[Core][Enable gcs scheduler 4/n] Fix PG scheduling by gcs scheduler #27084
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Chong-Li
force-pushed
the
refactor_gcs_sched_4/n
branch
from
3447ccb
to
0c83c99
Compare
Chong-Li
added
the
@author-action-required
rkooo567
removed
the
@author-action-required
Chong-Li
force-pushed
the
refactor_gcs_sched_4/n
branch
from
1d53218
to
a86107f
Compare
wumuzi520
reviewed
Jul 31, 2022
wumuzi520
reviewed
Jul 31, 2022
wumuzi520
reviewed
Jul 31, 2022
wumuzi520
reviewed
Jul 31, 2022
wumuzi520
reviewed
Jul 31, 2022
Chong-Li
force-pushed
the
refactor_gcs_sched_4/n
branch
from
daa6fa8
to
446ad7d
Compare
scv119
reviewed
Aug 15, 2022
scv119
reviewed
Aug 15, 2022
added 3 commits
August 16, 2022 15:55
Signed-off-by: Chong-Li <[email protected]>
Signed-off-by: Chong-Li <[email protected]>
Signed-off-by: Chong-Li <[email protected]>
Chong-Li
force-pushed
the
refactor_gcs_sched_4/n
branch
from
19c9dfe
to
37164f0
Compare
Chong-Li
requested review from
ericl,
richardliaw,
jjyao,
edoakes,
simon-mo,
jiaodong,
shrekris-anyscale,
sihanwang41 and
a team
as code owners
Chong-Li
removed request for
amogkam,
ArturNiederfahrenhorst,
edoakes,
Yard1,
suquark,
ckw017,
krfricke,
AmeerHajAli,
jovany-wang,
simon-mo,
ijrsvt,
SongGuyang,
kouroshHakha,
avnishn,
DmitriGekhtman,
xwjiang2010,
shrekris-anyscale and
jianoaix
added 3 commits
August 25, 2022 11:43
Signed-off-by: Chong-Li <[email protected]>
Signed-off-by: Chong-Li <[email protected]>
scv119
approved these changes
Aug 25, 2022
|
Do you have any more suggestions to this PR? @rkooo567 @iycheng |
7 tasks
XiaodongLv
pushed a commit
to XiaodongLv/ray
that referenced
this pull request
Sep 2, 2022
…ay-project#27084) This the first split PR of ray-project#25075, which tried to enable gcs scheduler by default. This split PR mainly includes: In GcsPlacementGroupScheduler::CommitBundleResources() and ReturnBundleResources(), we have to trigger pending actors (in gcs) because resources have been updated. Still in the above two functions, we have to update PG wildcard resources in a special way. A PG's wildcard resources (in a certain node) has to be the sum of all related bundle resources. Even though CommitBundleResources() uses ToNodeBundleResourcesMap() to sum up bundle resources, it does not handle the scenario that a single (or subset) bundle is rescheduled, in which this single bundle's wildcard resources would wrongly override the existing one. (see test_placement_group_reschedule_when_node_dead for such a scenario). Fix the remaining issues from ([Core][Enable gcs scheduler 3/n] integrate placement group with gcs scheduler ray-project#24842 (comment)).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Why are these changes needed?
This the first split PR of #25075, which tried to enable gcs scheduler by default.
This split PR mainly includes:
In
GcsPlacementGroupScheduler::CommitBundleResources()andReturnBundleResources(), we have to trigger pending actors (in gcs) because resources have been updated.Still in the above two functions, we have to update PG wildcard resources in a special way. A PG's wildcard resources (in a certain node) has to be the sum of all related bundle resources. Even though
CommitBundleResources()usesToNodeBundleResourcesMap()to sum up bundle resources, it does not handle the scenario that a single (or subset) bundle is rescheduled, in which this single bundle's wildcard resources would wrongly override the existing one. (seetest_placement_group_reschedule_when_node_deadfor such a scenario).Fix the remaining issues from ([Core][Enable gcs scheduler 3/n] integrate placement group with gcs scheduler #24842 (comment)).
Related issue number
#25075, #24842
Checks
scripts/format.shto lint the changes in this PR.