Merge pull request #110 from qiangwei1983/master

Updated Iluvatar GPU related documentation

Author: hami-robot[bot]

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

@@ -2,7 +2,6 @@
 title: Enable Illuvatar GPU Sharing
 ---
 
-
 ## Introduction
 
 **We now support iluvatar.ai/gpu(i.e MR-V100、BI-V150、BI-V100) by implementing most device-sharing features as nvidia-GPU**, including:
@@ -28,52 +27,67 @@ title: Enable Illuvatar GPU Sharing
 
 > **NOTICE:** *Install only gpu-manager, don't install gpu-admission package.*
 
-* Identify the resource name about core and memory usage(i.e 'iluvatar.ai/vcuda-core', 'iluvatar.ai/vcuda-memory')
-
-* set the 'iluvatarResourceMem' and 'iluvatarResourceCore' parameters when install hami
-
+* set the devices.iluvatar.enabled=true when install hami
 ```
-helm install hami hami-charts/hami --set scheduler.kubeScheduler.imageTag={your kubernetes version} --set iluvatarResourceMem=iluvatar.ai/vcuda-memory --set iluvatarResourceCore=iluvatar.ai/vcuda-core -n kube-system
+helm install hami hami-charts/hami --set scheduler.kubeScheduler.imageTag={your kubernetes version} --set devices.iluvatar.enabled=true
 ```
 
-> **NOTE:** The default resource names are:
-> - `iluvatar.ai/vgpu` for GPU count
-> - `iluvatar.ai/vcuda-memory` for memory allocation
-> - `iluvatar.ai/vcuda-core` for core allocation
->
-> You can customize these names using the parameters above.
+**Note:** The currently supported GPU models and resource names are defined in (https://github.com/Project-HAMi/HAMi/blob/master/charts/hami/templates/scheduler/device-configmap.yaml):
+```yaml
+    iluvatars:
+    - chipName: MR-V100
+      commonWord: MR-V100
+      resourceCountName: iluvatar.ai/MR-V100-vgpu
+      resourceMemoryName: iluvatar.ai/MR-V100.vMem
+      resourceCoreName: iluvatar.ai/MR-V100.vCore
+    - chipName: MR-V50
+      commonWord: MR-V50
+      resourceCountName: iluvatar.ai/MR-V50-vgpu
+      resourceMemoryName: iluvatar.ai/MR-V50.vMem
+      resourceCoreName: iluvatar.ai/MR-V50.vCore
+    - chipName: BI-V150
+      commonWord: BI-V150
+      resourceCountName: iluvatar.ai/BI-V150-vgpu
+      resourceMemoryName: iluvatar.ai/BI-V150.vMem
+      resourceCoreName: iluvatar.ai/BI-V150.vCore
+    - chipName: BI-V100
+      commonWord: BI-V100
+      resourceCountName: iluvatar.ai/BI-V100-vgpu
+      resourceMemoryName: iluvatar.ai/BI-V100.vMem
+      resourceCoreName: iluvatar.ai/BI-V100.vCore
+```
 
 ## Device Granularity
 
 HAMi divides each Iluvatar GPU into 100 units for resource allocation. When you request a portion of a GPU, you're actually requesting a certain number of these units.
 
 ### Memory Allocation
 
-- Each unit of `iluvatar.ai/vcuda-memory` represents 256MB of device memory
+- Each unit of `iluvatar.ai/<card-type>.vMem` represents 256MB of device memory
 - If you don't specify a memory request, the system will default to using 100% of the available memory
 - Memory allocation is enforced with hard limits to ensure tasks don't exceed their allocated memory
 
 ### Core Allocation
 
-- Each unit of `iluvatar.ai/vcuda-core` represents 1% of the available compute cores
+- Each unit of `iluvatar.ai/<card-type>.vCore` represents 1% of the available compute cores
 - Core allocation is enforced with hard limits to ensure tasks don't exceed their allocated cores
 - When requesting multiple GPUs, the system will automatically set the core resources based on the number of GPUs requested
 
