[PD Disaggregation] support DP via v1 router and decouple DP and EP (#5197)

* [fix] support DP via v1 router and decouple DP and EP

* [fix] fix scripts

* [fix] reset model path

* [fix] dp use get_output_ep, fix router port type, update scripts

* [merge] merge with latest code

* [chore] remove some debug log

* [fix] fix code style check

* [fix] fix test_multi_api_server for log_dir name

* [chore] reduce logs

* Apply suggestions from code review

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
This commit is contained in:
Yonghua Li
2025-12-04 15:38:43 +08:00
committed by GitHub
parent 5cd17fd662
commit f4119d51b4
15 changed files with 394 additions and 146 deletions
@@ -280,7 +280,7 @@ class PrefixCacheManager:
+ f" --rdma_port {cache_config.rdma_comm_ports[i] if cache_config.rdma_comm_ports is not None else '0'}"
+ f" --speculative_config '{self.speculative_config.to_json_string()}'"
+ (" --create_cache_tensor" if create_cache_tensor else "")
+ f" >{log_dir}/launch_cache_transfer_manager_{int(device_ids[i])}.log 2>&1"
+ f" >{log_dir}/launch_cache_transfer_manager_tprank{i}.log 2>&1"
)
logger.info(f"Launch cache transfer manager, command:{launch_cmd}")
cache_manager_processes.append(subprocess.Popen(launch_cmd, shell=True, preexec_fn=os.setsid))
@@ -372,7 +372,7 @@ class PrefixCacheManager:
+ f" --engine_pid {pid_suffix}"
+ f" --rdma_port {cache_config.rdma_comm_ports[i] if cache_config.rdma_comm_ports is not None else '0'}"
+ f" --speculative_config '{self.speculative_config.to_json_string()}'"
+ f" >{log_dir}/launch_cache_messager_{int(device_ids[i])}.log 2>&1"
+ f" >{log_dir}/launch_cache_messager_tprank{i}.log 2>&1"
)
logger.info(f"Launch cache messager, command:{launch_cmd}")
cache_messager_processes.append(subprocess.Popen(launch_cmd, shell=True, preexec_fn=os.setsid))