19:59:59 <dconde> #startmeeting gbp-status 19:59:59 <odl_meetbot> Meeting started Thu May 1 19:59:59 2014 UTC. The chair is dconde. Information about MeetBot at http://ci.openstack.org/meetbot.html. 19:59:59 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:59:59 <odl_meetbot> The meeting name has been set to 'gbp_status' 20:02:18 <dlenrow> What is google hangout link for today please? 20:03:58 <dconde> just a moment 20:04:19 <dconde> #link for hangout is https://plus.google.com/hangouts/_/calendar/ZHZvcmtpbkBub2lyb25ldHdvcmtzLmNvbQ.nf43n3np1b0kc48it41dsg4v2o 20:04:37 <dconde> dlenrow - let me know if that works 20:04:55 <dlenrow> At least it's easy to memorize :-) 20:05:12 <dconde> #chair alagalah 20:05:12 <odl_meetbot> Current chairs: alagalah dconde 20:05:44 <dconde> starting in 5 minutes 20:05:51 <dvorkinista> #table dconde 20:06:12 <mickey_spiegel> #info here 20:06:18 <dvorkinista> #info here 20:06:41 * tbachman #info here 20:07:02 <dconde> #table, chair, sofa 20:07:21 <dconde> #info dconde here 20:07:24 <s3wong> #info here 20:07:52 <dconde> #topic start of meeting 20:07:59 <dconde> #info agenda item 20:08:10 <dconde> who will be committing code 20:08:30 <dconde> #info we see the list of people who will be committed 20:08:37 <dconde> #info anyone else should speak up. 20:08:52 <dconde> #agreed that the list if what we have for now. 20:09:27 <dconde> #info liemmn is 20%, right? but may ramp in future? 20:09:50 <dconde> #info liemmn said starting at less than 20 but may ramp to 40% 20:10:03 <dconde> #info uchau may ramp up in similar way 20:10:34 <dconde> #info Hemanth adds 20% to himself and another TBD from oneConvergence. 20:10:53 <dconde> #info A4, a4 in spreadsheet is a reference to tasks in the spreadsheet. 20:11:11 <dconde> #info concept of dummy renderer. 20:11:19 <dconde> #info we see a diagram on the hangout. 20:12:14 <dconde> #info this idagram is going to change since renderers have two components. we have common store for the 4 boxes. repo, ep reg, enforcement repo, op states. 20:12:45 <dconde> #info benefits in parallel development by creating a dummy renderer that statically responds (like a stub?) same for SB renderer 20:13:12 <dconde> #info it's a stub with a statically defined field. 20:13:34 <dconde> #info if I know the subject,etc. we can stuff the subject into the stub 20:14:14 <dconde> #info we need to work with the scorer for the scalar generation algorithm. Someone in APIC team has done that. So alagalah will contact them for background. 20:14:28 <dconde> #info same for thresholding and acknowledgement 20:14:39 <dconde> #info the boxes have very little intelligence in them. 20:14:49 <dconde> #info the renderer has all the intelligence in them. 20:15:15 <dconde> #topic parallel implementation 20:15:24 <jmedved> there are folks trying to join the hangout and can’t - google says the video call is full… 20:15:40 <dconde> if you cannot get into hangout, please telephone in. 20:16:08 <jmedved> what’s the bridge #? 20:16:11 <dconde> dvorkinista states that renderer … (lost audio?) 20:16:25 <dconde> US/Canada: +1.866.432.9903 United Kingdom: +44.20.8824.0117 20:17:29 <dconde> #info the DSE talk to the Endpoint registry / mapper 20:17:45 <dconde> #info DSE / Governance / Congress /etc. needs to fit in. 20:18:00 <dconde> #info DSE stands for data services engine 20:18:10 <dconde> #info Plexxi guys contributed that to the Congress project. 20:18:36 <dconde> #info DSE pulls out network metadata from other sources 20:18:56 <dconde> it can influence policies and influence groups of endpoints. 20:19:36 <dconde> #info there is a placeholer in the release plan. 20:19:51 <dconde> #info we start with OpenFlow and OpFlex moved into incubation this morning 20:20:03 <dconde> #info so that's a real project. 20:20:58 <dconde> #info we want to implement REnderer common and migrate OF Strawman 20:21:18 <dconde> mickey_spiegel says that we need to figure out the unknown unknown 20:21:41 <dconde> if the renderer is stateless and transofrmational, then we have more restrictions on subscription, 20:21:54 <dlenrow> What subjects we plan to render for needs to be defined in more detail for Helium 20:22:11 <dconde> #info we will talk about that in the ARCH meeting tomorrow. 20:22:27 <dconde> #info we think subscription and rendering is intertwined. 20:22:38 <dconde> #info everyting in column A should be a sheet. 20:22:53 <dconde> #info interdependencies are critical. 20:23:13 <dconde> #topic release plan schedule 20:23:24 <dconde> #info we are looking at gates. 20:23:30 <dconde> #info M1 to M3 is easy 20:23:34 <dconde> #info M4 and 5 are harder 20:24:41 <dconde> #info all things are gated on beginning of a week except we have a conflict with Labor day for code freeze 20:25:24 <dconde> #info probability for working on labor day is low 20:25:36 <dconde> #info dvorkinista reviewing release plan. 20:25:49 <dconde> #info this is aggressive. 