I think its too easy, thats the problem. For <$1000 (excluding bandwidth/ccts) you can buy a box, connect to your two providers, get an ASN and IPs and you're away. Compare to the telephone network, to 'multihome' you need to get licenses, allocations of numbers and codes thats not so easy, get some SS7 kit and do your data builds.. you're talking quite a lot more money and certainly a lot more difficult technically. Perhaps we should make the Internet more difficult :) I dont agree that connecting to two+ upstreams makes you better. In my experience end networks have a couple of orders of magnitude more downtime than a PoP in any reasonably large ISP. Ie the percentage theoretical improvement is small. In addition you seriously increase the complexity of your system, chances are you're using the cheapest kit you could find (or at least cheaper and smaller than what I would use).. its not great at BGP and may fall over when you get a minor DoS attack, you probably generate flaps quite a bit from adhoc changes and if you're announcing a /24 then thats going to get you dampened quickly.. so you actually create a new weakest link. Also most of the corporates I've dealt with take defaults rather than full tables.. so if the provider does have an issue you still forward the traffic, theres no failover of outbound routing. Even if you spend (waste) the money on some decent gear, you're on your own and when a problem occurs the ISPs are going to be less helpful to you (not by choice, I mean they dont have control of your network any more.. there knowledge of whats causing problems is limited to the bit that they provide to you), so chances are your problems may be more serious and take longer to diagnose and fix. IMHO avoid multihoming. You will know when you are big enough and you *need* to do it, if you're not sure or you only want to do it cause you heard everyone else is and its real cool then I suggest you dont. Steve On Thu, 11 Mar 2004, John Neiberger wrote:
On another list we've been having multihoming discussions again and I wanted to get some fresh opinions from you.
For the past few years it has been fairly common for non-ISPs to multihome to different providers for additional redundancy in case a single provider has problems. I know this is frowned upon now, especially since it helped increase the number of autonomous systems and routing table prefixes beyond what was really necessary. It seems to me that a large number of companies that did this could just have well ordered multiple, geographically separate links to the same provider.
What is the prevailing wisdom now? At what point do you feel that it is justified for a non-ISP to multihome to multiple providers? I ask because we have three links: two from Sprint and one from Global Crossing. I'm considering dropping the GC circuit and adding another geographically-diverse connection to Sprint, and then removing BGP from our routers.
I see a few upsides to this, but are there any real downsides?
Flame on. :-)
Thanks, John --