Re: Query about the Amount of Multicast State information

From: Adesh Chaudhary <er.adeshchaudhary_at_gmail.com>
Date: Sun, 31 Mar 2013 11:26:09 +0530

On Sun, Mar 31, 2013 at 12:21 AM, Yuri Bank <yuribank_at_gmail.com> wrote:

> Each (S,G) has multiple timers, outgoing interface list that may be
> constantly changing etc.... Overall, an entry in the mroute table could
> require a lot of overhead in the control-plane. A BGP route on the other
> hand is basically stateless, a prefix is learned and installed into the
> RIB, nothing else happens with it until the neighbor it was learned from
> goes down or an update is received.
>
> -YuriB
>
>
>
>
> On Sat, Mar 30, 2013 at 2:45 AM, Adesh Chaudhary <
> er.adeshchaudhary_at_gmail.com> wrote:
>
>> Hi Folks,
>>
>> The amount of State information (especially S,G) seems like a big concern,
>> when I am referring to Multicast texts / trainings.
>> I am unable to understand why? I mean Unicast routing table can handle
>> full
>> BGP prefix information. Then why is "Large amount of Multicast State
>> information (S,G) considered such a big deal"?
>> Can't the Multicast table handle it? OR What?
>>
>> Please let me know about your views on this. I am not able to get it.
>>
>> --
>> Thanks & Regards,
>> Adesh
>> +91 99996 10511 (Delhi)
>> +91 99860 10511 (Bangalore)
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>>
>>
>>
>>
>>
>>
>
Thanks Yuri, Agreed to your point regarding maintenance operations on
MROUTE table. Now I get it better.
Cheers!

-- 
Thanks & Regards,
Adesh
+91 99996 10511 (Delhi)
+91 99860 10511 (Banglore)
Blogs and organic groups at http://www.ccie.net
Received on Sun Mar 31 2013 - 11:26:09 ART

This archive was generated by hypermail 2.2.0 : Wed Apr 03 2013 - 19:06:19 ART