17:01:53 <colindixon> #startmeeting tws
17:01:54 <odl_meetbot> Meeting started Mon Aug 10 17:01:53 2015 UTC.  The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html.
17:01:54 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:01:54 <odl_meetbot> The meeting name has been set to 'tws'
17:02:00 <colindixon> #topic agenda bashing
17:02:08 <alagalah> colindixon: Are you able to use the webex?
17:02:12 <colindixon> #chair alagalah tbachman dfarrell07
17:02:12 <odl_meetbot> Current chairs: alagalah colindixon dfarrell07 tbachman
17:02:15 <alagalah> colindixon: As soon as I reclaimed host it booted me ?
17:02:15 <colindixon> alagalah: yeah, I’m on
17:02:27 <alagalah> colindixon: Can you record ?
17:02:30 <colindixon> oh, yeah… that means somebody else has it :-(
17:02:33 <abhijitkumbhare> alagalah: why did it boot you?
17:02:42 <adetalhouet> colindixon, I do have it
17:02:42 <alagalah> abhijitkumbhare: Read above
17:02:56 <alagalah> adetalhouet: For ?
17:03:16 <adetalhouet> for nothing, it gave it to me by default I guess
17:03:28 <colindixon> alagalah: I have it
17:03:47 <abhijitkumbhare> we have ended our meeting alagalah - can you see if you can reclaim the host role (ideally it should not be related)
17:03:52 <colindixon> alagalah: we are recording
17:03:56 <colindixon> alagalah: you ahve the ball
17:04:53 * colindixon will try to take notes, but would love help if people are around
17:05:03 <alagalah> #topic https://wiki.opendaylight.org/view/Tech_Work_Stream:Main
17:05:07 <alagalah> #link https://wiki.opendaylight.org/view/Tech_Work_Stream:Main
17:05:10 <colindixon> #undo
17:05:10 <odl_meetbot> Removing item from minutes: <MeetBot.ircmeeting.items.Link object at 0x189f190>
17:05:11 <colindixon> #undo
17:05:11 <odl_meetbot> Removing item from minutes: <MeetBot.ircmeeting.items.Topic object at 0x1cccf10>
17:05:17 <colindixon> #link https://wiki.opendaylight.org/view/Tech_Work_Stream:Main the TWS main page
17:05:19 <alagalah> https://lists.opendaylight.org/pipermail/discuss/2015-August/005478.html
17:05:48 <abhijitkumbhare> I will be skipping the meeting today - but let me know if you guys still have an issue with the webex - not that I know exactly how to help in the webex issues :)
17:05:58 <colindixon> #link https://lists.opendaylight.org/pipermail/discuss/2015-August/005478.html this is the topic alagalah proposed for today and nobody else had anything, so here we are
17:07:20 <colindixon> #info the 3 proposed topics were: (1) app coexitence via layered/stacked pipelines, (2) growing a proper overlay layer that can meet the needs of other layers, e.g., Policy and SFC, (3) Honeycomb: building a disgtributed ODL agend out of ODL parts that could run on the local server
17:08:03 <colindixon> #info these topics came out of the summit and edwarnicke shares out the slides
17:08:35 <edwarnicke> #link https://docs.google.com/presentation/d/1hPqNVgrvTDbgw0FmMdQ6nEWve6D7fYYCrtPz2qYztEM/edit#slide=id.g63c24fb28_0_0
17:09:16 <colindixon> #info edwarnicke also notes that these are somewhat different than prior efforts as they don’t generally fit inside a single project and fall more cross-project
17:09:32 <colindixon> #topic app coexistence through a stack/layer approach
17:09:50 <colindixon> #info edwarnicke says when it comes to OpenFlow coexistence we have a series of approaches
17:10:15 <colindixon> #info a common one is the global arbiter that says yes/no and/or modifies flows, edwarnicke doesn’t sense an appetite for this from OF app devs
17:10:43 <colindixon> #info another approach was used by GBP/SFC in Lithium, was a light-handoff between different functions (essentially passing between tables)
17:11:21 <colindixon> #info in networking, we generally succeed with layers, edwarnicke has one idea of three layers with policy over SFC over Overlay
17:12:42 <colindixon> #info edwarnicke says there are a few things that need to be worked out: (1) how do handoffs actually happen and (2) how we allocate tables to logical functions
17:13:32 <colindixon> #info alagalah asks if we want to try to solve this in the general case for OpenFlow or for OVS specifically
17:14:10 <colindixon> #info jan medved says that this *has* to be OVS-specific because the appraoch of put nearly anything in any table and create tables as you see fit just doesn’t exist in h/w switches
17:15:12 <colindixon> #info Uri suggests that we do go along the lines of investigating how to work with hardware even if over the period of years
17:17:08 <colindixon> #info colindixon brings up two clarifications: first are we asking h/w OF vs. s/w OF or OVS vs. non-OVS?
17:17:21 <colindixon> #undo
17:17:21 <odl_meetbot> Removing item from minutes: <MeetBot.ircmeeting.items.Info object at 0x192eb90>
17:18:01 <colindixon> #info colindixon asks are we asking h/w OF vs. s/w OF or OVS OF vs. non-OVS OF?
