16:00:27 <readams1> #startmeeting
16:00:27 <odl_meetbot> Meeting started Mon Apr 21 16:00:27 2014 UTC.  The chair is readams1. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:27 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:00:40 <readams1> #chair alagalah
16:00:40 <odl_meetbot> Current chairs: alagalah readams1
16:01:30 <alagalah> #topic End point registry and mapper
16:04:01 <alagalah> #link https://wiki.opendaylight.org/view/Project_Proposals:Group_Based_Policy_Plugin#Group_Policy_Components_and_Plugins describes architecture of system
16:04:42 <s3wong> hemanthravi: ping
16:04:47 <alagalah> Please join the ghangout
16:04:49 <hemanthravi> s3wong: hi
16:04:55 <alagalah> hemanthravi: please join the hangout
16:05:12 <s3wong> hemanthravi: are you going to join the endpoint mapper / policy repo meeting on G hangout?
16:05:29 <alagalah> #topic Policy repository, end-point mapper group meeting
16:05:36 <hemanthravi> yes
16:06:22 <alagalah> #info need to fabricate a set of requirements on MD-SAL, what are the latency requirements, resolution requirements on the MD-SAL
16:06:57 <alagalah> #info dvorkinista Says natural thing since our stuff and MD-SAL are both tree based. Depends on how selectors and relators evolved, so they can be supported in a native way in the MD-SAL
16:07:51 <alagalah> #info need to study and understand the capabilities of MD-SAL and see if applicable (working backwards) and then as selectors and relators evolve (working forward) and work towards each other
16:10:13 <alagalah> #info idea is minimize changes to both approaches.
16:12:36 <alagalah> #info dvorkinista The actual Endpoint registry may not require MD-SAL datastore. Its a simple hash table implementation. We need to investigate using MD-SAL as a message bus but not a datastore for EP registry (because it may not require a tree structure)
16:16:08 <alagalah> #info readams1 Wants to stub out a sample MD-SAL implementation for EP Registry
16:16:42 <alagalah> #action alagalah and readams1 to work on a date for date
16:18:19 <alagalah> #action target to have a skeleton structure by next week's Monday
16:20:12 <alagalah> #info dvorkinista We are assuming MD-SAL, and should look at making both trees, even though EP registry could be implemented of K,V Map
16:20:59 <alagalah> #info Rest of team has nothing to add
16:21:06 <readams1> #endmeeting