Production-scale NAT64

Mark Tinka mark.tinka at seacom.mu
Wed Aug 26 15:59:24 UTC 2015


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256



On 26/Aug/15 17:16, Valdis.Kletnieks at vt.edu wrote:

>
> So I'm guessing that 75% of the traffic flows with better latency than
> the 25% IPvhorse-n-buggy traffic? ;)

Practically, when we've tested NAT64 at reasonable scale, it does not
add any noticeable slow-down provided your hardware is decent and you're
operating the forwarding plane within the limits supported by the vendor.

Yes, I know this can quickly become a cost run-away problem, but for
better or worse, that is what separates the wheat from the other thing...

The point is you need a transition tech. solution if you are serious
about providing a service to your customers. Assuming you don't is
living in denial.

Mark.
-----BEGIN PGP SIGNATURE-----
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=nQN9
-----END PGP SIGNATURE-----




More information about the NANOG mailing list