17:00:10 <colindixon> #startmeeting tws
17:00:11 <odl_meetbot> Meeting started Mon Aug 17 17:00:10 2015 UTC.  The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html.
17:00:11 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:11 <odl_meetbot> The meeting name has been set to 'tws'
17:00:59 <colindixon> #topic agenda bashing
17:01:19 <colindixon> #into today’s main topic is covering some of the user story work that’s been done
17:01:29 <colindixon> #link https://wiki.opendaylight.org/view/Tech_Work_Stream:Main#Upcoming_Meeting_Agendas
17:01:57 <colindixon> #info if people have ideas for future meetings, feel free to post them there or bring them up during the beginning/end of TWS calls
17:02:18 <colindixon> #chair phrobb tbachman
17:02:18 <odl_meetbot> Current chairs: colindixon phrobb tbachman
17:05:37 <colindixon> #info the other topic for today is maros and jmedved on the performance of NETCONF
17:05:43 <colindixon> julim: are you having audio issues?
17:05:56 <julim> colindixon - having trouble connecting to webex
17:06:04 <nyechiel> colindixon: not able to dial
17:06:04 <julim> meeting number code / access code invalid
17:06:05 <colindixon> julim: anything we can do to help?
17:06:12 <colindixon> Meeting number: 196 592 514
17:06:16 <colindixon> huh
17:06:28 <julim> I tried that mtg # with both toll-free and toll # but it does not work
17:06:28 <colindixon> I’m  in using VoIP
17:06:36 <julim> webex still setting up (for 10 min now...)
17:06:52 <nyechiel> same for me. Can't use WebEx for audio on Linux...
17:08:30 <colindixon> julim: can you have it call you?
17:08:31 <julim> colindixon - is there an international (israel) #?
17:08:38 <julim> I can't get webex even on.
17:08:53 <julim> I have to reboot I think as something got hung. brb in 3 min
17:09:07 <colindixon> you’re using these numbers: +1-855-797-9485 (US Toll free)
17:09:11 <colindixon> +1-415-655-0002 (US Toll)
17:09:29 <julim> will try #
17:09:45 <julim> ^^ nyechiel -- see new #'s from colindixon (different from website)
17:09:58 <colindixon> #topic netconf performance
17:10:38 <julim> colindixon - I managed to dial-in
17:10:41 <julim> just no webex
17:10:45 <colindixon> julim: thanks!
17:10:47 <colindixon> it’s all good
17:10:52 <nyechiel> colindixon, julim: I am in as well, thanks!
17:10:53 <colindixon> if you’re just going to present the wiki, that works fine
17:10:59 <julim> awesome nyechiel!
17:11:06 <julim> we'll use wiki.
17:11:10 <nyechiel> ack
17:11:12 <colindixon> nyechiel and julim: I’ll give you the ball back around half past
17:11:21 <julim> sounds good. thx colindixon
17:11:23 <colindixon> #info netconf scale testing
17:12:13 <colindixon> #info connecting to 10,000 devices, it took about 15 minutes to connect to every device, then 5-6 inutes to make an RPC call to each device
17:12:23 <colindixon> #info all tests using 4 core CPU with 6G of RAM for ODL
17:12:36 <colindixon> #info the above was with the clustered data store
17:12:55 <colindixon> #info with the in-memory data store, it took ~13 minutes (vs. 15 to do the same thing)
17:13:31 <colindixon> #info can mount 12-12.5k devices total with 4G of RAM
17:13:57 <colindixon> #info colindixon asks where these scripts are running, maros says so far private, but working on getting them into integration
17:14:09 <colindixon> #info netconf peformance testing
17:14:26 <colindixon> #info netconf-test-client => netconf server in ODL => MD-SAL’s data store (in-memory)
17:14:34 <colindixon> #info running on 16 core, 32G RAM machine
17:14:58 <colindixon> #info data writes are simple L2-FIB-style data
17:15:31 <colindixon> #info without baching, we get to 2400 writes/sec sync and 5600 writes/sec async
17:16:04 <colindixon> #info could get more than 40,000 writes/sec with async and batching 1000 writes/sec
17:17:02 <colindixon> #info jmedved points out that this is vs. ~1000-1500 writes/sec in OpenFlow
17:17:19 <colindixon> #info colindixon asks why that’s the case since they should both be just writing to the data sore
17:18:15 <colindixon> #info jmedved says part of that was that with openflow it’s a write, read, and then write, with batching the number goes 10-12k writes per second even with openflow, which begins to look not too dissimilar
17:18:36 <colindixon> #info with multiple clients and batching, it goes as high as 130,000+ writes/second
17:19:34 <colindixon> #info how many L2-FIB-link entries could be stored in limited memory something like 1.4 million decreasing to 300k as you go from 2G RAM to 500M RAM
17:19:46 <colindixon> #info netconf notification performance
17:20:11 <colindixon> #info notifications are basically a summary of a few routes
17:21:17 <colindixon> #info notifications/second range from 3k notifications/second to 22k notifications/second (increase due to smaller size and reduced complexity)
17:21:28 <colindixon> #info end-to-end netconf performance
17:21:59 <colindixon> #info restconf-test-client => restconf => MD-SAL binding aware app => netconf connector => netconf device
17:22:26 <colindixon> #info simulated netconf device was parsing data and throwing it way
17:22:33 <colindixon> #info data was also a list of routes
17:23:19 <colindixon> #info ranging from 600 routes/second without batching, multiple clients, or asynchrony to 6000 routes/second with things tuned up
17:24:00 <colindixon> #info this went up to 180,000 routes/second when moving to something avoiding restconf, restconf seems to be what’s slowling things down
17:24:24 <colindixon> #info jmedved asks when memory constrained, what was the peformance, maros says it was pretty fast until it got to ~90% of capacity
17:24:37 <colindixon> #info jmedved points out that this would be useful for the honeycomb ideas
17:27:26 <colindixon> #info colindixon asks about memory efficiency, it looks like we use ~1.5G RAM to store ~1,000,000 extra L2 FIB entries, which is ~1.5KB/entry, which isn’t great, but probably OK of Java
