aliyun / aliyun/SimAI

vidur-alibabacloud使用aicb后端时,单请求的TPOT、TTFT以及E2E时间是否能与真实的集群的时间对齐。

Open
#305 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.2k
Forks
184
Avg merge
13h 4m
Merged PRs (30d)
1

Description

如题,或者是有没有人知道这个误差能控制到什么范围呀?我测试了Qwen3-next-80B在DP4 EP4的四卡环境上,有20%的误差,是我做的有问题吗?还是实际上这个误差是符合预期的?有人做过相同的工作吗?

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 by tracing the aicb backend setup used for Qwen3-next-80B on the DP4 EP4 four-card configuration, then compare its single-request TPOT, TTFT, and E2E measurements with the real-cluster timings. Establish whether the reported 20% difference comes from the test setup or is within the expected accuracy range, documenting the resulting bounds or reproduction details.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.