Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
songlinfeng
container-toolkit
Commits
59cf14f2
Commit
59cf14f2
authored
Nov 08, 2025
by
songlinfeng
💬
Browse files
Update README.md
parent
fdcfde81
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
11 additions
and
0 deletions
+11
-0
README.md
README.md
+11
-0
No files found.
README.md
View file @
59cf14f2
...
@@ -92,6 +92,7 @@ DCU Tracker用来监控使用--gpus和-e DTK_VISIBLE_DEVICES启动容器的DCU
...
@@ -92,6 +92,7 @@ DCU Tracker用来监控使用--gpus和-e DTK_VISIBLE_DEVICES启动容器的DCU
DCU Tracker 提供命令行来控制容器对DCU的访问,可以被设置为shared或exclusive.
DCU Tracker 提供命令行来控制容器对DCU的访问,可以被设置为shared或exclusive.
-
shared 表示DCU可以同时被多个容器一起使用,这是默认状态
-
shared 表示DCU可以同时被多个容器一起使用,这是默认状态
-
exclusive 表示DCU同时只能被一个容器使用。
-
exclusive 表示DCU同时只能被一个容器使用。
```
sh
```
sh
$
dtk-ctk dcu-tracker
-h
$
dtk-ctk dcu-tracker
-h
NAME:
NAME:
...
@@ -124,7 +125,9 @@ OPTIONS:
...
@@ -124,7 +125,9 @@ OPTIONS:
```
```
###使用DCU Tracker
###使用DCU Tracker
通过rocm-smi来查看节点上的DCUs
通过rocm-smi来查看节点上的DCUs
```
sh
```
sh
$
rocm-smi
$
rocm-smi
...
@@ -141,6 +144,7 @@ GPU Temp (DieEdge) AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU%
...
@@ -141,6 +144,7 @@ GPU Temp (DieEdge) AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU%
```
```
-
查看DCU Tracker Status
-
查看DCU Tracker Status
如果DCU Tracker enabled,DCU默认被赋予 shared 权限
如果DCU Tracker enabled,DCU默认被赋予 shared 权限
```
sh
```
sh
$
dtk-ctk dcu-tracker status
$
dtk-ctk dcu-tracker status
------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------
...
@@ -156,7 +160,9 @@ GPU Id UUID Accessibility Container Ids
...
@@ -156,7 +160,9 @@ GPU Id UUID Accessibility Container Ids
$
dtk-ctk dcu-tracker status
$
dtk-ctk dcu-tracker status
DCU Tracker is disabled
DCU Tracker is disabled
```
```
-
开启 DCU Tracker
-
开启 DCU Tracker
```
sh
```
sh
$
dtk-ctk dcu-tracker status
$
dtk-ctk dcu-tracker status
...
@@ -168,13 +174,17 @@ GPU Id UUID Accessibility Container Ids
...
@@ -168,13 +174,17 @@ GPU Id UUID Accessibility Container Ids
$
dtk-ctk dcu-tracker
enable
$
dtk-ctk dcu-tracker
enable
DCU Tracker is already enabled
DCU Tracker is already enabled
```
```
-
关闭 DCU Tracker
-
关闭 DCU Tracker
```
sh
```
sh
$
dtk-ctk dcu-tracker disable
$
dtk-ctk dcu-tracker disable
DCU Tracker has been disabled
DCU Tracker has been disabled
```
```
-
设置DCU的访问权限
-
设置DCU的访问权限
当DCU Tracker开启时,启动容器时会自动记录容器使用DCU的情况
当DCU Tracker开启时,启动容器时会自动记录容器使用DCU的情况
```
sh
```
sh
$
docker run
--name
slf_dmps
-e
DTK_VISIBLE_DEVICES
=
0,1
-it
a4dd5be0ca23
$
docker run
--name
slf_dmps
-e
DTK_VISIBLE_DEVICES
=
0,1
-it
a4dd5be0ca23
...
@@ -201,6 +211,7 @@ GPU Id UUID Accessibility Container Ids
...
@@ -201,6 +211,7 @@ GPU Id UUID Accessibility Container Ids
2 0x73873C7A6EB040A1 Shared None
2 0x73873C7A6EB040A1 Shared None
```
```
-
设置DCU 为exclusive属性
-
设置DCU 为exclusive属性
```
sh
```
sh
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment