forked from docs/doc-exports
qiujiandong1
bf8cd71a69
Reviewed-by: Eotvos, Oliver <oliver.eotvos@t-systems.com> Co-authored-by: qiujiandong1 <qiujiandong1@huawei.com> Co-committed-by: qiujiandong1 <qiujiandong1@huawei.com>
342 lines
30 KiB
HTML
342 lines
30 KiB
HTML
<a name="cce_02_0415"></a><a name="cce_02_0415"></a>
|
|
|
|
<h1 class="topictitle1">CCE AI Suite (NVIDIA GPU)</h1>
|
|
<div id="body32001227"><div class="section" id="cce_02_0415__en-us_topic_0000001788047466_section5841216112318"><h4 class="sectiontitle">Add-on Overview</h4><p id="cce_02_0415__en-us_topic_0000001788047466_p17232621173315">CCE AI Suite (NVIDIA GPU) is a device management add-on that supports GPUs in containers. To use GPU nodes in a cluster, this add-on must be installed.</p>
|
|
</div>
|
|
<div class="section" id="cce_02_0415__en-us_topic_0000001788047466_section13564204912269"><h4 class="sectiontitle">Add-on Parameters</h4>
|
|
<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="cce_02_0415__en-us_topic_0000001788047466_table7237514152812" frame="border" border="1" rules="all"><caption><b>Table 1 </b>Parameters</caption><thead align="left"><tr id="cce_02_0415__en-us_topic_0000001788047466_row152371514142818"><th align="left" class="cellrowborder" valign="top" width="16.650000000000002%" id="mcps1.3.2.2.2.5.1.1"><p id="cce_02_0415__en-us_topic_0000001788047466_p8237101412810">Parameter</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="17.61%" id="mcps1.3.2.2.2.5.1.2"><p id="cce_02_0415__en-us_topic_0000001788047466_p4237131416284">Mandatory</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="16.919999999999998%" id="mcps1.3.2.2.2.5.1.3"><p id="cce_02_0415__en-us_topic_0000001788047466_p15237171413287">Type</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="48.82%" id="mcps1.3.2.2.2.5.1.4"><p id="cce_02_0415__en-us_topic_0000001788047466_p1237171432815">Description</p>
|
|
</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody><tr id="cce_02_0415__en-us_topic_0000001788047466_row4237101416283"><td class="cellrowborder" valign="top" width="16.650000000000002%" headers="mcps1.3.2.2.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p3237141418281">basic</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="17.61%" headers="mcps1.3.2.2.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p723710140287">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.2.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1237111472817">object</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.2.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1723741413286">Basic add-on configuration parameters</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row5237914182816"><td class="cellrowborder" valign="top" width="16.650000000000002%" headers="mcps1.3.2.2.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p112371214132813">custom</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="17.61%" headers="mcps1.3.2.2.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p192371914182813">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.2.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p4237181416289"><a href="#cce_02_0415__en-us_topic_0000001788047466_table109631116113412">Table 3</a> object</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.2.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p02371514192810">Custom parameters</p>
|
|
</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
</div>
|
|
|
|
<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" id="cce_02_0415__en-us_topic_0000001788047466_table43632037103817" frame="border" border="1" rules="all"><caption><b>Table 2 </b>Configuration of basic</caption><thead align="left"><tr id="cce_02_0415__en-us_topic_0000001788047466_row12363193793812"><th align="left" class="cellrowborder" valign="top" width="25%" id="mcps1.3.2.3.2.5.1.1"><p id="cce_02_0415__en-us_topic_0000001788047466_p9363133713383">Parameter</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="24.98%" id="mcps1.3.2.3.2.5.1.2"><p id="cce_02_0415__en-us_topic_0000001788047466_p14363163713389">Mandatory</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="25.019999999999996%" id="mcps1.3.2.3.2.5.1.3"><p id="cce_02_0415__en-us_topic_0000001788047466_p193637373386">Type</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="25%" id="mcps1.3.2.3.2.5.1.4"><p id="cce_02_0415__en-us_topic_0000001788047466_p1936323713812">Description</p>
|
|
</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody><tr id="cce_02_0415__en-us_topic_0000001788047466_row6363143710382"><td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p03631137143819">cluster_version</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="24.98%" headers="mcps1.3.2.3.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p16363123713814">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25.019999999999996%" headers="mcps1.3.2.3.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p6363113717385">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p83636374386">CCE cluster version</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row9363113718389"><td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p17363837113814">device_version</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="24.98%" headers="mcps1.3.2.3.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p4363113773813">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25.019999999999996%" headers="mcps1.3.2.3.