[jhorn@terremark.com: RIS in Miami]
Hi, What do you think about this offer? Arife ----- Forwarded message from Josh Snowhorn <jhorn@terremark.com> ----- From: "Josh Snowhorn" <jhorn@terremark.com> To: <arife@ripe.net> Subject: RIS in Miami Date: Mon, 6 Feb 2006 22:56:41 -0500 Hi Arife, I just noticed that you are the tech contact for RIS. I already have K-Root down at the NAP of the Americas in my lab cabinets. Why not put a RIS collector here too? It would be fee of course :-)... See you soon! Josh Joshua Snowhorn Director NAP of the Americas, Inc. A Terremark Worldwide Company (Amex:TWW) 305-205-8327 Cell snowhorn@terremark.com <mailto:snowhorn@terremark.com> AIM:snowhorn70 -- This message has been scanned for viruses and dangerous content, and is believed to be clean. ----- End forwarded message ----- -- Arife Vural SED, RIPE NCC
Hi Arife,
What do you think about this offer?
hmm, miami ... want to go there and install it yourself? :-) I guess the answer depends on what a box at NAP would add and how the revised RIS infrastructure would perform when more data are coming in. New York and Palo Alto don't have that many peers compared to the RIS boxes at European exchanges (amsix,linx,decix,vix). If you would get more or others at NAP it might be worth while. This also relates to Joao's question at RIPE51 following your presentation in the routing-wg: "You are adding more and more collectors. How many are enough?" If you know how to answer that, you know how to answer all future requests/offers to host a RIS route collector. ;) -- Rene
What do you think about this offer?
hmm, miami ... want to go there and install it yourself? :-)
That would be nice ha, after having all grey days here :)
I guess the answer depends on what a box at NAP would add and how the revised RIS infrastructure would perform when more data are coming in. New York and Palo Alto don't have that many peers compared
That makes me wonder too actually, maybe we should spend more time to increase the number of peers in certain places. US boxes, and South America, Japan we do not have much boxes. Arife
to the RIS boxes at European exchanges (amsix,linx,decix,vix). If you would get more or others at NAP it might be worth while.
This also relates to Joao's question at RIPE51 following your presentation in the routing-wg:
"You are adding more and more collectors. How many are enough?"
If you know how to answer that, you know how to answer all future requests/offers to host a RIS route collector. ;)
-- Rene
Hi all,
"You are adding more and more collectors. How many are enough?"
This question came up in a workshop on BGP where there was a proposal to install 1 RRC-like device at _every_ IX in the world. That means something like 100-200 of them. This is a non-trivial operation to install and maintain, so the question came up: what can one do with N points, what does an N+1'th add. Nobody of the BGP analysis folks could give a good answer. (I did suggest that this would be an interesting study: take 1 RRC, do some analysis, add a second, repeat and see what you add, etc.) The handwaving argument is, of course, list all AS that we peer with, list AS present at the new location, if there is a huge difference, you'll learn a lot by adding an IX. If the difference is small, there isn't. Also, we should ask if/how NAP (or whoever) wants to promote peering with the RRC. This is a problem at some locations. Henk ------------------------------------------------------------------------------ Henk Uijterwaal Email: henk.uijterwaal(at)ripe.net RIPE Network Coordination Centre http://www.amsterdamned.org/~henk P.O.Box 10096 Singel 258 Phone: +31.20.5354414 1001 EB Amsterdam 1016 AB Amsterdam Fax: +31.20.5354445 The Netherlands The Netherlands Mobile: +31.6.55861746 ------------------------------------------------------------------------------ 1160438400. Watch this space...
participants (3)
-
Arife Vural -
Henk Uijterwaal -
Rene Wilhelm