Open Stack

On 13 Aug 2019, at 18:19, Thierry Carrez <thierry at openstack.org<mailto:thierry at openstack.org>> wrote:

Tim Bell wrote:
We’re also thinking about starting a “Large Scale” SIG so people could
collaborate in tackling more of the scaling issues together. Thierry
Carrez (ttx) and I will be looking into that by mentioning the idea to
LINE and YahooJapan (as some perspective at-scale operators)to see
what they think and also make a list of organizations that could be
interested. Rico Lin (ricolin) will also update the SIG guidelines
documents to make the whole process easier and Jim Rollenhagen (jroll)
will try and bring this up at Verizon Media.

How about a forum brainstorm in Shanghai ?

Yes, my goal is to get a number of interested stakeholders to meet in
Shanghai and see if there is enough alignment in goals to openly
collaborate on those questions.

We have had several groups tackling facets of the "large scale" problem
in the past (performance WG, large deployments workgroup, LCOO...) -- my
thinking here would be to narrow it down to addressing scaling
limitations in cluster sizes (think: RabbitMQ falling down after a given
number of compute nodes).


We’ll have some CERN people in Shanghai and would be happy to participate. We’re following up on a number of scaling issues (such as https://techblog.web.cern.ch/techblog/post/nova-ironic-at-scale/ and Neutron) and would be happy to share best practise with other large deployments.

Tim
--
Thierry Carrez (ttx)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-discuss/attachments/20190813/04af808e/attachment-0001.html>