I've been having an offline debate and wanted to take a quick lithmus test...Mostly aimed at the folks who don't currently peer at any of the 'nationally recognized peering points', but I'll take answers from anyone...Ya'll are always so short on opinions. :-} If someone presented you with the following options: 1) $X connect to a local peering point and peer with other local/regional ISPs (MLPA) You retain current transit. 2) $X+$Y connect to a local peering point and peer with other local/regional ISPs plus the host (a decent sized national carrier) You retain current transit. 3) $X+$Y+$Z Purchase transit from the host, not including access to the local peering points Would purchasing #3 prevent or discourage you from purchasing #1 and would purchasing #1 or #2 prevent or discourage you from purchasing #3? How do folks feel about the concept of local peering points? Comments, public or private, are, of course, welcome. bob iii (not speaking for anyone but myself, & certainly not my employer...)
On Thu, 29 May 1997 bownes@web9.com wrote:
I've been having an offline debate and wanted to take a quick lithmus test...Mostly aimed at the folks who don't currently peer at any of the 'nationally recognized peering points', but I'll take answers from anyone...Ya'll are always so short on opinions. :-}
If someone presented you with the following options:
1) $X connect to a local peering point and peer with other local/regional ISPs (MLPA) You retain current transit. 2) $X+$Y connect to a local peering point and peer with other local/regional ISPs plus the host (a decent sized national carrier) You retain current transit. 3) $X+$Y+$Z Purchase transit from the host, not including access to the local peering points
Would purchasing #3 prevent or discourage you from purchasing #1 and would purchasing #1 or #2 prevent or discourage you from purchasing #3? How do folks feel about the concept of local peering points?
Comments, public or private, are, of course, welcome.
bob iii (not speaking for anyone but myself, & certainly not my employer...)
I'm currently at a local peering point hosted by DRA, when looking for additional bandwidth, I didn't look very hard at DRA at that time based on the fact I was already peering with them, I was very wrong at that time, but I think your question is based on how I viewed things at that time, not now. I also feel others would think along similar lines if put in the same position at other reagional peering points. ============================================================== Tim Flavin Internet Access for St Louis & Chicago Internet 1st, Inc Toll Free Sales & Support 800-875-3173 http://www.i1.net For more information email info@i1.net ==============================================================
I've been having an offline debate and wanted to take a quick lithmus test...Mostly aimed at the folks who don't currently peer at any of the 'nationally recognized peering points', but I'll take answers from anyone...Ya'll are always so short on opinions. :-}
*smile* And we're always so fast to leap in with them, even when we're still 200 messages behind, and not likely to catch up before responding.
If someone presented you with the following options:
1) $X connect to a local peering point and peer with other local/regional ISPs (MLPA) You retain current transit. 2) $X+$Y connect to a local peering point and peer with other local/regional ISPs plus the host (a decent sized national carrier) You retain current transit. 3) $X+$Y+$Z Purchase transit from the host, not including access to the local peering points
Would purchasing #3 prevent or discourage you from purchasing #1 and would purchasing #1 or #2 prevent or discourage you from purchasing #3? How do folks feel about the concept of local peering points?
Much of this depends on the relative orders of magnitude for X, Y, and Z. If X is relatively small compared to Z, I don't think many people would even have to think once before purchasing both #3 and #1, so long as there was no contractual limitation on what other peering could be obtained when purchasing option #3. Assuming a reasonable transit carrier already, #2 would definitely preclude #3. There's greater benefit (IF you already have a good transit carrier) in the redundancy afforded by hearing routes from multiple sources, and having multiple outbound announcements. If a peering session drops in #2, only that peer loses your announcements. Case #3, peer drops, you lose everything at that location. :( Local peering points are best used for just that. Exchange traffic with others in your area. I don't think LOCAL exchange points should be used as places to try to offload traffic destined for far-reaching endpoints multiple hops away. That's what transit carriers are for, and I think transit carriage of traffic out of local regions is only going to increase, as the major players stop seeing value at the local exchange points. It's a great market for a company that wants to stop selling to end-users, and who simply wants to provide transit pipes to regional and local exchange points. Of course, it's past 1am, and I'm just babbling, so feel free to totally ignore this. :)
Comments, public or private, are, of course, welcome. bob iii (not speaking for anyone but myself, & certainly not my employer...)
Matt Petach, barely able to speak for himself, let alone an employer.
participants (3)
-
bownes＠web9.com -
Matthew Petach -
Tim Flavin