20:26:08 <dconde> #info if people at 20% then if you are not familiar with ODL it's hard. 20:26:10 <dlenrow> The problem space for which one can define subjects and associated rendering is nearly infinite. What subset do we foresee delivering? 20:26:50 <dconde> #info we can do stuff like pair work, etc to ramp it up. alagalah says we need to experience first hand if this is sane. 20:27:23 <dconde> #info if the goal is to render everything forever, it's hopeless. 20:27:30 <dconde> #info we need to understand the subset. 20:27:58 <dconde> #info dlenrow is talking about stretch goals, etc. a few subjects to render 20:28:26 <dconde> #info dvorkinista says if we do it well, it's better than something fast that does not work. 20:29:07 <dconde> #info mickey_spiegal says that contracts and such need to be put in early. 20:29:27 <dconde> #info dvorkinista says it needs to be fast too. 20:29:48 <dconde> #info alagalah says a subset is doable. 20:30:18 <dconde> #info dvorkinista 20:30:43 <dconde> #info most affects are on service chaining, but dvorkinista says we can hold off on that but it needs to be done well, and we can wait for that. 20:31:12 <dconde> #info we need to be very careful and not over commit. 20:31:25 <dconde> #topic service chaining 20:31:29 <dconde> Derick is talking 20:31:50 <dconde> #info looking at diagram 20:32:06 <dconde> #info the diagram will be mailed out. 20:32:15 <dconde> #info dvorkinista we need to make sure it is extensible. 20:32:23 <dconde> #info OK to skip functions to edit without rewrite. 20:32:57 <dconde> #info dvorkiinsta not worried about policy. EPR needs to be fast. Renderer is where all crazy behaviour is. 20:33:18 <dconde> #info it seems to make sense for parallel development. 20:33:37 <dconde> #info alagalah says we need to worry about extensibility in renderer. 20:33:58 <dconde> #info we need to describe a policy that governs that. 20:35:10 <dconde> #info all of these things can be subclassed, says dvorkinista 20:35:20 <dconde> #info OF stuff is down below, says mickey_spiegel 20:36:15 <dconde> #info this is in the OF render, not part of the common renderer. 20:37:00 <dconde> #info dvorkiista says it's another layer of abstraction. OpenFlow based, affinity, etc. 20:37:34 <dconde> #topic renderer 20:37:47 <dconde> #info we are redrawing diagram. 20:38:01 <dconde> #info adding OF renderer below Flow based renderer 20:38:54 <dconde> #info idea to make the dummy object is statically defined and we invoke this into flows. 20:39:22 <dconde> #info flow based rendering is not stateless. 20:39:58 <dconde> #info render common is constantly rendering. Are we worried about passing vs. storing. 20:41:29 <dconde> #info we talked about schedule. 20:41:35 <dconde> #topic recap 20:41:42 <dconde> #info is this a sane plan? 20:43:16 <dconde> #info keep everything equal. and statically define it. 20:43:47 <dconde> #info readams asks.does the dummy version need to be created first? 20:44:26 <dconde> #info readams says we cannot make progress until model is defined. 20:45:22 <dconde> #info readams what is A5 for renderer? 20:46:09 <dconde> #info alagalah asks which parts do you want to contrib to? 20:46:21 <dconde> #topic call fo volunteers 20:46:31 <dconde> #topic call for volunteers 20:47:34 <dconde> #topic we have a proposal now. 20:47:52 <dconde> #info liemnm how do we do tests? 20:48:40 <dconde> #info we do have dependencies -- says lenrow 20:48:50 <dconde> #info alagalah UI extensions are unknown. 20:49:18 <dconde> #ifo DLUX is horizon port to ODL 20:50:05 <dconde> #info we need to look at QOS, openflow 1.3 metering impl, etc. 20:50:11 <dconde> #info this may limit what labels we can use. 20:50:31 <dconde> #action alagalah to follow on edwarnicke ODL plans 20:51:02 <dvorkinista> dconde: why would it affect labels? This is insane! 20:51:35 <dconde> OK. I think I mis-scribed. 20:52:11 <dconde> dvorkinista the cross talk got me rendering improperly on IRC 20:52:38 <dvorkinista> dconde: it's all good 20:52:51 <dconde> #info test dev platform == mininet. but ideally we need real set-up. 20:53:20 <dconde> #info tbachman to document the ovsdb / openstack / ovs dev environment 20:53:28 <dconde> #info also for openflow on mininet. 20:53:57 <tbachman> lol -- right when alagalah mentions my name, ghangout drops on me! 20:53:58 <dconde> #info for OpFlex, it's nascent, so we need to figure out. probably mininet too,b ut it may be ovs switches directly on Linux. Who knows. 20:54:44 <dconde> #info spredsheet was emailed out. 20:55:10 <dconde> #info sharing is failing. 20:56:13 <dconde> #info sharing of gDocs done. 20:56:53 <dconde> #topic anything else? 20:57:22 <dconde> #info monday COB has the final commitment. 20:57:48 <dconde> dlenrow says ghangout max was reached. 20:59:07 <dconde> #info if you cannot get on ghangout then go to IRC and say you cannot get on, and then we can fire off a WebEx. 20:59:40 <dconde> #info of use YouTube live to stream video. 20:59:53 <dconde> #endmeeting