批量更新实例
批量更新实例
Path Parameters
The service id.
Request Body required
- Array [
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- ]
服务名
instances object[] required
实例
实例名称
spec object required
实例规格
cluster object required
关联集群
订阅
资源组
可用区
工作空间
关联集群名
虚拟专有网络ID
model object required
模型规格
名称
ref object required
git引用
标签
分支
提交
Possible values: [TYPE_AMP_PRIVATE_MODEL_UNSPECIFIED
, TYPE_AMP_PUBLIC_MODEL
, TYPE_AFT_FINETUNE_MODEL
]
类型
framework object required
框架规格
名称
版本
镜像仓库
镜像标签
命令行
envs object[]
框架环境变量
名称
值
Possible values: [IMAGE_TYPE_NONE_UNSPECIFIED
, IMAGE_TYPE_OFFICIAL
, IMAGE_TYPE_PRIVATE
]
镜像类型
sidecar object
Sidecar参数
线程池大小
线程池等待请求数
Possible values: <= 1
负载阈值
replicas object required
副本数规格
Possible values: <= 100
值
resources object required
资源规格
cpu
内存
acceleration_hardwares object[]
加速硬件
Possible values: [TYPE_NONE_UNSPECIFIED
, TYPE_GPU_NVIDIA
]
硬件类型
Possible values: [MODEL_NONE_UNSPECIFIED
, MODEL_GPU_NVIDIA_A100
, MODEL_GPU_NVIDIA_A800
, MODEL_GPU_NVIDIA_L40S
, MODEL_GPU_NVIDIA_V100
, MODEL_GPU_NVIDIA_T4
]
硬件型号
数量
最小库存单元名称
volume_mounts object[]
Possible values: <= 9
卷挂载, 仅支持实例停止时更新
挂载路径
AI文件存储配置
可用区
finetune_models object[]
微调模型
名称
ref object required
git引用
标签
分支
提交
Possible values: [TYPE_AMP_PRIVATE_MODEL_UNSPECIFIED
, TYPE_AMP_PUBLIC_MODEL
, TYPE_AFT_FINETUNE_MODEL
]
类型
破坏性升级
status object
实例状态
render object
渲染状态
reasons object[]
原因
health object
健康状态
resource_group object
底层资源组
resources object[]
底层资源
replicas object[]
副本状态
health object
健康状态
owner object
底层资源
- 200
- default
OK
Schema
- Array [
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- Array [
- ]
- ]
instances object[]
实例
实例名称
spec object required
实例规格
cluster object required
关联集群
订阅
资源组
可用区
工作空间
关联集群名
虚拟专有网络ID
model object required
模型规格
名称
ref object required
git引用
标签
分支
提交
Possible values: [TYPE_AMP_PRIVATE_MODEL_UNSPECIFIED
, TYPE_AMP_PUBLIC_MODEL
, TYPE_AFT_FINETUNE_MODEL
]
类型
framework object required
框架规格
名称
版本
镜像仓库
镜像标签
命令行
envs object[]
框架环境变量
名称
值
Possible values: [IMAGE_TYPE_NONE_UNSPECIFIED
, IMAGE_TYPE_OFFICIAL
, IMAGE_TYPE_PRIVATE
]
镜像类型
sidecar object
Sidecar参数
线程池大小
线程池等待请求数
Possible values: <= 1
负载阈值
replicas object required
副本数规格
Possible values: <= 100
值
resources object required
资源规格
cpu
内存
acceleration_hardwares object[]
加速硬件
Possible values: [TYPE_NONE_UNSPECIFIED
, TYPE_GPU_NVIDIA
]
硬件类型
Possible values: [MODEL_NONE_UNSPECIFIED
, MODEL_GPU_NVIDIA_A100
, MODEL_GPU_NVIDIA_A800
, MODEL_GPU_NVIDIA_L40S
, MODEL_GPU_NVIDIA_V100
, MODEL_GPU_NVIDIA_T4
]
硬件型号
数量
最小库存单元名称
volume_mounts object[]
Possible values: <= 9
卷挂载, 仅支持实例停止时更新
挂载路径
AI文件存储配置
可用区
finetune_models object[]
微调模型
名称
ref object required
git引用
标签
分支
提交
Possible values: [TYPE_AMP_PRIVATE_MODEL_UNSPECIFIED
, TYPE_AMP_PUBLIC_MODEL
, TYPE_AFT_FINETUNE_MODEL
]
类型
破坏性升级
status object
实例状态
render object
渲染状态
reasons object[]
原因
health object
健康状态
resource_group object
底层资源组
resources object[]
底层资源
replicas object[]
副本状态
health object
健康状态
owner object
底层资源
{
"instances": [
{
"name": "string",
"spec": {
"cluster": {
"subscription": "string",
"resource_group": "string",
"zone": "string",
"workspace": "string",
"aec2_name": "string",
"vpc_id": "string"
},
"model": {
"name": "string",
"ref": {
"tag": "string",
"branch": "string",
"hash": "string"
},
"type": "TYPE_AMP_PRIVATE_MODEL_UNSPECIFIED"
},
"framework": {
"name": "string",
"display_name": "string",
"version": "string",
"image_repository": "string",
"image_tag": "string",
"commands": [
"string"
],
"envs": [
{
"name": "string",
"value": "string"
}
],
"image_type": "IMAGE_TYPE_NONE_UNSPECIFIED",
"sidecar": {
"pool_size": 0,
"pool_wait": 0,
"load_threshold": 0
}
},
"replicas": {
"value": 0
},
"resources": {
"cpu": "string",
"memory": "string",
"acceleration_hardwares": [
{
"type": "TYPE_NONE_UNSPECIFIED",
"model": "MODEL_NONE_UNSPECIFIED",
"count": "string"
}
],
"cci_sku_name": "string"
},
"volume_mounts": [
{
"mount_path": "string",
"afs_id": "string"
}
],
"zone": "string",
"finetune_models": [
{
"name": "string",
"ref": {
"tag": "string",
"branch": "string",
"hash": "string"
},
"type": "TYPE_AMP_PRIVATE_MODEL_UNSPECIFIED"
}
],
"destructive_update": true
},
"status": {
"render": {
"success": true,
"reasons": [
{
"field": "string",
"message": "string"
}
]
},
"health": {
"message": "string",
"status": "HEALTH_STATE_CODE_UNKNOWN_UNSPECIFIED"
},
"resource_group": {
"subscription": "string",
"resource_group": "string",
"zone": "string",
"workspace": "string",
"name": "string",
"resources": [
{
"name": "string",
"type": "string",
"content": "string"
}
]
},
"replicas": [
{
"name": "string",
"health": {
"message": "string",
"status": "HEALTH_STATE_CODE_UNKNOWN_UNSPECIFIED"
},
"owner": {
"name": "string",
"type": "string",
"content": "string"
}
}
],
"app_name": "string"
}
}
]
}
Default error response
Schema
- Array [
- ]
The status code, which should be an enum value of [google.rpc.Code][google.rpc.Code].
A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the [google.rpc.Status.details][google.rpc.Status.details] field, or localized by the client.
details object[]
A list of messages that carry the error details. There is a common set of message types for APIs to use.
The type of the serialized message.
{
"code": 0,
"message": "string",
"details": [
{
"@type": "string"
}
]
}