wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-08-13 12:30:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

Actor checkpointing with object lineage reconstruction (#1004)

Browse Source 
* Worker reports error in previous task, actor task counter is incremented after task is successful

* Refactor actor task execution

- Return new task counter in GetTaskRequest
- Update worker state for actor tasks inside of the actor method
  executor

* Manually invoked checkpoint method

* Scheduling for actor checkpoint methods

* Fix python bugs in checkpointing

* Return task success from worker to local scheduler instead of actor counter

* Kill local schedulers halfway through actor execution instead of waiting for all tasks to execute once

* Remove redundant actor tasks during dispatch, reconstruct missing dependencies for actor tasks

* Make executor for temporary actor methods

* doc

* Set default argument for whether the previous task was a success

* Refactor actor method call

* Simplify checkpoint task submission

* lint

* fix philipp's comments

* Add missing line

* Make actor reconstruction tests run faster

* Unimportant whitespace.

* Unimportant whitespace.

* Update checkpoint method signature

* Documentation and handle exceptions during checkpoint save/resume

* Rename get_task message field to actor_checkpoint_failed

* Fix bug.

* Remove debugging check, redirect test output
This commit is contained in:
Stephanie Wang
2017-10-12 09:53:32 -07:00
committed by Robert Nishihara
parent b585001881
commit 3764f2f2e1
14 changed files with 608 additions and 210 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/local_scheduler/local_scheduler_algorithm.h
+5 -2
View File
@@ -178,12 +178,15 @@ void handle_worker_removed(LocalSchedulerState *state,
*
* @param state The state of the local scheduler.
* @param algorithm_state State maintained by the scheduling algorithm.
* @param wi Information about the worker that is available.
* @param worker The worker that is available.
* @param actor_checkpoint_failed If the last task assigned was a checkpoint
* task that failed.
* @return Void.
*/
void handle_actor_worker_available(LocalSchedulerState *state,
SchedulingAlgorithmState *algorithm_state,
LocalSchedulerClient *worker);
LocalSchedulerClient *worker,
bool actor_checkpoint_failed);
/**
* Handle the fact that a new worker is available for running an actor.