apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 00:33:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Sync Code] develop to release/2.0.3 (#2873)

Browse Source 
* [LLM] support send batch data and aggregate data (#2860)

* [LLM] support send batch data and aggregate data

* [LLM] fix ci bugs

* [LLM] fix ci bugs

* [LLM] fix ci bugs

* [LLM] fix ci bugs

* [LLM] update

* [LLM] Update Multinode Deployment (#2830)

* [LLM] fix multinode bugs

* [LLM] update multinode deployment

* [LLM] update multinode deployment

* [LLM] update multinode deployment

* [LLM] update multinode deployment

* [LLM] update multinode deployment

* [LLM] fix ci bugs

* Update fastdeploy/engine/args_utils.py

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

* [LLM] update random port

* [LLM] update random port

* [LLM] fix ci bugs

* fix ci bugs

---------

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

---------

Co-authored-by: ltd0924 <32387785+ltd0924@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
This commit is contained in:
Jiang-Jia-Jun
2025-07-16 23:44:26 +08:00
committed by GitHub
parent 63d6e7ce06
commit 09d0073fdc
18 changed files with 375 additions and 264 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
fastdeploy/engine/expert_service.py
View File

@@ -49,8 +49,8 @@ class ExpertService(object):
cfg (Config): Config object containing all the configuration parameters.
"""
self.cfg = cfg
start_pos = local_data_parallel_id * self.cfg.tensor_parallel_size
end_pos = (local_data_parallel_id + 1) * self.cfg.tensor_parallel_size
start_pos = (local_data_parallel_id * self.cfg.tensor_parallel_size) % self.cfg.worker_num_per_node
end_pos = ((local_data_parallel_id + 1) * self.cfg.tensor_parallel_size) % self.cfg.worker_num_per_node
self.cfg.cache_config.rdma_comm_ports = self.cfg.cache_config.rdma_comm_ports[
start_pos:end_pos]
self.cfg.local_device_ids = self.cfg.device_ids.split(
@@ -65,7 +65,7 @@ class ExpertService(object):
self.cfg.parallel_config.local_data_parallel_id = local_data_parallel_id
address = (cfg.pod_ips[0], cfg.engine_worker_queue_port)
address = (cfg.master_ip, cfg.engine_worker_queue_port)
self.engine_worker_queue = EngineWorkerQueue(
address=address,
is_server=False,
@@ -118,7 +118,7 @@ class ExpertService(object):
cache_config=self.cfg.cache_config,
tensor_parallel_size=self.cfg.tensor_parallel_size,
device_ids=self.cfg.local_device_ids,
pod_ip=self.cfg.pod_ips[0],
pod_ip=self.cfg.master_ip,
engine_worker_queue_port=self.cfg.engine_worker_queue_port,
pid_suffix=f"{local_data_parallel_id}_{ipc_signal_suffix}"
)