[Usage]: How to run multi-node with trtllm-bench on kubernetes (LWS)
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 14.7k
- Forks
- 2.8k
- Avg merge
- 2d 23h
- Merged PRs (30d)
- 489
Description
System Info
System Information:
- OS:
- Python version:
- CUDA version:
- GPU model(s):
- Driver version:
- TensorRT-LLM version: v1.3.0rc17
How would you like to use TensorRT-LLM
Hi, I'm working on benchmarking the inference performance of trtllm on a customized model config.json, yet searched through docs and notice there is currently only clear guidelines on how to run trtllm-bench on single (kubernetes) node setup. Is there any way to run trtllm-bench on multi-node with LeaderWorkerSet?
I wish to run on 8 GPUs, with each node having 4 GPUs. We may consider running on 16 / 32 GPUs in future.
Before submitting a new issue...
- Make sure you already searched for relevant issues, and checked the documentation and examples for answers to frequently asked questions.
Contributor guide
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 TensorRT-LLM documentation and examples linked in the issue, focusing on the existing single-node trtllm-bench workflow. Determine whether a multi-node setup using LeaderWorkerSet is supported for the requested 8-GPU arrangement and what configuration is needed; done should be clear, reproducible guidance for running the benchmark across nodes.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- kubernetes
- Domain
- devops, distributed-systems
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 38/100