][Test] Avoid Port-Reuse to DeFlake (#16697)
Co-authored-by: mwtian <81660174+mwtian@users.noreply.github.com>
2021-06-28 23:54:06 -07:00
Ian Rodney
a9df1b7a67
[Test][Modin] Actually run test_modin (#16719) 2021-06-28 20:39:30 -07:00
SangBin Cho
804a867b3d
Revert revert OBOD pubsub PR (#16487)
* Revert "Revert "[Pubsub] Use a pubsub module for Ownership based object directory (#16407)" (#16486)"

This reverts commit b986938f0f.

* revert the obod problem.

* Add stats.

* Fix a possible regression.

* in another progress

* debugging

* Fix stats bug

* update

* Add more stats.

* Add stats

* lint

* Fix issue

* remove spammy logs

* lint

* better error msg for debugging

* Add even more logging

* Remove spammy logs

* Fix iterator invalidation issue

* more debugging info

* fix

* Add more debug logs

* add debug logs

* Remove the problematic line for confirmation

* Completed

* Fixed a broken test.

* experiment

* Lint

* Add a better error message

* try out

* revert the build file.

* In progress again

* IP

* Formatting

* Revert the log level

* Unskip test array

* final clean up.

* fix a build issue

* debug logs

* remove

* .

* Add more critical logs.

* format

* tmp

* log

* log

* issue fix

* Upgrade

* test experiment

* Fix an issue

* Fix issues.

* Lint

* remove unnecessary code

* last clean up.

Co-authored-by: Stephanie Wang <swang@cs.berkeley.edu>
2021-06-28 20:30:31 -07:00
SongGuyang
41b9a5102b
[C++ worker] support build C++ worker during python setup (#16636) 2021-06-29 10:29:47 +08:00
Amog Kamsetty
322b9531f6
[SGD] Add __init__ file to tf.examples (#16726) 2021-06-28 19:23:22 -07:00
Ian Rodney
1a357a7e4f
[Client] Auto-Run ray.client().connect() (#16259) 2021-06-28 17:01:26 -07:00
Travis Addair
e5dfa4cfb9
[tune] Only use TBXLoggerCallback when torch is installed (#16695)
* [tune] Only use TBXLoggerCallback when torch is installed

* Fix lint

* fix

* Update python/ray/tune/utils/callback.py

Co-authored-by: Amog Kamsetty <amogkamsetty@yahoo.com>
Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com>
2021-06-28 16:34:20 -07:00
Alex Wu
0811ae4231
Use the same worker id in python and C++ (#16712)
Co-authored-by: Alex <alex@anyscale.com>
2021-06-28 15:42:37 -07:00
architkulkarni
b9f6132c08
skip flaky conda env fixture on MacOS (#16710) 2021-06-28 09:38:17 -07:00
Tao Wang
38157a3166
[Core]support external redis address when starting ray processes (#13170)
* support external redis address when starting ray processes

* use a more general name

* add cli option

* handle some details

* fix set shards logic

* reuse --address instead of introduce a new one

* lint

* tiny

* lint and fix
2021-06-28 09:22:40 -07:00
Kai Fricke
04bfba1274
[tune] Move reporter detection to utility function (#16673)
Test failures seem unrelated
2021-06-28 12:55:05 +01:00
Amog Kamsetty
54ce8092ab
[Tune] Update transformers to 4.6.1 (#16397)
* add examples

* update dask docs

* add build file

* formatting

* fix ci command

* fix

* Update python/ray/util/dask/BUILD

* newline

* fix pytest fixtures

* fixes

* formatting

* fix shuffle example

* update

* dont log to wandb
2021-06-26 14:10:47 -07:00
Eric Liang
aa882ed52d
Make it more convenient to develop ray.data by setting RAY_EXPERIMENTAL_DATA_API=1 (#16685)
* make it convenient to import ray.data

