ffn server use vllm serve and dp

[jiangkuaixue123]

Purpose

FFN DP拉起
Node1:

vllm serve /home/y00889327/DSV2LiteWeight \
	-dp 2 \
	--data-parallel-size-local 1 \
	--data-parallel-address 127.0.0.1 \
	--data-parallel-rpc-port 13345 \
	--max_num_batched_tokens 72 \
	--max_num_seqs 72 \
	--max-model-len 4096 \
	--enforce-eager \
	--enable_expert_parallel \
	--afd-config '{"afd_connector":"m2nconnector", "num_afd_stages":"1", "afd_role": "ffn", "afd_extra_config":{"afd_size":"2A2F"}, "compute_gate_on_attention": "True"}' > ffn_node1.log 2>&1 &

Node2:

vllm serve /home/y00889327/DSV2LiteWeight \
	-dp 2 \
	--data-parallel-size-local 1 \
	--data-parallel-address 127.0.0.1 \
	--data-parallel-rpc-port 13345 \
	--data-parallel-start-rank 1 \
	--max_num_batched_tokens 72 \
	--max_num_seqs 72 \
	--max-model-len 4096 \
	--enforce-eager \
	--enable_expert_parallel \
	--headless \
	--afd-config '{"afd_connector":"m2nconnector", "num_afd_stages":"1", "afd_role": "ffn", "afd_extra_config":{"afd_size":"2A2F"}, "compute_gate_on_attention": "True"}' > ffn_node2.log 2>&1 &

Test Plan

Test Result

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
• (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.