在运行H100实验的时候观察到发向nvswitch的流量
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.2k
- Forks
- 184
- Avg merge
- 13h 4m
- Merged PRs (30d)
- 1
Description
现象及问题
在运行H100实验的时候观察到发向nvswitch的流量，我们想确认这个流量是否是预期内的，以及这个流量的实际意义。
复现步骤
1. 在代码中加入流的源、目的输出
2. 运行官方demo
$ AS_SEND_LAT=3 AS_NVLS_ENABLE=1 ./bin/SimAI_simulator -t 16 -w ./example/microAllReduce.txt -n ./Spectrum-X_128g_8gps_100Gbps_H100 -c astra-sim-alibabacloud/inputs/config/SimAI.conf
3. 从log中观察到流量发送到nvswitch
从拓扑中可以看到nvswitch的编号为8
可以看到在实验的最开始，有流量发送到nvswitch，随后有流量从nvswitch发回到GPU
如果将拓扑文件中的H100更换为A100，则没有这些流量。
提前感谢大家的时间和帮助！
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the reproduction command for bin/SimAI_simulator, using example/microAllReduce.txt, SimAI.conf, and the Spectrum-X_128g_8gps_100Gbps_H100 topology. Compare the logs and topology behavior with the A100 variant; done means determining whether the initial NVSwitch traffic is expected and documenting its actual meaning or identifying the relevant correction.
Written by the indexing model from the issue text.
Assessment
- Domain
- distributed-systems, networking
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100