nanog mailing list archives
Re: Scalability issues in the Internet routing system
From: Blaine Christian <blaine () blaines net>
Date: Thu, 27 Oct 2005 10:30:41 -0400
> I think you hit the nail on the head regarding record locking. This isNeat! So you were thinking you would leave the actual route selection process monolithic and create separate processes per peer? I have seen folks doing something similar with separate MBGP routing instances. Had not heard of anyone attempting this for a "global" routing table with separate threads per neighbor (as opposed to per table). What do you do if you have one neighbor who wants to send you all 2M routes though? I am thinking of route reflectors specifically but also confederation EIBGP sessions.> the thing that is going to bite you if anything will. I have heard > none of the usual suspects speak up so I suspect that either this> thread is now being ignored or no one has heard of an implementation> like the one you just described. In BGP there is no 'global' route (actually path) selection in BGP.Everything is done per prefix+path. In the RIB you can just lock the prefix, insert the new path and recalculate which one wins. Then issue the updateto the FIB, if any. Work done. Statistically there is very littlecontention on the prefix and the path records. For contention two updates for the same prefix would have to arrive at the same time from two different peers handled by different CPU's. I'd guess the SMP scaling factor for BGP is around 1.98. The 0.02 go lost for locking overhead and negative caching effects. Real serialization happens only at the FIB change queue. Howeverserializing queues can be handled very efficiently on SMP too.
Hey Andre,If you are intending to break the BGP process into per neighbor threads this does not sound like it would have beneficial impact on a single neighbor with the majority of the routes (thinking specifically of EIBGP and/or Route Reflectors). Was your idea specifically related to per neighbor processing or were you thinking you could break the BGP process itself into chunks?
Current thread:
- Re: Scalability issues in the Internet routing system, (continued)
- Re: Scalability issues in the Internet routing system JORDI PALET MARTINEZ (Oct 26)
- Re: Scalability issues in the Internet routing system Alexei Roudnev (Oct 26)
- Re: Scalability issues in the Internet routing system Valdis . Kletnieks (Oct 26)
- Re: Scalability issues in the Internet routing system Blaine Christian (Oct 26)
- Re: Scalability issues in the Internet routing system Andre Oppermann (Oct 26)
- Re: Scalability issues in the Internet routing system Christopher L. Morrow (Oct 26)
- Re: Scalability issues in the Internet routing system Blaine Christian (Oct 26)
- Re: Scalability issues in the Internet routing system Andre Oppermann (Oct 26)
- Re: Scalability issues in the Internet routing system Blaine Christian (Oct 26)
- Re: Scalability issues in the Internet routing system Andre Oppermann (Oct 27)
- Re: Scalability issues in the Internet routing system Blaine Christian (Oct 27)
- Re: Scalability issues in the Internet routing system Michael . Dillon (Oct 27)
- Re: Scalability issues in the Internet routing system william(at)elan.net (Oct 27)
- Re: Scalability issues in the Internet routing system Lincoln Dale (Oct 26)
- Re: Scalability issues in the Internet routing system Blaine Christian (Oct 26)
- Re: Scalability issues in the Internet routing system Alexei Roudnev (Oct 27)
- Re: Scalability issues in the Internet routing system Lincoln Dale (Oct 27)
- Re: Scalability issues in the Internet routing system Andre Oppermann (Oct 19)
- Re: Scalability issues in the Internet routing system Paul Jakma (Oct 19)