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1836315371384">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p636343733819">Add-on version</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row1536323733820"><td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p113632373384">driver_version</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="24.98%" headers="mcps1.3.2.3.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1336310378384">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25.019999999999996%" headers="mcps1.3.2.3.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p6363837113814">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p136312372380">Image tag of an add-on pod where a driver is installed. Generally, the value is the same as that of <strong id="cce_02_0415__en-us_topic_0000001788047466_b4261285197">device_version</strong>.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row10363173716381"><td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p6363133716385">obs_url</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="24.98%" headers="mcps1.3.2.3.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p0363637113810">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25.019999999999996%" headers="mcps1.3.2.3.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p193636375387">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p13631376381">When a GPU driver is downloaded from the default driver address, the value is the GPU driver address.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row1536323763818"><td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p203631374384">swr_addr</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="24.98%" headers="mcps1.3.2.3.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p336315371382">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25.019999999999996%" headers="mcps1.3.2.3.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1736314370386">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1136310374384">Image repository address</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row499862012395"><td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p10998132033918">swr_user</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="24.98%" headers="mcps1.3.2.3.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p11998132010397">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25.019999999999996%" headers="mcps1.3.2.3.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p9998182063914">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="25%" headers="mcps1.3.2.3.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1899816204391">Tenant path of an image repository</p>
|
|
</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
</div>
|
|
|
|
<div class="tablenoborder"><a name="cce_02_0415__en-us_topic_0000001788047466_table109631116113412"></a><a name="en-us_topic_0000001788047466_table109631116113412"></a><table cellpadding="4" cellspacing="0" summary="" id="cce_02_0415__en-us_topic_0000001788047466_table109631116113412" frame="border" border="1" rules="all"><caption><b>Table 3 </b>Configuration of custom</caption><thead align="left"><tr id="cce_02_0415__en-us_topic_0000001788047466_row1696361663418"><th align="left" class="cellrowborder" valign="top" width="21.75%" id="mcps1.3.2.4.2.5.1.1"><p id="cce_02_0415__en-us_topic_0000001788047466_p79633167341">Parameter</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="12.509999999999998%" id="mcps1.3.2.4.2.5.1.2"><p id="cce_02_0415__en-us_topic_0000001788047466_p139631416173412">Mandatory</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="16.919999999999998%" id="mcps1.3.2.4.2.5.1.3"><p id="cce_02_0415__en-us_topic_0000001788047466_p17963111616348">Type</p>
|
|
</th>
|
|
<th align="left" class="cellrowborder" valign="top" width="48.82%" id="mcps1.3.2.4.2.5.1.4"><p id="cce_02_0415__en-us_topic_0000001788047466_p1696351619347">Description</p>
|
|
</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody><tr id="cce_02_0415__en-us_topic_0000001788047466_row196318165349"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p06393538345">compatible_with_legacy_api</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p169631416103416">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p4963016153419">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1396371683413">API compatibility switch</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p24141822165110">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b4995161511207">false</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p684021781011"><strong id="cce_02_0415__en-us_topic_0000001788047466_b181066208204">true</strong>: The add-on supports the GPU native mode and xGPU virtualization.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row2621435116"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p26213316114">component_schedulername</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p8621341118">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p4629316114">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p662163111114">Name of the scheduler used by the add-on.</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p105956341514">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b158353237218">default-scheduler</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row716715518141"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p6167145121419">disable_mount_path_v1</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p616719531415">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p151671571412">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p9129599524">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b33731117102018">false</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p7167135171413"><strong id="cce_02_0415__en-us_topic_0000001788047466_b63006517265">true</strong>: <strong id="cce_02_0415__en-us_topic_0000001788047466_b466183202717">/opt/cloud/cce/nvidia</strong> is not mounted to the <strong id="cce_02_0415__en-us_topic_0000001788047466_b2491131516278">/usr/lib/nvidia</strong> directory of a GPU container.