17:28:15 <colindixon> #info jmedved asks rovarga what could be done to further optimize memory usage, rovarga said we could use custom data structures or wait for Java 9 and some of it’s features
17:31:07 <colindixon> #Info points out we should look into this more deeply for honeycomb and what the trade-offs
17:32:06 <colindixon> #info there are some discussions about the relative reasons for openflow vs. netfconf performance differences
17:32:25 <colindixon> #info LuisGomez asks if it will get into ODL CI,  Maros says yes that’s the plan
17:34:17 <colindixon> #info Art says he’d like to put out tutorials of how NETCONF would work so he can help out, would like somebody to reach out to him
17:34:30 <colindixon> #info jmedved and Art will connect
17:34:39 <colindixon> #topic user stories
17:35:16 <julim> https://wiki.opendaylight.org/view/UserStories:Main
17:35:29 <colindixon> #link https://wiki.opendaylight.org/view/UserStories:Main the user stories we’re going to go through
17:36:15 <colindixon> julim and nyechiel: I’m presenting the page so people can see it over webex too
17:36:58 <colindixon> #info julim says that this is a list of user stories around OpenDaylight and OpenStack primarily, there are a variety of roles defined
17:37:41 <colindixon> #Info each user story is a list of scenarios that are either not yet covered by opendaylight or only partially covered
17:37:42 <julim> the fuller list of user stories (backlog) can be seen at https://wiki.opendaylight.org/view/File:ODL_User_Stories.pdf
17:37:58 <colindixon> #info nyechiel says these are the basic table stakes for working with OpenStack
17:38:14 <colindixon> #link https://wiki.opendaylight.org/view/File:ODL_User_Stories.pdf this is the full backlog of user stories that they have put together
17:39:32 <colindixon> #info edwarnicke asks about the first one, he was under the impression that security groups were working
17:39:49 <colindixon> #info edwarnicke asks they to check with group-based policy working
17:40:10 <colindixon> #info nyechiel points out that so far this is mostly looking at OVSDB, no GBP, but he’ll look at that
17:40:42 <colindixon> #info edwarnicke notes that he’s pretty sure that GBP supports security groups and IPv6 security groups fully working
17:43:15 <colindixon> #info edwarnicke believes everything here is working insofar as he knows except for DHCP/IPAM and conntracking features that require new OVS
17:43:55 <colindixon> #info nyechiel and julim say that this isn’t so much about these items, although they might be useful, but more that this is likely a better model for how to set priorities and focus in OpenDaylight
17:43:57 <julim> edwarnicke - pls. feel free to update the status column about things you believe are completed or are outstanding
17:44:08 <edwarnicke> julim: Link?
17:44:24 <julim> edwarnicke - https://wiki.opendaylight.org/view/UserStories:Main
17:44:32 <edwarnicke> julim: Many thanks :)
17:44:37 <julim> Thank you edwarnicke
17:44:40 <colindixon> #info nyechiel and julim also point out that if we want tration with the openstack community (and their users) this is likely a good approach
17:45:14 <colindixon> #link https://wiki.opendaylight.org/view/UserStories:Explanation put some really basic information about how to appraoch things with user stories
17:45:18 <colindixon> #undo
17:45:18 <odl_meetbot> Removing item from minutes: <MeetBot.ircmeeting.items.Link object at 0x19c1bd0>
17:45:24 <colindixon> #link https://wiki.opendaylight.org/view/UserStories:Explanation put some really basic information about how to appraoch things with user stories (courtesy of julim)
17:45:45 <colindixon> #info Prem_ asks how this changes with the big tent approach in OpenStack
17:46:24 <colindixon> #info nyechiel says that’s a longer discussion, but he doesn’t thing we’re far enough along in our user story planning for that to dramatically affect us
17:50:05 <colindixon> #info art says he really likes this, but he’d love to see this drift even further up the stack from feature-level things to even more use cases
17:50:16 <colindixon> #info julim says that soudns great, and ask for any next steps
17:50:35 <colindixon> #info art says he is trying to come up with a first stab at well-define use caes and is hoping for feedback soon
17:51:05 <colindixon> #info dneary asks if there are any high-priority use cases missing from here?
17:52:22 <colindixon> #info edwarnicke asks about NFV use cases, he sees NFV as the end-game and OpenStack is just stepping stone
17:55:42 <colindixon> #info edwarnicke says one idea about how to be more story/user-oriented would be to make sure we at least acknowledge when we do meet them
17:56:28 <colindixon> #info vishnoianil__ says that we have a broad set of good ideas and ways forward for things, but we often don’t end up with effective places to discuss the ideas and requirements in a single place
17:56:55 <colindixon> #info colindixon assuems that added onto vishnoianil__’s comments is the idea that we could attach user stories to it
17:57:18 <colindixon> #info edwarnicke says that odlforge would help a lot as having a way to place code somewhere that could be played with
17:58:37 <rovarga> I think the recent OF-CONFIG review was very instructive on the review process... I think breaking down use-cases to architecture is the first step ... and I think it always worked when someone took charge and produced a first cut ...
18:00:30 <colindixon> #info phrobb says that these other comments point to a broader thing about how to break down what we’re doing from a user-oriented, requirements-driven, approach for features
18:02:00 <colindixon> #info art asks how we do feature requests today, colindixon says maybe that would be a good topic for a future call
18:02:22 <colindixon> #endmeeting