通过Pod创建实例

使用指定的Pod创建GPU实例

POST https://www.chenyu.cn/api/open/v2/instance/create_by_pod

通过Pod创建实例

cURL

curl --request POST \
  --url https://www.chenyu.cn/api/open/v2/instance/create_by_pod \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "pod_uuid": "<string>",
  "pod_tag": "<string>",
  "gpu_uuid": "<string>",
  "gpu_nums": 123
}
'

200

{
  "code": 123,
  "msg": "<string>",
  "data": {
    "instance_uuid": "<string>",
    "status": 123,
    "title": "<string>",
    "create_time": 123,
    "start_time": 123,
    "save_image_status": 123,
    "charging_type": 123,
    "shutdown_regular": {
      "shutdown_time": 123,
      "enable": true
    },
    "image_uuid": "<string>",
    "image_name": "<string>",
    "image_tag": "<string>",
    "gpu_uuid": "<string>",
    "gpu_name": "<string>",
    "gpu_nums": 123
  }
}

通过Pod创建实例

使用应用市场中的Pod创建一个新的GPU实例。

请求参数

pod_uuid · string · required

Pod uuid，可通过查询应用市场Pod接口获取

pod_tag · string

Pod版本标签，默认使用最新版本

gpu_uuid · string · required

GPU型号uuid

gpu_nums · integer · required

使用GPU数量，可选值：1/2/4/8

响应参数

code · integer

响应码

msg · string

响应信息

data · object

返回的实例数据

Show data

instance_uuid · string

实例uuid

status · integer

实例状态

title · string

实例名称

create_time · integer

实例创建时间，秒级时间戳

start_time · integer

实例启动时间，秒级时间戳

save_image_status · integer

实例镜像保存状态

• 1: 上传中
• 2: 上传成功
• 3: 上传失败

charging_type · integer

计费类型

shutdown_regular · object

定时关机设置

Show shutdown_regular

shutdown_time · integer

关机时间，秒级时间戳

enable · boolean

是否启用定时关机

image_uuid · string

实例使用的镜像uuid

image_name · string

实例使用的镜像名称

image_tag · string

实例使用的镜像tag

gpu_uuid · string

实例使用的gpu uuid

gpu_name · string

实例使用的gpu名称

gpu_nums · integer

实例使用的gpu数量

代码示例

Python

import requests
from datetime import datetime

url = "https://www.chenyu.cn/api/open/v2/instance/create_by_pod"
headers = {
    "Authorization": "Bearer your_api_key",
    "Content-Type": "application/json"
}

data = {
    "pod_uuid": "pod_12345678-1234-1234-1234-123456789012",
    "pod_tag": "v2.0",
    "gpu_uuid": "gpu_87654321-4321-4321-4321-210987654321",
    "gpu_nums": 1
}

response = requests.post(url, headers=headers, json=data)
result = response.json()

if result['code'] == 0:
    instance = result['data']
    create_time = datetime.fromtimestamp(instance['create_time'])
    
    print(f"实例创建成功!")
    print(f"实例UUID: {instance['instance_uuid']}")
    print(f"实例名称: {instance['title']}")
    print(f"GPU名称: {instance['gpu_name']}")
    print(f"GPU数量: {instance['gpu_nums']}")
    print(f"创建时间: {create_time.strftime('%Y-%m-%d %H:%M:%S')}")
    
    if instance['server_url']:
        print("服务访问地址:")
        for url in instance['server_url']:
            print(f"  - {url}")
else:
    print(f"创建失败: {result['msg']}")

响应示例

{
  "code": 0,
  "msg": "success",
  "data": {
    "instance_uuid": "inst_12345678-1234-1234-1234-123456789012",
    "status": 1,
    "title": "PyTorch深度学习实例",
    "create_time": 1703145600,
    "start_time": 1703145660,
    "save_image_status": 2,
    "charging_type": 1,
    "shutdown_regular": {
      "shutdown_time": 1703232000,
      "enable": true
    },
    "image_uuid": "img_87654321-4321-4321-4321-210987654321",
    "image_name": "PyTorch 2.0 环境",
    "image_tag": "v2.0",
    "gpu_uuid": "gpu_87654321-4321-4321-4321-210987654321",
    "gpu_name": "NVIDIA RTX 4090",
    "gpu_nums": 1
  }
}