华为计算微信公众号
昇腾AI开发者公众号
华为计算微博
华为计算今日头条
"server_count":"4",
"server_list":[
{
"device":[
"device_id":"0",
"device_ip":"192.168.0.10",
"rank_id":"0"
},
"device_id":"1",
"device_ip":"192.168.0.11",
"rank_id":"1"
"device_id":"2",
"device_ip":"192.168.0.12",
"rank_id":"2"
"device_id":"3",
"device_ip":"192.168.0.13",
"rank_id":"3"
"device_id":"4",
"device_ip":"192.168.0.14",
"rank_id":"4"
"device_id":"5",
"device_ip":"192.168.0.15",
"rank_id":"5"
"device_id":"6",
"device_ip":"192.168.0.16",
"rank_id":"6"
"device_id":"7",
"device_ip":"192.168.0.17",
"rank_id":"7"
}
],
"server_id":"10.3.0.112",
"container_ip":"10.3.0.112"
"device_ip":"192.168.0.20",
"rank_id":"8"
"device_ip":"192.168.0.21",
"rank_id":"9"
"device_ip":"192.168.0.22",
"rank_id":"10"
"device_ip":"192.168.0.23",
"rank_id":"11"
"device_ip":"192.168.0.24",
"rank_id":"12"
"device_ip":"192.168.0.25",
"rank_id":"13"
"device_ip":"192.168.0.26",
"rank_id":"14"
"device_ip":"192.168.0.27",
"rank_id":"15"
"server_id":"10.3.0.162",
"container_ip":"10.3.0.162"
"device_ip":"192.168.0.30",
"rank_id":"16"
"device_ip":"192.168.0.31",
"rank_id":"17"
"device_ip":"192.168.0.32",
"rank_id":"18"
"device_ip":"192.168.0.33",
"rank_id":"19"
"device_ip":"192.168.0.34",
"rank_id":"20"
"device_ip":"192.168.0.35",
"rank_id":"21"
"device_ip":"192.168.0.36",
"rank_id":"22"
"device_ip":"192.168.0.37",
"rank_id":"23"
"server_id":"10.3.0.163",
"container_ip":"10.3.0.163"
"device_ip":"192.168.0.40",
"rank_id":"24"
"device_ip":"192.168.0.41",
"rank_id":"25"
"device_ip":"192.168.0.42",
"rank_id":"26"
"device_ip":"192.168.0.43",
"rank_id":"27"
"device_ip":"192.168.0.44",
"rank_id":"28"
"device_ip":"192.168.0.45",
"rank_id":"29"
"device_ip":"192.168.0.46",
"rank_id":"30"
"device_ip":"192.168.0.47",
"rank_id":"31"
"server_id":"10.3.0.233",
"container_ip":"10.3.0.233"
"status":"completed",
"version":"1.0"
} 以上是我的ranktable.json 但是我运行多机推理 提示我以下报错
[2025-03-19 14:49:49,054] [2721] [281473430219712] [llm] [ERROR] [logging.py-50] : World size does not match with ranktable file
[2025-03-19 14:49:49,055] [2721] [281473430219712] [llm] [ERROR] [logging.py-50] : Node num does not match with ranktable file
容器下都设置了
以下是运行命令
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 0 10.3.0.112
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 8 10.3.0.112
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 16 10.3.0.112
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 24 10.3.0.112
我要发帖子
"server_count":"4",
"server_list":[
{
"device":[
{
"device_id":"0",
"device_ip":"192.168.0.10",
"rank_id":"0"
},
{
"device_id":"1",
"device_ip":"192.168.0.11",
"rank_id":"1"
},
{
"device_id":"2",
"device_ip":"192.168.0.12",
"rank_id":"2"
},
{
"device_id":"3",
"device_ip":"192.168.0.13",
"rank_id":"3"
},
{
"device_id":"4",
"device_ip":"192.168.0.14",
"rank_id":"4"
},
{
"device_id":"5",
"device_ip":"192.168.0.15",
"rank_id":"5"
},
{
"device_id":"6",
"device_ip":"192.168.0.16",
"rank_id":"6"
},
{
"device_id":"7",
"device_ip":"192.168.0.17",
"rank_id":"7"
}
],
"server_id":"10.3.0.112",
"container_ip":"10.3.0.112"
},
{
"device":[
{
"device_id":"0",
"device_ip":"192.168.0.20",
"rank_id":"8"
},
{
"device_id":"1",
"device_ip":"192.168.0.21",
"rank_id":"9"
},
{
"device_id":"2",
"device_ip":"192.168.0.22",
"rank_id":"10"
},
{
"device_id":"3",
"device_ip":"192.168.0.23",
"rank_id":"11"
},
{
"device_id":"4",
"device_ip":"192.168.0.24",
"rank_id":"12"
},
{
"device_id":"5",
"device_ip":"192.168.0.25",
"rank_id":"13"
},
{
"device_id":"6",
"device_ip":"192.168.0.26",
"rank_id":"14"
},
{
"device_id":"7",
"device_ip":"192.168.0.27",
"rank_id":"15"
}
],
"server_id":"10.3.0.162",
"container_ip":"10.3.0.162"
},
{
"device":[
{
"device_id":"0",
"device_ip":"192.168.0.30",
"rank_id":"16"
},
{
"device_id":"1",
"device_ip":"192.168.0.31",
"rank_id":"17"
},
{
"device_id":"2",
"device_ip":"192.168.0.32",
"rank_id":"18"
},
{
"device_id":"3",
"device_ip":"192.168.0.33",
"rank_id":"19"
},
{
"device_id":"4",
"device_ip":"192.168.0.34",
"rank_id":"20"
},
{
"device_id":"5",
"device_ip":"192.168.0.35",
"rank_id":"21"
},
{
"device_id":"6",
"device_ip":"192.168.0.36",
"rank_id":"22"
},
{
"device_id":"7",
"device_ip":"192.168.0.37",
"rank_id":"23"
}
],
"server_id":"10.3.0.163",
"container_ip":"10.3.0.163"
},
{
"device":[
{
"device_id":"0",
"device_ip":"192.168.0.40",
"rank_id":"24"
},
{
"device_id":"1",
"device_ip":"192.168.0.41",
"rank_id":"25"
},
{
"device_id":"2",
"device_ip":"192.168.0.42",
"rank_id":"26"
},
{
"device_id":"3",
"device_ip":"192.168.0.43",
"rank_id":"27"
},
{
"device_id":"4",
"device_ip":"192.168.0.44",
"rank_id":"28"
},
{
"device_id":"5",
"device_ip":"192.168.0.45",
"rank_id":"29"
},
{
"device_id":"6",
"device_ip":"192.168.0.46",
"rank_id":"30"
},
{
"device_id":"7",
"device_ip":"192.168.0.47",
"rank_id":"31"
}
],
"server_id":"10.3.0.233",
"container_ip":"10.3.0.233"
}
],
"status":"completed",
"version":"1.0"
} 以上是我的ranktable.json 但是我运行多机推理 提示我以下报错
[2025-03-19 14:49:49,054] [2721] [281473430219712] [llm] [ERROR] [logging.py-50] : World size does not match with ranktable file
[2025-03-19 14:49:49,055] [2721] [281473430219712] [llm] [ERROR] [logging.py-50] : Node num does not match with ranktable file
容器下都设置了
以下是运行命令
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 0 10.3.0.112
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 8 10.3.0.112
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 16 10.3.0.112
bash run.sh pa_bf16 full_CEval 5 1 deepseekv2 /deepseek/Deepseek-R1-bf16-hfd-w8a8/ /home/user/rank_table_file.json 16 2 24 10.3.0.112