17:19:26 <colindixon> #info edwarnicke says that from his memory of cpqd was that it stayed very close the OF spec, apps tended to require OVS-specific things like registers, and cpqd was a pain in the ass to get running at all
17:19:51 <colindixon> #info Uri says that we should focus on one thing for the short term to get things working, but again keeping in mind working more broad would help
17:20:24 <colindixon> #info edwarnicke’s notes on cpqd were in response to Jan’s question if other vSwitches would just work with OVS, the answer is seemingly not easily
17:23:07 <colindixon> #info alagalah says that tony had talked about an idea that would translate a general pipeline into more h/w or vSwitch-specific rules
17:24:23 <colindixon> #info there’s a lot of discussion TTPs as way to handle diverse switch table formats, but maybe in the longer run
17:31:35 <alagalah> #action alagalah to reschedule followup for OpenFlow App co-existance... key thing to resolve: general vs OVS solution.
17:31:38 <colindixon> #topic overlay manager layer
17:32:18 <colindixon> #info this stems from the issue that every app/layer (or at least many of them) are managing their own overlays complete with tunnels and everything else
17:32:29 <alagalah> colindixon: FYI folks, I'd like to save 5min at the end of this to schedule next week (also prompt me to send out an email calling for topics :D) and colindixon I am out next week
17:33:31 <colindixon> #info edwarnicke’s idea is to first have a way to register a key that points to an overlay topology, a key could be an IP subnet, VLAN, etc.
17:34:42 <colindixon> #info in effect this is what the LISP mapping service in OpenDaylight already does
17:36:04 <colindixon> #info LuisGomez says that this is too hard for him to follow without an example
17:36:13 <colindixon> #info vishnoianil asks what the problem statement is
17:37:41 <colindixon> #info edwarnicke says this stems from gettting SFC and GBP to work together and it comes down to two problems: (1) how do we pass things back and forth between apps, e.g., SFC and GBP, (2) handling exit from chains to the next handler
17:39:54 <colindixon> #info vishnoianil says shouldn’t the exit be handled by that layer, but edwarnicke says that it’s no the way ti works, right now SFC doesn’t handle egress forwarding and maybe shoudln’t since it woudl require it to be nearly omniscient
17:41:10 <colindixon> #info alagalah adds that the SFC spec deosn’t specify egress filters, and so you need some logical equivalent to handle dealing with egress
17:42:08 <colindixon> #info alagalah says that it has to deal with *all* corner cases otherwise the amount of state that the consumers need to maintain may make this harder to use rather than not use
17:44:05 <colindixon> #info LuisGomez asks if this will work in a generic OpenFlow network, colindixon says that there was some discussion, but the zeitgeist seemed to be that we’d focus on OVS to actually get things done
17:44:13 <colindixon> #topic Honeycomb
17:45:08 <colindixon> #info edwarnicke says the problem statement here is that we do a lot of handling packet_ins, but forwarding packet_ins to a controller is expensive and making global decisions on packet_ins is even more so
17:45:29 <colindixon> #info instead, we could run a stripped down version of ODL on the servers by the vSwitches to handle most packet_in events locally
17:45:54 <colindixon> #info this “looks like” it would scale much better and might open up other alternative design points to OpenDaylight
17:47:08 <colindixon> #info Uri says we need to look at what should go in the stripped down ODL instance, what footprint it should have, how it performs, etc.
17:50:23 <colindixon> #info colindixon says the two big things he worries about are: (1) the engineering challenges to effectively build an ODL unikernel for ODL apps and (2) what does this buy us in terms of our creating a custom ODL to stripped-down ODL
17:50:46 <colindixon> #info on the first one, we desperately need to see how well we can scale it down while still doing useful things
17:51:40 <colindixon> #info on the second one, desiging the protocol between ODL and stripped-down ODL so that it’s not so “fat” that interactions bottleneck performance, but not so “thin” that it’s negligibly better than what your replacing, e.g., packet_ins
17:53:35 <colindixon> #info jan points out that resource utilization and performance are different things and that we have pretty impressive performacne numbers
17:53:53 <colindixon> #info vina_ermagan asks for jan’s numbers, jan says he’ll get her scripts and test setups to replicate them
17:55:06 <colindixon> #topic future meetings
17:55:47 <colindixon> #info it’s looking like there will be a meeting on Thursday morning (and future Thursday mornings)
17:56:25 <colindixon> edwarnicke: do we have a time?
17:56:35 <edwarnicke> 7am PT
17:56:44 <colindixon> edwarnicke: is there a link?
17:56:44 <edwarnicke> I'll email out the meeting shortly
17:56:47 <colindixon> ok
17:56:48 <colindixon> cool
17:57:01 <colindixon> #info the meeting will be at 7a pacific time on thursday
17:57:03 <alagalah> edwarnicke: Can we get a time out please ?
17:57:10 <alagalah> edwarnicke: few minutes to go...
17:58:10 <colindixon> #action edwarnicke will send an invite for the Thursday 7a pacific follow-up meeting over e-mail soon
17:59:49 <colindixon> #action colindixon to try to find some topic for next week, hopefully user stories, maybe intern projects, maybe Jan’s performance numbers, etc.
18:00:53 <colindixon> #info vishnoianil asks for some kind of blueprint and/ro comprehensible writeup that we can start to work out details on
18:01:35 <colindixon> #endmeeting