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row14611402167"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p8744018164">disable_nvidia_gsp</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p471640121615">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p3713407168">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p19784428135210">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b15990175252112">true</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p3714041612"><strong id="cce_02_0415__en-us_topic_0000001788047466_b9567575217">true</strong>: The GPU GSP firmware is disabled.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row17882247171619"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p178821847131614">driver_mount_paths</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p2088224741610">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p5882104712162">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p15882154714167">Driver file directory that needs to be automatically mounted to a GPU container</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p167091298537">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b8984113782217">"bin,lib64"</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row295219112176"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p17953419170">enable_fault_isolation</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p129534116176">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p119535110178">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p399711373534">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b14184165462118">true</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p4953413179"><strong id="cce_02_0415__en-us_topic_0000001788047466_b1125514455223">true</strong>: The add-on detects hardware faults or driver issues of a GPU and then sets the GPU to be unavailable.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row19767781713"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p169767710173">enable_health_monitoring</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p179767718174">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1397647101720">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p73455407585">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b13185205413214">true</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p497677191718"><strong id="cce_02_0415__en-us_topic_0000001788047466_b1029423442318">true</strong>: The add-on detects hardware faults or driver issues of a GPU.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row981910187173"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p5819718111718">enable_metrics_monitoring</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p11819171813172">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p58191918161716">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p16265175011583">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b918616548214">true</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p981911817175"><strong id="cce_02_0415__en-us_topic_0000001788047466_b6863958162319">true</strong>: The add-on collects GPU metrics and reports these metrics to Prometheus.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row36114349174"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p46123441720">enable_simple_lib64_mount</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p861183491719">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1861143412172">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1750151445918">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b518735412111">true</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p2611034111711"><strong id="cce_02_0415__en-us_topic_0000001788047466_b1898722018247">true</strong>: Only the <strong id="cce_02_0415__en-us_topic_0000001788047466_b35141125122419">libxxx.so.x</strong> file is mounted to a container.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row18899144341719"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p389964381712">enable_xgpu</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p19899114371710">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1289934321714">Bool</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p2725123613590">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b937461713206">false</strong></p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p158991943141717">Whether to enable xGPU virtualization.