[launcher] look ma, no more zombies#714
Merged
jeffra merged 4 commits intodeepspeedai:masterfrom Feb 4, 2021
Merged
Conversation
stas00
requested review from
RezaYazdaniAminabadi,
ShadenSmith,
arashashari,
awan-10,
cli99,
conglongli,
eltonzheng,
jeffra,
minjiaz,
niumanar,
samyam and
tjruwase
as code owners
Collaborator
|
Thank you @stas! This has been on our backlog to investigate for a long time and a huge annoyance. Can’t wait to try this out soon. |
Collaborator
Author
|
Let's trade - I will fix annoying issues and perhaps someone could give me a bit of guidance with the 2D integration - as it looks I'm not making any progress - your docs say it supports non-DeepSpeed pipeline but it doesn't look it is the case :( #710 really could use help from someone who has the deepspeed internals understanding related to more than 1D parallelism. |
jeffra
approved these changes
Feb 4, 2021
Collaborator
|
Thanks @stas00 for the help here! We've tested on our side and this is awesome. One remaining note though, this will fix the zombie problem on a single-node/multi-process setup however the issue currently still remains on a multi-node/multi-process setup. I think it shouldn't be too bad to extend this to work on multi-node though now that we have the single-node/multi-process setup working. |
ConnorJL
pushed a commit
to EleutherAI/DeeperSpeed
that referenced
this pull request
Feb 8, 2021
* Dist testing backend fixes, etc. (deepspeedai#708) * set_batch_fn and remove old sanity check (deepspeedai#712) * properly set engine.local_rank if it's set to -1 * Add executable permission to `ds_elastic` and `ds_report` in `bin`. (deepspeedai#711) * Add executable permission to `ds_elastic` and `ds_report` in `bin`. * Automatic `ds_elastic` formatting Co-authored-by: Jeff Rasley <jerasley@microsoft.com> * local rank of -1 means not set (deepspeedai#720) * bump to 0.3.11 * [launcher] look ma, no more zombies (deepspeedai#714) Co-authored-by: Jeff Rasley <jerasley@microsoft.com> Co-authored-by: Jeff Rasley <jerasley@microsoft.com> Co-authored-by: Shaden Smith <Shaden.Smith@microsoft.com> Co-authored-by: Jon Eyolfson <eyolfson@gmail.com> Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
sdtblck
added a commit
to EleutherAI/DeeperSpeed
that referenced
this pull request
Feb 11, 2021
* Dist testing backend fixes, etc. (deepspeedai#708) * set_batch_fn and remove old sanity check (deepspeedai#712) * properly set engine.local_rank if it's set to -1 * Add executable permission to `ds_elastic` and `ds_report` in `bin`. (deepspeedai#711) * Add executable permission to `ds_elastic` and `ds_report` in `bin`. * Automatic `ds_elastic` formatting Co-authored-by: Jeff Rasley <jerasley@microsoft.com> * local rank of -1 means not set (deepspeedai#720) * bump to 0.3.11 * [launcher] look ma, no more zombies (deepspeedai#714) Co-authored-by: Jeff Rasley <jerasley@microsoft.com> * Improve starred expressions (deepspeedai#696) * Improve starred expressions `deepspeed/profiling/flops_profiler/profiler.py` uses starred expressions that are no longer valid with [PEP 617][1]. The new Python parser is in 3.9, and this change allows DeepSpeed to run with the newest Python version. I have not checked all locations that has this issue. However, this change allows me to run simple examples. [1]: https://www.python.org/dev/peps/pep-0617/ * Match style for "Improve starred expressions", although readability suffers The style guide might need to be updated for this new use case of expressions. Python [Issue 40631][1] includes more discussion on the change. [1]: https://bugs.python.org/issue40631 Co-authored-by: Cheng Li <pistasable@gmail.com> * Fixed typo in Readme. (deepspeedai#737) * 1bit_adam dependencies (deepspeedai#742) * Clickable screenshots (deepspeedai#746) * Fix docstring * Make screenshots clickable for easier viewing * Add flops profiler tutorial (deepspeedai#682) * work on flops profiler tutorial * update flops profiler tutorial * add flops profiler tutorial and fix names * work on flops profiler tutorial * update flops profiler tutorial * add flops profiler tutorial and fix names * fix tailing ws * fix names * remove multistep profiling and update docs * fix cases where functionals and submodules coexist in a parent module, update readme * fix typo * always invoke post hook function * fix module flops sum and update tests * update tutorial * Only initialize distributed if required (deepspeedai#734) Co-authored-by: Jeff Rasley <jerasley@microsoft.com> Co-authored-by: Jeff Rasley <jerasley@microsoft.com> Co-authored-by: Shaden Smith <Shaden.Smith@microsoft.com> Co-authored-by: Jon Eyolfson <eyolfson@gmail.com> Co-authored-by: Stas Bekman <stas00@users.noreply.github.com> Co-authored-by: Cheng Li <pistasable@gmail.com> Co-authored-by: TheDudeFromCI <thedudefromci@gmail.com> Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com> Co-authored-by: Sean Naren <sean@grid.ai>
This file contains hidden or 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
2 participants
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.
Almost identical to the solution merged into pytorch pytorch/pytorch#49305, this PR automatically takes care of killing zombies, should one of the processes fail (or get killed) or the parent process gets manually terminated by the operator.
As I already explained the 2 problems and the solution in full details here: pytorch/pytorch#49305 I hope it's ok if I don't repeat it again.
Thank you.