deepspeedai / deepspeedai/DeepSpeedExamples

About available pinned memory limit

Open
#732 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
6.8k
Forks
1.1k
Avg merge
2d 16h
Merged PRs (30d)
1

Description

This statement the amount of pinned memory available on a system is much less than the total CPU memory mentioned in this link
https://github.com/microsoft/DeepSpeedExamples/tree/master/inference/huggingface/zero_inference#performance-tuning-tips is a little confusing. The limit of pinned memory available is due to os/kernel constraint or somewhat else? I'm afraid there's no explicit limit in Linux kernel.

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.

Research direction

Start with the performance-tuning tips section in the linked Hugging Face ZeRO inference documentation and verify the statement about pinned memory against Linux kernel behavior. Clarify whether the limit is imposed by the operating system or another constraint, and update the wording or add an explanatory reference when the cause is established.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface, linux
Domain
documentation, operating-systems
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
32/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.