[Question] Detailed introduction about apex DistributedDataParallel

Open
#597 5 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
35/100
Issue type
Documentation
Clarity
Needs clarification
Activity status
Stale
Tech stack
python, pytorch
Domain
distributed-systems, documentation

Research direction

Start with the DistributedDataParallel documentation in parallel.html and the implementation in apex/parallel/distributed.py around line 129. Trace how communication and computation overlap during backward() and how gradient transfers are bucketed. Done means adding a clear explanation of these mechanisms for readers who cannot infer them from the source.

Written by the indexing model from the issue text.

Description

Dear apex developers,
I read from the documentation here:

DistributedDataParallel is optimized for use with NCCL. It achieves high performance by overlapping communication with computation during backward() and bucketing smaller gradient transfers to reduce the total number of transfers required.

I also read the source code of DistributedDataParallel , but I still can't understand the so-called overlapping.

Could you provide some more detailed explanation for me ?

Dominant language
Python
Stars
9k
Forks
1.5k
Avg merge
2d 4h
Merged PRs (30d)
3

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from NVIDIA/apex

All issues in NVIDIA/apex

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.