Re: A couple of possibly interesting stats.
The first table focuses on the current classful routes being passed around by AS and the amount of gain CIDR would give *IF* we could aggregate as the AS level. It shows the number of nets per AS so we see who is announcing what in terms of BGP as well as what the number of CIDR nets would be advertised if we could aggregate at this level...
Tony,
Tony, On average each net learned by AS 690 is announced in 2.5 places. I think the range is typically 1-5 with the most common number being 2 (not sure on that). In any case, I think your assumption that we can aggregate at the AS level may be highly optimistic (at least for the near term - until we get far more sophisticated tools to track topology and aggregation implications). Of course, if CIDR either didn't prefer the more specific route but relied entirely on policy or punched holes in aggregates rather than overlap, this wouldn't be an issue, but that's the way CIDR works so it is an issue. It makes life much more interesting for Routing Authorities. :-) Curtis
On average each net learned by AS 690 is announced in 2.5 places.
This is an interesting peice of data, but does not tell us anything when we are okking at how much CIDR can by us. Looking at how many places the NSFNET sees inbound announcements of nets does not matter at all. All it tells you is that the global Internet is redundant in terms of links. Look at home ASs of nets - if you aggregate at the home AS, that is what is going to count. --asp@uunet.uu.net (Andrew Partan)
Looking at how many places the NSFNET sees inbound announcements of nets does not matter at all. All it tells you is that the global Internet is redundant in terms of links.
Look at home ASs of nets - if you aggregate at the home AS, that is what is going to count.
--asp@uunet.uu.net (Andrew Partan)
Andrew, A lot of the regional networks have more than one attachment point to ANSnet. Some use one as primary and one as backup. Most of the regional networks announce some of their networks as primary at one attachment point and other networks at another attachment point. This means that that regional could not aggregate before reaching ANSnet because to do so would mean losing the load balancing. One longer term way around this is to have ANS accept more specific routes plus an aggregate and propogate the more specific routes into it's IBGP but not propogate them further (just the aggregate). In any case, configuration is going to get a lot harder for many regionals to make sure that they aggregate equally to all their peers but still are capable of passing more specific routes to ANS or any peer network to which they attach at multiple places and for which they need to load balance. Knowing that the average is 2.5:1 just indicates that aggregating entire AS is not going to be easy and that the figures arrived at using that assumption are going to be very optimistic at least for the near term. The second point I made was that a major role of the routing arbitrator will be to provide enough knowlege of topology to make more aggregation possible so we can approach the kind of figures Tony projected. Maybe that point got lost because of the way it was presented. Regards, Curtis
If they have more than one attachement, then they are (currently) using more than one AS - so with "one AS one policy", they should be fine. [If folks are violating the "one AS one policy" rule, then they are going to have problems that they have brought upon themselves. BGP does not really support more than one AS per policy.] If there are 2 ASs that touch in more than one point, then BGP4's inter-AS metric should take care of this (you will need to have some more care in what routes you pass between these 2 ASs and what routes they pass on to their neighbors), but as you need to coordinate to have 2 ASs touch in more than one point anyhow, you should just need to add a more to the coordination. In any case, downstream ASs should not know or care or need to do anything to make this work. You certainly should be able to do aggregation by site w/o any problem (even for ASs that touch in more than 1 place).
The second point I made was that a major role of the routing arbitrator will be to provide enough knowlege of topology to make more aggregation possible so we can approach the kind of figures Tony projected. Maybe that point got lost because of the way it was presented.
Actaully, the role of the RA is to get people to give it their (local) data so that the RA can get a database that has (nearly) complete global info. Projections of how much we are going to get from CIDR will be somewhat soft until we started getting a couple of real life examples. I have poked at what we would get under various CIDR strategeies, but as I have not done it for "real", I am not 100% sure yet. Here is some results from a run that I did of my data a while back: 1381 total nets 646 CIDR routes if CIDRize by site 619 CIDR routes if CIDRize by AUP (NSFNET vs non-NSFNET) 1026 NSFNET nets; 468 NSFNET CIDR routes 355 non-NSFNET nets; 151 non-NSFNET CIDR routes 500 CIDR routes if CIDRize everything together Even if we just CIDRized by site, it looks like we will get nearly all of the savings that we would get if we fully CIDRized. --asp@uunet.uu.net (Andrew Partan)
If they have more than one attachement, then they are (currently) using more than one AS - so with "one AS one policy", they should be fine.
[If folks are violating the "one AS one policy" rule, then they are going to have problems that they have brought upon themselves. BGP does not really support more than one AS per policy.]
Where did this come from? I was talking about load balancing between predominantly T1 networks that touch the T3 network in two geographically distant places and require the T3 network to take an exit point that minimizes the slower networks internal distance.
Projections of how much we are going to get from CIDR will be somewhat soft until we started getting a couple of real life examples. I have poked at what we would get under various CIDR strategeies, but as I have not done it for "real", I am not 100% sure yet.
Here is some results from a run that I did of my data a while back: 1381 total nets 646 CIDR routes if CIDRize by site 619 CIDR routes if CIDRize by AUP (NSFNET vs non-NSFNET) 1026 NSFNET nets; 468 NSFNET CIDR routes 355 non-NSFNET nets; 151 non-NSFNET CIDR routes 500 CIDR routes if CIDRize everything together
Even if we just CIDRized by site, it looks like we will get nearly all of the savings that we would get if we fully CIDRized. --asp@uunet.uu.net (Andrew Partan)
Alternet has grown very substantially since CIDR allocation was in place, much more so than many other networks (gov agencies, some regionals that were well established early on, etc). So Alternet represents a good case to benfit from CIDR, but maybe not a typical case. I'm glad to hear that Alternet alone can eliminate 700-800 of the routes. Things look optimistic for getting relief from the 19,000 route situation we are now in, if we can just get rolling. Curtis
participants (2)
-
asp＠uunet.uu.net -
Curtis Villamizar