## Why are these changes needed?
This is a part of redis removal. This PR remove redis kv in function table.
rpush related code is not updated in this PR.
## Related issue number
This PR adds support for automatic block splitting on read and map transforms, to keep block size bounded to ~500MiB. This avoids potential OOM situations where a map task may consume too much intermediate Python heap memory, or too much object store shared memory for one block.
## Why are these changes needed?
The .boto files are already added to the base image and ACL'ed to root, adding them again during app config build causes permission issues.
## Related issue number
## Why are these changes needed?
Ray wheel size limit is still at 100MB. Removing debug symbols would decrease Ray Linux wheel sizes.
## Related issue number
## Checks
* [xgboost] Fix release test app configs
* Revert full app config
* Update base docker image
* Only change cpu base image
* default
* Pin xgboost to 1.5. in cpu tests
* Remove numpy hack
* Revert one line
Co-authored-by: Amog Kamsetty <amogkamsetty@yahoo.com>
This PR removes global named actor and global PGs.
I believe these APIs are not used widely in OSS.
CPP part is not included in this PR.
@kfstorm @clay4444 @raulchen Please take a look if this change is reasonable.
IMPORTANT NOTE: This is a Java API change and will lead backward incompatibility in Java global named actor and global PG usage.
CPP part is not included in this PR.
INCLUDES:
Remove setGlobalName() and getGlobalActor() APIs.
Remove getGlobalPlacementGroup() and setGlobalPG
Add getActor(name, namespace) API
Add getPlacementGroup(name, namespace) API
Update doc pages.
This second PR in the stack that supports out or order execution for threaded/async actors. Previous PR #20148 Next PR #20150
At a high level, threaded actor/async actor already don't guarantee execution order, and the current "sequential" order implementation has caused some confusion and inconvenience. Please refer to #19822 for detailed discussion.
This PR we further separate out the logic for ordering actor requests on the client side. In the next PR, we will implement a different type of queue that supports out of order execution.
This is part of stack that enable out-of-order execution for actors. Previous PR #20150 Next PR #20176
Refactor the actor receiver code, by separating classes into their own header/cc files. specifically:
scheduling_queue.h for ScheduleQueue interface;
actor_scheduling_util.h for InBountRequest/DependencyWaiter/DependencyWaiterImpl
actor_scheduling_queue.h for ActorScheudlingQueue (the sequential execution queue)
normal_scheduling_queue.h for NormalSchedulingQueue (the task execution queue)
fiber_state_manager.h for FiberStateManager
thread_pool_manager.h for PoolManager and BoundedExecutor
## Why are these changes needed?
Since we are using gcs client as kv backend, we need to make it auto-reconnect in case of a failure. This PR adds this feature.
This PR adds auto_reconnect decorator to gcs-utils and in case of a failure it'll try to reconnect to gcs until it succeeds.
This feature right now support redis which should be deleted later once we finished bootstrap since kv will always go to gcs.
## Related issue number
Some Ray client users are likely seeing an issue similar to #7084. Inside a container, connecting to localhost: fails but connecting to 127.0.0.1: succeeds. Changing Ray client to use 127.0.0.1 for localhost connection / serving should fix the issue.