* update

* Update python/ray/experimental/data/read_api.py

Co-authored-by: Alex Wu <itswu.alex@gmail.com>

Co-authored-by: Alex Wu <itswu.alex@gmail.com>
2021-06-26 09:17:30 -07:00
Eric Liang
6bfa97eed7
Check in the first iteration of an Arrow-based dataset api (#16648) 2021-06-25 18:45:13 -07:00
Dmitri Gekhtman
7b58ec9ae5
[autoscaler] rsync bootstrap flag (#16667) 2021-06-25 15:26:47 -07:00
Eric Liang
9b17c35bee
Fix PullManager handling of get requests and liveness issues (#16394) 2021-06-25 13:01:46 -07:00
Kai Fricke
696334ff08
[tune] Fix Tee utility class properties (#16674) 2021-06-25 18:19:01 +01:00
architkulkarni
b15ab2d60b
[Core] [runtime env] Support specifying runtime env in @ray.remote decorator (#16660) 2021-06-25 09:37:40 -05:00
SongGuyang
e74d9d3ded
[runtime env] Download runtime env(conda) in agent instead of setup_worker (#16525) 2021-06-25 19:39:05 +08:00
dependabot[bot]
2e3771cc29
[tune](deps): Bump tensorflow-probability in /python/requirements/tune (#16561)
Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-06-25 11:50:35 +01:00
architkulkarni
8587f9d738
[Core] [runtime env] Fix conda/pip filepaths relative to working_dir (#16186) 2021-06-24 16:43:25 -05:00
architkulkarni
4637298d36
Delete conda env before creating to deflake test_runtime_env_complicated (#16628) 2021-06-24 12:13:26 -05:00
architkulkarni
e8c25a2fa4
[Core] [runtime env] Merge child's runtime_env["env_vars"] with that of parent (#16553) 2021-06-24 12:13:13 -05:00
Simon Mo
aabdfe2989
[Serve] Fix HTTP headers (#16647) 2021-06-24 11:59:43 -05:00
Amog Kamsetty
53d16365b0
[Release] Convert Horovod and SGD release tests (#15999) 2021-06-24 15:56:02 +01:00
Kai Fricke
ef97bdd407
[release] Fix app config: Install latest releases. Bump xgboost-ray version (#16581) 2021-06-24 12:56:21 +01:00
Gabriele Oliaro
3e2f608145
Work stealing! (#15475)
* work_stealing one commit squash

* using random task id to request workers

* inlining methods in direct_task_transport.h

* faster checking for presence of stealable tasks in RequestNewWorkerIfNeeded

* linting

* fixup! using random task id to request workers

* estimating number of tasks to steal based only on tasks in flight

* linting

* fixup! linting

* backup of changes

* fixed issue in scheduling queue test after merge

* linting

* redesigned work stealing. compiles but not tested

* all tests passing locally

* fixup! all tests passing locally

* fixup! fixup! all tests passing locally

* fixed big bug in StealTasksIfNeeded

* rev1

* rev2 (before removing the work_stealing param)

* removed work_stealing flag, fixed existing unit tests

* added unit tests; need to figure out how to assign distinct worker ids in GrantWorkerLease

* fixed work stealing test

* revisions, added two more unit/regression tests

* test
2021-06-23 17:08:28 -07:00
Frank Luan
9249287a36
Object spilling threshold (#16558)
* Object spilling threshold

* clang-format

* Make tests more lenient

* Fix tests

* Fix tests

* Address comments

* Fix tests lint

* Refactor

* Fix tests

* Fix cpp tests

* Address comments
2021-06-23 16:54:41 -07:00
SangBin Cho
f816f613c7
[Test] Handle flaky tests (#16602)
* Handle flaky tests.

* lint

* tag more

* add test_scheduling

* Remove global gc

Co-authored-by: Amog Kamsetty <amogkamsetty@yahoo.com>
2021-06-23 16:24:12 -07:00
Amog Kamsetty
b9e5ca4c18
[tune] Deflake mnist_ptl_mini (#16555) 2021-06-23 14:26:40 -07:00
SangBin Cho
ccb02dacb6
Mark the global gc test unflaky (#16601) 2021-06-23 13:38:32 -07:00
architkulkarni
9cb65d5e2f
[Core] Move wheel URL utils from test_utils to utils (#16386) 2021-06-23 13:41:02 -05:00
chenk008
82d92d0d61
[Core]Use worker shim PID to check worker registration (#16398) 2021-06-22 21:12:53 -07:00
Kai Fricke
a1765ac627
[tune] move to local parameter registry for tune.with_parameters() (#16611) 2021-06-22 17:58:11 -07:00
Chris K. W
b4f2cbce02
[Client] Disconnect on dataclient error (#16588)
* disconnect when main thread finds dataclient shut down, update error messages

* Add test_dataclient_disconnect to small tests

* drop unused var

* add __main__ section to test

* avoid direct ray import

* rerun
2021-06-22 16:46:10 +03:00