这是驱动程序包中包含的文档。
在vSphere上，您无法指定每个VM的放置位置，但您可以确定放置策略。
默认是你目前所拥有的，替代选项是加载每个GPU直到它完全加载，然后移动到下一个。
在这种情况下，所有4个VM都将位于同一GPU上。
设置它，
在每个ESX主机上，将“vGPU.consolidation = true”添加到文件/ etc / vmware / config
https://en.wikipedia.org/wiki/PLX_Technology
以上来自于谷歌翻译
以下为原文
This is in the documentation included in the driver bundle.
On vSphere you cannot specify where each VM is placed, but you can determine the placement policy. Default is what you have at present, the alternate option is to load up each GPU until it's fully loaded, then move to the next. In that instance, all 4 VM's would be on the same GPU.
To set it,
On each ESX host, add "vGPU.consolidation = true" to file /etc/vmware/config
https://en.wikipedia.org/wiki/PLX_Technology
这是驱动程序包中包含的文档。
在vSphere上，您无法指定每个VM的放置位置，但您可以确定放置策略。
默认是你目前所拥有的，替代选项是加载每个GPU直到它完全加载，然后移动到下一个。
在这种情况下，所有4个VM都将位于同一GPU上。
设置它，
在每个ESX主机上，将“vGPU.consolidation = true”添加到文件/ etc / vmware / config
https://en.wikipedia.org/wiki/PLX_Technology
以上来自于谷歌翻译
以下为原文
This is in the documentation included in the driver bundle.
On vSphere you cannot specify where each VM is placed, but you can determine the placement policy. Default is what you have at present, the alternate option is to load up each GPU until it's fully loaded, then move to the next. In that instance, all 4 VM's would be on the same GPU.
To set it,
On each ESX host, add "vGPU.consolidation = true" to file /etc/vmware/config
https://en.wikipedia.org/wiki/PLX_Technology
举报