</p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row13261145491715"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p5261135410170">gpu_driver_config</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p112616546173">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1926135401712">Map</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1626165491711">Configurations of the GPU driver for a single node pool</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p6799113216">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b1638530192512">{}</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row191252191814"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1791362121815">health_check_xids_v2</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1691312217186">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1591310281815">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1091382151813">GPU error range for the add-on health checks</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p9472131417113">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b13926131792516">"74,79"</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row1539314191811"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p9391914151814">inject_ld_Library_path</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p839181415189">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p2392144182">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1439141421815">Value of the <strong id="cce_02_0415__en-us_topic_0000001788047466_b1032132914253">LD_LIBRARY_PATH</strong> environment variable automatically injected by the add-on to a GPU container</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p168718405215">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b103914452255">""</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row9626331111812"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p11626173111813">lib64_container_paths</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p962633118181">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1862683119187">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1626113161819">Mount path of NVIDIA lib64 in a GPU container</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p28171225355">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b117336622617">"/usr/lib64,/usr/lib/x86_64-linux-gnu"</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row28241838191812"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p88249383184">metrics_delete_interval</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p2824938121820">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p78244383189">int</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p382443811813">Timeout threshold for deleting a metric when the metric cannot be obtained. The unit is millisecond.</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p52506513620">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b614381716264">30000</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row1949104811187"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p194912481183">metrics_monitor_interval</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p12491184831817">No</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p12491124811187">int</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p7491948111818">Interval for obtaining metrics, in milliseconds.</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p16275152011719">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b74829297267">15000</strong></p>
|
|
</td>
|
|
</tr>
|
|
<tr id="cce_02_0415__en-us_topic_0000001788047466_row4101115512185"><td class="cellrowborder" valign="top" width="21.75%" headers="mcps1.3.2.4.2.5.1.1 "><p id="cce_02_0415__en-us_topic_0000001788047466_p0101105512181">nvidia_driver_download_url</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="12.509999999999998%" headers="mcps1.3.2.4.2.5.1.2 "><p id="cce_02_0415__en-us_topic_0000001788047466_p1510135571814">Yes</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="16.919999999999998%" headers="mcps1.3.2.4.2.5.1.3 "><p id="cce_02_0415__en-us_topic_0000001788047466_p111011755111814">String</p>
|
|
</td>
|
|
<td class="cellrowborder" valign="top" width="48.82%" headers="mcps1.3.2.4.2.5.1.4 "><p id="cce_02_0415__en-us_topic_0000001788047466_p151015553187">Path for downloading the NVIDIA driver</p>
|
|
<p id="cce_02_0415__en-us_topic_0000001788047466_p9277176182">Default value: <strong id="cce_02_0415__en-us_topic_0000001788047466_b11785545182518">""</strong></p>
|
|
</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
</div>
|
|
</div>
|
|
<div class="section" id="cce_02_0415__en-us_topic_0000001788047466_section5790442114119"><h4 class="sectiontitle">Example Request</h4><pre class="screen" id="cce_02_0415__en-us_topic_0000001788047466_screen13762558164116">{
|
|
"kind": "Addon",
|
|
"apiVersion": "v3",
|
|
"metadata": {
|
|
"name": "gpu-beta",
|
|
},
|
|
"spec": {
|
|
"clusterID": "80c9e306-***-***-***-0255ac100043",
|
|
"version": "2.0.69",
|
|
"addonTemplateName": "gpu-beta",
|
|
"values": {
|
|
"basic": {
|
|
"cluster_version": "v1.27",
|
|
"device_version": "2.0.69",
|
|
"driver_version": "2.0.69",
|
|
"obs_url": "***",
|
|
"region": "***",
|
|
"swr_addr": "***",
|
|
"swr_user": "***"
|
|
},
|
|
"custom": {
|
|
"compatible_with_legacy_api": true,
|
|
"component_schedulername": "kube-scheduler",
|
|
"disable_mount_path_v1": false,
|
|
"disable_nvidia_gsp": true,
|
|
"driver_mount_paths": "bin,lib64",
|
|
"enable_fault_isolation": true,
|
|
"enable_health_monitoring": true,
|
|
"enable_metrics_monitoring": true,
|
|
"enable_simple_lib64_mount": true,
|
|
"enable_xgpu": true,
|
|
"gpu_driver_config": {},
|
|
"health_check_xids_v2": "74,79",
|
|
"inject_ld_Library_path": "",
|
|
"lib64_container_paths": "/usr/lib64,/usr/lib/x86_64-linux-gnu",
|
|
"metrics_delete_interval": 30000,
|
|
"metrics_monitor_interval": 15000,
|
|
"nvidia_driver_download_url": ""
|
|
},
|
|
}
|
|
}
|
|
}</pre>
|
|
</div>
|
|
</div>
|
|
<div>
|
|
<div class="familylinks">
|
|
<div class="parentlink"><strong>Parent topic:</strong> <a href="cce_02_0366.html">Add-on Instance Parameters</a></div>
|
|
</div>
|
|
</div>
|
|
|