<div dir="auto"><div>(Please excuse the top-posting)</div><div dir="auto"><br></div><div dir="auto">The other possibility is that the Cyborg managed devices are plumbed in via IP in guest network space. Then "attach" isn't so much a Nova problem as a Neutron one - probably similar to Manila.</div><div dir="auto"><br></div><div dir="auto">Has the Cyborg team considered a RESTful-API proxy driver, i.e., something that wraps a vendor-specific accelerator service and makes it friendly to a multi-tenant OpenStack cloud? Quantum co-processors might be a compelling example which fit this model.</div><div dir="auto"><br></div><div dir="auto">Cheers,<br><br><div class="gmail_quote" dir="auto"><div dir="ltr">On Sun., 20 May 2018, 23:28 Chris Friesen, <<a href="mailto:chris.friesen@windriver.com">chris.friesen@windriver.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On 05/19/2018 05:58 PM, Blair Bethwaite wrote:<br>
> G'day Jay,<br>
><br>
> On 20 May 2018 at 08:37, Jay Pipes <<a href="mailto:jaypipes@gmail.com" target="_blank" rel="noreferrer">jaypipes@gmail.com</a>> wrote:<br>
>> If it's not the VM or baremetal machine that is using the accelerator, what<br>
>> is?<br>
><br>
> It will be a VM or BM, but I don't think accelerators should be tied<br>
> to the life of a single instance if that isn't technically necessary<br>
> (i.e., they are hot-pluggable devices). I can see plenty of scope for<br>
> use-cases where Cyborg is managing devices that are accessible to<br>
> compute infrastructure via network/fabric (e.g. rCUDA or dedicated<br>
> PCIe fabric). And even in the simple pci passthrough case (vfio or<br>
> mdev) it isn't hard to imagine use-cases for workloads that only need<br>
> an accelerator sometimes.<br>
<br>
Currently nova only supports attach/detach of volumes and network interfaces. <br>
Is Cyborg looking to implement new Compute API operations to support hot <br>
attach/detach of various types of accelerators?<br>
<br>
Chris<br>
<br>
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</blockquote></div></div></div>