deepspeedai / deepspeedai/DeepSpeedExamples

About multiple-thread attention computation on CPU using zero-inference example.

Open
#886 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

Hi,

I am trying to test the attention computation on the CPU with zero-interference.

I use the following command to run the script.

BSZ=96
LOG_DIR=$BASE_LOG_DIR/${MODEL_NAME}_bs${BSZ}
mkdir -p  $LOG_DIR
deepspeed --num_gpus 1 run_model.py --dummy --model ${FULL_MODEL_NAME} --batch-size ${BSZ} --cpu-offload --pin-memory 1 --offload-dir /tmp/data/dxu --gen-len 32 --pin-memory 1 --kv-offload --async_kv_offload

Duing the exection, I only saw one core is used.
Screenshot 2024-04-04 at 3 11 39 PM

However, there are many processes are created for the run.

Are there any other parameters or env I should configure to enable multiple-cores?

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 run_model.py and the supplied DeepSpeed command, focusing on the CPU offload, pin-memory, KV offload, and async KV offload options. Reproduce the zero-interference run and inspect process and core usage; done means identifying the missing configuration or confirming that the example cannot use multiple CPU cores.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.