[AusNOG] powerful routers in core/edge routing/switching

[Ankit Agrawal]

Macca, agree, it might not be a necessity for majority of people.

Need to understand what is best on a technical level first and obviously
commercials will follow.

I haven't really seen anyone criticising what they are using from the
options that I listed below which is sort of what I thought as well. They
all can do the job which sort of makes it harder to decide which one to
pick. You just need to look at what is going to best suite the
infrastructure where the product will be deployed and obviously cater for
future scalability.

Ankit.

From:  McDonald Richards <McDonald.Richards at vocus.com.au>
Date:  Thursday, 14 February 2013 9:02 AM
To:  "AusNOG at lists.ausnog.net" <ausnog at lists.ausnog.net>
Subject:  Re: [AusNOG] powerful routers in core/edge routing/switching

You're right ­ some times it's not enough, but I don't think that is the
case for the majority of people who are carrying a full BGP table. If you
can afford all that connectivity and caching, you can probably afford to buy
a router that is suitable to support it all (provided all the afore
mentioned caching and connectivity is actually a requirement of the
business).

Macca


From: Ankit Agrawal <ankitagrawals at gmail.com>
Date: Wednesday, 13 February 2013 12:50 PM
To: McDonald Richards <McDonald.Richards at vocus.com.au>,
"AusNOG at lists.ausnog.net" <ausnog at lists.ausnog.net>
Subject: Re: [AusNOG] powerful routers in core/edge routing/switching

Sometimes default route is not enough and you do need to be aware of full
routing table, including local peering to have optimum routing both from
cost and user experience perspective. This is especially the case when you
are not running a single IP feed from one or two providers but infact have
several transit links and peering/caching services.

Of course there are ways to overcome these issues, but then as you said, you
end up in a web of network that only you can understand and support and its
beyond documentation.

Ankit.

From: McDonald Richards <McDonald.Richards at vocus.com.au>
Date: Wednesday, 13 February 2013 8:15 AM
To: "AusNOG at lists.ausnog.net" <ausnog at lists.ausnog.net>
Subject: Re: [AusNOG] powerful routers in core/edge routing/switching

What has port density, capacity or throughput got to do with routing?

Why do people with networks in a single geographic region, even if
multi-homed, need to run default-fee? You know you can use a default route
and a routing subset to achieve both redundancy and faster convergence?

I'm pretty happy with the current generation of hardware and where it sits
price-wise. There have been a lot of good suggestions in the thread so I
won't throw anymore in.There are cheaper and smarter ways to do things, but
with smarts comes the risk that nobody else can support your tangled web of
network magic.

Macca


From: Joshua D'Alton <joshua at railgun.com.au>
Date: Wednesday, 13 February 2013 11:06 AM
To: "AusNOG at lists.ausnog.net" <ausnog at lists.ausnog.net>
Subject: Re: [AusNOG] powerful routers in core/edge routing/switching

The problem is routing has lagged far behind switching in terms of port
density, capacity, throughput etc. Obviously a switching engine is peanuts
compared to a routing engine, but it is exaggerated by the massive amounts
of features they put in routing engines.

Seems to me we almost need a new breed of edge routers, ones that just talk
BGP to other providers, and the current edge can stay as they are handling
fancy things like MPLS which is really more of an internal routing,
therefore switching, feature. Or not :)

_______________________________________________ AusNOG mailing list
AusNOG at lists.ausnog.net http://lists.ausnog.net/mailman/listinfo/ausnog
_______________________________________________ AusNOG mailing list
AusNOG at lists.ausnog.net http://lists.ausnog.net/mailman/listinfo/ausnog

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20130214/fb5fa191/attachment.html>