 ## Running Iluvatar jobs
 
 Iluvatar GPUs can now be requested by a container
-using the `iluvatar.ai/vgpu`, `iluvatar.ai/vcuda-memory` and `iluvatar.ai/vcuda-core`  resource type:
+using the `iluvatar.ai/BI-V150-vgpu`, `iluvatar.ai/BI-V150.vMem` and `iluvatar.ai/BI-V150.vCore`  resource type:
 
 ```yaml
 apiVersion: v1
 kind: Pod
 metadata:
-  name: poddemo
+  name: BI-V150-poddemo
 spec:
   restartPolicy: Never
   containers:
-  - name: poddemo
-    image: harbor.4pd.io/vgpu/corex_transformers@sha256:36a01ec452e6ee63c7aa08bfa1fa16d469ad19cc1e6000cf120ada83e4ceec1e
+  - name: BI-V150-poddemo
+    image: registry.iluvatar.com.cn:10443/saas/mr-bi150-4.3.0-x86-ubuntu22.04-py3.10-base-base:v1.0
     command:
     - bash
     args:
@@ -87,19 +101,17 @@ spec:
       sleep 360000
     resources:
       requests:
-        iluvatar.ai/vgpu: 1
-        iluvatar.ai/vcuda-core: 50
-        iluvatar.ai/vcuda-memory: 64
+        iluvatar.ai/BI-V150-vgpu: 1
+        iluvatar.ai/BI-V150.vCore: 50
+        iluvatar.ai/BI-V150.vMem: 64
       limits:
-        iluvatar.ai/vgpu: 1
-        iluvatar.ai/vcuda-core: 50
-        iluvatar.ai/vcuda-memory: 64
+        iluvatar.ai/BI-V150-vgpu: 1
+        iluvatar.ai/BI-V150.vCore: 50
+        iluvatar.ai/BI-V150.vMem: 64
 ```
 
 > **NOTICE1:** *Each unit of vcuda-memory indicates 256M device memory*
 
-> **NOTICE2:** *You can find more examples in [examples/iluvatar folder](https://github.com/Project-HAMi/HAMi/tree/release-v2.6/examples/iluvatar/)*
-
 ## Device UUID Selection
 
 You can specify which GPU devices to use or exclude using annotations:
@@ -111,15 +123,13 @@ metadata:
   name: poddemo
   annotations:
     # Use specific GPU devices (comma-separated list)
-    iluvatar.ai/use-gpuuuid: "node1-iluvatar-0,node1-iluvatar-1"
+    hami.io/use-<card-type>-uuid: "device-uuid-1,device-uuid-2"
     # Or exclude specific GPU devices (comma-separated list)
-    iluvatar.ai/nouse-gpuuuid: "node1-iluvatar-2,node1-iluvatar-3"
+    hami.io/no-use-<card-type>-uuid: "device-uuid-1,device-uuid-2"
 spec:
   # ... rest of pod spec
 ```
 
-> **NOTE:** The device ID format is `{node-name}-iluvatar-{index}`. You can find the available device IDs in the node status.
-
 ### Finding Device UUIDs
 
 You can find the UUIDs of Iluvatar GPUs on a node using the following command:
@@ -131,7 +141,7 @@ kubectl get pod <pod-name> -o yaml | grep -A 10 "hami.io/<card-type>-devices-all
 Or by examining the node annotations:
 
