[rabit improvement] support rabit worker set/get configs from tracker #94
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
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Outdated suggestions cannot be applied.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.
native rabit checkpoint restore were failing in XGB.
In order to let restart worker avoid run allreduce before loadcheckpoint, we plan to save those config(s) in dmlc-tracker when training job starts first time(e.g number of columns of partitioned training data set)
If worker failure, instead of calling allreduce and break rabit recovery assumption, we can fetch configs from tracker. This would allow checkpoint load correctly and starts at right iteration number.
If tracker die, training job will die anyway. We might leverage spark hdfs checkpoint and recover entire cluster from there.
More detail here
dmlc/xgboost#4250 (comment)