Skip to content

Commit ac74724

Browse files
committed
add node gpu/nouse-gpuuuid md
Signed-off-by: ZhengW22 <s123zz123@gmail.com>
1 parent 2ba97bc commit ac74724

6 files changed

Lines changed: 40 additions & 0 deletions

File tree

docs/userguide/Enflame-device/enable-enflame-gcu-sharing.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -82,6 +82,8 @@ spec:
8282
8383
## Device UUID Selection
8484
85+
### Pod Annotation
86+
8587
You can specify which GPU devices to use or exclude using annotations:
8688
8789
```yaml
@@ -98,6 +100,11 @@ spec:
98100
# ... rest of pod spec
99101
```
100102

103+
### Node Annotation
104+
105+
`enflame.com/nouse-gpuuuid`:
106+
String type, supports multiple device IDs. Multiple devices should be separated by comma, ie: "node1-enflame-2,node1-enflame-3". If set, HAMI will not allocate any devices defined in this string to Pods during scheduling.
107+
101108
> **NOTE:** The device ID format is `{node-name}-enflame-{index}`. You can find the available device IDs in the node status.
102109
103110
### Finding Device UUIDs

docs/userguide/Iluvatar-device/enable-illuvatar-gpu-sharing.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -102,6 +102,8 @@ spec:
102102
103103
## Device UUID Selection
104104
105+
### Pod Annotation
106+
105107
You can specify which GPU devices to use or exclude using annotations:
106108
107109
```yaml
@@ -118,6 +120,11 @@ spec:
118120
# ... rest of pod spec
119121
```
120122

123+
### Node Annotation
124+
125+
`iluvatar.ai/nouse-gpuuuid`:
126+
String type, supports multiple device IDs. Multiple devices should be separated by comma, ie: "node1-iluvatar-2,node1-iluvatar-3". If set, HAMI will not allocate any devices defined in this string to Pods during scheduling.
127+
121128
> **NOTE:** The device ID format is `{node-name}-iluvatar-{index}`. You can find the available device IDs in the node status.
122129
123130
### Finding Device UUIDs

docs/userguide/configure.md

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -81,6 +81,12 @@ helm install hami hami-charts/hami --set devicePlugin.deviceMemoryScaling=5 ...
8181
| `hami.io/gpu-scheduler-policy` | String | GPU scheduling policy: `"binpack"` allocates the pod to the same GPU card for execution. `"spread"` allocates the pod to different GPU cards for execution. | `"binpack"` or `"spread"` |
8282
| `nvidia.com/vgpu-mode` | String | The type of vGPU instance this pod wishes to use. | `"hami-core"` or `"mig"` |
8383

84+
## Node configs: annotations
85+
86+
| Argument | Type | Description | Example |
87+
|----------|------|-------------|---------|
88+
| `nvidia.com/nouse-gpuuuid` | String | If set, HAMI will not allocate any devices defined in this string to Pods during scheduling. | `"GPU-AAA,GPU-BBB"` |
89+
8490
## Container configs: env
8591

8692
| Argument | Type | Description | Default |

i18n/zh/docusaurus-plugin-content-docs/current/userguide/Enflame-device/enable-enflame-gcu-sharing.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -78,6 +78,8 @@ spec:
7878
7979
## 设备 UUID 选择
8080
81+
### Pod注解
82+
8183
你可以通过 Pod 注解来指定要使用或排除特定的 GPU 设备:
8284
8385
```yaml
@@ -94,6 +96,11 @@ spec:
9496
# ... rest of pod spec
9597
```
9698

99+
### Node注解
100+
101+
`enflame.com/nouse-gpuuuid`
102+
字符串类型,支持多个设备ID,使用,分割多个设备,如:"node1-enflame-2,node1-enflame-3"。如果设置,hami在进行调度将不会为Pod使用字符串中定义的任何设备。
103+
97104
> **说明:** 设备 ID 格式为 `{节点名称}-enflame-{索引}`。你可以在节点状态中找到可用的设备 ID。
98105
99106
### 查找设备 UUID

i18n/zh/docusaurus-plugin-content-docs/current/userguide/Iluvatar-device/enable-illuvatar-gpu-sharing.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -100,6 +100,8 @@ spec:
100100
101101
## 设备 UUID 选择
102102
103+
### Node annotation
104+
103105
你可以通过 Pod 注解来指定要使用或排除特定的 GPU 设备:
104106
105107
```yaml
@@ -116,6 +118,11 @@ spec:
116118
# ... 其余 Pod 配置
117119
```
118120

121+
### Pod Annotation
122+
123+
`iluvatar.ai/nouse-gpuuuid`
124+
字符串类型,支持多个设备ID,使用,分割多个设备,如:"node1-iluvatar-2,node1-iluvatar-3"。如果设置,hami在进行调度将不会为Pod使用字符串中定义的任何设备。
125+
119126
> **说明:** 设备 ID 格式为 `{节点名称}-iluvatar-{索引}`。你可以在节点状态中找到可用的设备 ID。
120127
121128
### 查找设备 UUID

i18n/zh/docusaurus-plugin-content-docs/current/userguide/configure.md

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -64,6 +64,12 @@ helm install hami hami-charts/hami --set devicePlugin.deviceMemoryScaling=5 ...
6464
| `hami.io/gpu-scheduler-policy` | 字符串 | GPU 卡调度策略:`"binpack"` 表示将 Pod 分配到同一块 GPU 卡上执行,`"spread"` 表示分配到不同的 GPU 卡上执行。 | `"binpack"``"spread"` |
6565
| `nvidia.com/vgpu-mode` | 字符串 | 指定该 Pod 希望使用的 vGPU 实例类型。 | `"hami-core"``"mig"` |
6666

67+
## Node 配置:注解
68+
69+
| 参数 | 类型 | 描述 | 示例 |
70+
| --- | ---- | --- | ----- |
71+
| `nvidia.com/nouse-gpuuuid` | 字符串 | 如果设置了此字段,则该 Pod 分配的设备 **必须** 是此字符串中定义的 GPU UUID 之一。 | `"GPU-AAA,GPU-BBB"` |
72+
6773
## 容器配置:环境变量
6874

6975
| 参数 | 类型 | 描述 | 默认值 |

0 commit comments

Comments
 (0)