 ```bash
-kubectl get node <node-name> -o yaml | grep -A 10 "hami.io/node-register-<card-type>"
+kubectl get node <node-name> -o yaml | grep -A 10 "hami.io/node-<card-type>-register"
 ```
 
 Look for annotations containing device information in the node status.
@@ -149,6 +159,6 @@ Look for annotations containing device information in the node status.
 
 2. Virtualization takes effect only for containers that apply for one GPU(i.e iluvatar.ai/vgpu=1 ). When requesting multiple GPUs, the system will automatically set the core resources based on the number of GPUs requested.
 
-3. The `iluvatar.ai/vcuda-memory` resource is only effective when `iluvatar.ai/vgpu=1`.
+3. The `iluvatar.ai/<card-type>.vMem` resource is only effective when `iluvatar.ai/<card-type>-vgpu=1`.
 
-4. Multi-device requests (`iluvatar.ai/vgpu > 1`) do not support vGPU mode.
+4. Multi-device requests (`iluvatar.ai/<card-type>-vgpu= > 1`) do not support vGPU mode.

docs/userguide/Iluvatar-device/examples/allocate-BI-V150.md

@@ -0,0 +1,39 @@
+---
+title: Allocate BI-V150 slice
+---
+
+To allocate core and memory resources for the container, you only need to configure a certain size of GPU core `iluvatar.ai/BI-V150.vCore` and GPU memory resource `iluvatar.ai/BI-V150.vMem`.
+
+```yaml
+apiVersion: v1
+kind: Pod
+metadata:
+  name: BI-V150-poddemo
+spec:
+  restartPolicy: Never
+  containers:
+  - name: BI-V150-poddemo
+    image: registry.iluvatar.com.cn:10443/saas/mr-bi150-4.3.0-x86-ubuntu22.04-py3.10-base-base:v1.0
+    command:
+    - bash
+    args:
+    - -c
+    - |
+      set -ex
+      echo "export LD_LIBRARY_PATH=/usr/local/corex/lib64:$LD_LIBRARY_PATH">> /root/.bashrc
+      cp -f /usr/local/iluvatar/lib64/libcuda.* /usr/local/corex/lib64/
+      cp -f /usr/local/iluvatar/lib64/libixml.* /usr/local/corex/lib64/
+      source /root/.bashrc
+      sleep 360000
+    resources:
+      requests:
+        iluvatar.ai/BI-V150-vgpu: 1
+        iluvatar.ai/BI-V150.vCore: 50
+        iluvatar.ai/BI-V150.vMem: 64
+      limits:
+        iluvatar.ai/BI-V150-vgpu: 1
+        iluvatar.ai/BI-V150.vCore: 50
+        iluvatar.ai/BI-V150.vMem: 64
+```
+
+> **NOTE:** *Each `iluvatar.ai/<card-type>.vCore` unit represents 1% of an available compute core, and each `iluvatar.ai/<card-type>.vMem` unit represents 256MB of device memory*

docs/userguide/Iluvatar-device/examples/allocate-MR-V100.md

@@ -0,0 +1,40 @@
+---
+title: Allocate MR-V100 slice
+translated: true
+---
+
+To allocate core and memory resources for the container, you only need to configure a certain size of GPU core `iluvatar.ai/MR-V100.vCore` and GPU memory resource `iluvatar.ai/MR-V100.vMem`.
+
+```yaml
+apiVersion: v1
+kind: Pod
+metadata:
+  name: MR-V100-poddemo
+spec:
+  restartPolicy: Never
+  containers:
+  - name: MR-V100-poddemo
+    image: registry.iluvatar.com.cn:10443/saas/mr-bi150-4.3.0-x86-ubuntu22.04-py3.10-base-base:v1.0
+    command: 
+    - bash
+    args:
+    - -c
+    - |
+      set -ex
+      echo "export LD_LIBRARY_PATH=/usr/local/corex/lib64:$LD_LIBRARY_PATH">> /root/.bashrc
+      cp -f /usr/local/iluvatar/lib64/libcuda.* /usr/local/corex/lib64/
+      cp -f /usr/local/iluvatar/lib64/libixml.* /usr/local/corex/lib64/
