On 5/11/2017 1:44 PM, Georg Kunz wrote:
Nevertheless, one concrete thing which came to my mind, is this proposed
improvement of the interaction between Nova and Neutron:
https://review.openstack.org/#/c/390513/
In a nutshell, the idea is that Neutron adds more information to a port
object so that Nova does not need to make multiple calls to Neutron to
collect all required networking information. It seems to have stalled
for the time being, but bringing forward the edge computing use case
might increase the interest again.
Yes we've needed a sort of bulk query capability with the networking API
for years. I started going down a path the other night of trying to see
if we could bulk query floating IPs when building the internal instance
network info cache [1] but it looks like that's not supported. The REST
API docs for Neutron say that you can't OR filter query parameters
together, but at the time looking at the code it seemed like it might be
possible.
Chris Friesen from Wind River has also been looking at some of this
lately, see [2].
But getting people looking at doing performance profiling at scale and
then identifying the major pain points would be really really helpful
for the upstream development team that don't have access to those types
of resources. Then we could prioritize investigating ways to fix those
issues to improve performance.
[1] https://review.openstack.org/#/c/465792/
[2] http://lists.openstack.org/pipermail/openstack-dev/2017-May/117096.html
--
Thanks,
Matt
__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
