17:59:11 <dconde> #startmeeting ODL-GBP-ARCH
17:59:11 <odl_meetbot> Meeting started Fri Jun 13 17:59:11 2014 UTC.  The chair is dconde. Information about MeetBot at http://ci.openstack.org/meetbot.html.
17:59:11 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:59:11 <odl_meetbot> The meeting name has been set to 'odl_gbp_arch'
17:59:24 <dconde> #info Joel is joining to talkabout services today.
18:00:47 <dconde> #info Joel Halpern, to be specific of Ericsson
18:05:05 <dconde> #info we hit limite of 10
18:05:21 <dconde> #info on hangout so we hope we are OK with all participants
18:05:28 <tbachman> #clarification: that’s cigarrettes, folks
18:06:43 <tbachman> we lost Joel
18:08:17 <dconde> #info Joel is giving us the world of services insertion, tc.
18:08:35 <dconde> #info joel is not driving ODL work but he knows what's going on
18:09:01 <dconde> #info realize sfc by using IETF data plane work, NOT the cntrol
18:09:29 <dconde> #info we lost joel
18:09:33 <dconde> #info dialing
18:11:01 <dconde> #info IETF is on data plane.  odl is work is to take requests and drive it. and drive some intrfaces to service chain X, Y, Z
18:11:53 <dconde> dlenrow: we are losing your voice
18:12:49 <dconde> #info SFCin ODL will DRIVE traffic classification rules / pkt handling rules into the edges / clssifiers
18:13:00 <dconde> #if PKT FOO goes into chain BAR to satisfy BAR.
18:13:18 <dconde> #info we need something SFC logic
18:14:24 <dconde> #info there is some overlap
18:14:31 <dconde> #info between two grps
18:18:04 <dconde> #info external to SFC, operators stand up functions and some OSS will drives some service that uses XYZ (FW, Adult content filter, etc.)
18:33:14 <dconde> #info there is an assumption that it drives SB directly.
18:34:01 <dconde> #info GBP wants to leverage SFC work too.
18:35:49 <dconde> #info trying to use PCE, OF, NetConf, etc.  -- build a logic to use existing SB plug ins
18:36:44 <dconde> #info is OK to put an indirection layer within SFC
18:37:11 <dconde> #info but we ahve a collosion problem to deal with
18:43:42 <dconde> #info two DPs - one is transport level
18:43:51 <dconde> #info that can extend across SF forwarders.
18:43:59 <dconde> #info IEEE VLANS, MPLS
18:44:23 <dconde> #info second layer! header -- under transport that carries services meatadata
18:44:29 <dconde> #info to be used for forwarding
19:00:35 <dconde> #endmeeting