On 13/12/13 22:40 +0000, Kurt Griffiths wrote: >FWIW, Marconi can easily deliver sub-second latency even with lots of >clients fast-polling. We are also considering a long-polling feature that >will reduce latency further for HTTP clients. And there's already some work going on the `websocket` side[0]. It's still pretty much an experiment since its completion depends on some features that are under development. Targeting i-2. Cheers, FF [0] https://github.com/FlaPer87/marconi-websocket > >On 12/13/13, 2:41 PM, "Fox, Kevin M" <kevin.fox at pnnl.gov> wrote: > >>A second or two of latency perhaps shaved off by using something like >>AMQP or STOMP might not be justified for its added complexity. > > >_______________________________________________ >OpenStack-dev mailing list >OpenStack-dev at lists.openstack.org >http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev -- @flaper87 Flavio Percoco -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 836 bytes Desc: not available URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20131214/62309f60/attachment.pgp>