+      source /root/.bashrc
+      sleep 360000
+    resources:
+      requests:
+        iluvatar.ai/MR-V100-vgpu: 1
+        iluvatar.ai/MR-V100.vCore: 50
+        iluvatar.ai/MR-V100.vMem: 64
+      limits:
+        iluvatar.ai/MR-V100-vgpu: 1
+        iluvatar.ai/MR-V100.vCore: 50
+        iluvatar.ai/MR-V100.vMem: 64
+```
+
+> **NOTE:** *Each `iluvatar.ai/<card-type>.vCore` unit represents 1% of an available compute core, and each `iluvatar.ai/<card-type>.vMem` unit represents 256MB of device memory*

docs/userguide/Iluvatar-device/examples/allocate-device-core-and-memory-to-container.md

@@ -0,0 +1,34 @@
+---
+title: Allocate exclusive BI-V150 device
+---
+
+To allocate multiple BI-V150 devices, you only need to assign `iluvatar.ai/BI-V150-vgpu` with no other fields required.
+
+```yaml
+apiVersion: v1
+kind: Pod
+metadata:
+  name: BI-V150-poddemo
+spec:
+  restartPolicy: Never
+  containers:
+  - name: BI-V150-poddemo
+    image: registry.iluvatar.com.cn:10443/saas/mr-bi150-4.3.0-x86-ubuntu22.04-py3.10-base-base:v1.0
+    command: 
+    - bash
+    args:
+    - -c
+    - |
+      set -ex
+      echo "export LD_LIBRARY_PATH=/usr/local/corex/lib64:$LD_LIBRARY_PATH">> /root/.bashrc
+      cp -f /usr/local/iluvatar/lib64/libcuda.* /usr/local/corex/lib64/
+      cp -f /usr/local/iluvatar/lib64/libixml.* /usr/local/corex/lib64/
+      source /root/.bashrc
+      sleep 360000
+    resources:
+      requests:
+        iluvatar.ai/BI-V150-vgpu: 2
+      limits:
+        iluvatar.ai/BI-V150-vgpu: 2
+```
+> **Note:** *When applying for exclusive use of a GPU, `iluvatar.ai/<card-type>-vgpu=1`, you need to set the values ​​of `iluvatar.ai/<card-type>.vCore` and `iluvatar.ai/<card-type>.vMem` to the maximum number of GPU resources. `iluvatar.ai/<card-type>-vgpu>1` no longer supports the vGPU function, so you don't need to fill in the core and memory values*

docs/userguide/Iluvatar-device/examples/allocate-exclusive-BI-V150.md

@@ -0,0 +1,34 @@
+---
+title: Allocate exclusive BI-V100 device
+---
+
+To allocate multiple BI-V100 devices, you only need to assign `iluvatar.ai/BI-V150-vgpu` with no other fields required.
+
+```yaml
+apiVersion: v1
+kind: Pod
+metadata:
+  name: MR-V100-poddemo
+spec:
+  restartPolicy: Never
+  containers:
+  - name: MR-V100-poddemo
+    image: registry.iluvatar.com.cn:10443/saas/mr-bi150-4.3.0-x86-ubuntu22.04-py3.10-base-base:v1.0
+    command: 
+    - bash
+    args:
+    - -c
+    - |
+      set -ex
+      echo "export LD_LIBRARY_PATH=/usr/local/corex/lib64:$LD_LIBRARY_PATH">> /root/.bashrc
+      cp -f /usr/local/iluvatar/lib64/libcuda.* /usr/local/corex/lib64/
+      cp -f /usr/local/iluvatar/lib64/libixml.* /usr/local/corex/lib64/
+      source /root/.bashrc
+      sleep 360000
+    resources:
+      requests:
+        iluvatar.ai/MR-V100-vgpu: 2
+      limits:
+        iluvatar.ai/MR-V100-vgpu: 2
+```
+> **Note:** *When applying for exclusive use of a GPU, `iluvatar.ai/<card-type>-vgpu=1`, you need to set the values ​​of `iluvatar.ai/<card-type>.vCore` and `iluvatar.ai/<card-type>.vMem` to the maximum number of GPU resources. `iluvatar.ai/<card-type>-vgpu>1` no longer supports the vGPU function, so you don't need to fill in the core and memory values*