how-to block ads
|
|
Share Topic
 |
 |
|
|
 tubbynetreminds me of the danse russePremium,MVM
join:2008-01-16
Chandler, AZ | reply to dvd536
Re: I Wonder. . . . . said by dvd536:How many things this is going to break. i am not sure what you're getting at.
many service providers have already been thinking about the ipv6 transition (and many have ipv6 functionality in the core). the issue is that there are many major manufacturers that don't *fully* support ipv6 in the same way that ipv4 is supported. service providers (especially those of small mpls-based networks) often implement "non-standard" solutions in order to provide value-added services to their customers. in doing so, they implement route-leaking, vrf importation, security measures, etc that aren't supported fully in ipv6, or its not supported *in hardware* (meaning the ipv6 packets must be punted to the supervisor/route-processor for software handling, rather than being able to be switching using something similar to cisco express forwarding (cef) in hardware).
as i see it, comcast makes no mention of going "fully ipv6" in the core, but offering ipv6 to the customer endpoints via dhcpv6 and then providing a typical 6-to-4 tunnel at some point before it hits the core - allowing existing infrastructure to be used.
will this affect a few things - sure. cpe has to support ipv6. comcast has to work out the bugs on each of their respective markets to make sure that everything works as promised. this is the point of a "trial". its not like their going cold-turkey.
q.
--
"...if I in my north room dance naked, grotesquely before my mirror waving my shirt round my head and singing softly to myself..." | | |
|
 jjmb
join:2009-12-01
USA | said by tubbynet:said by dvd536:How many things this is going to break. i am not sure what you're getting at. many service providers have already been thinking about the ipv6 transition (and many have ipv6 functionality in the core). the issue is that there are many major manufacturers that don't *fully* support ipv6 in the same way that ipv4 is supported. service providers (especially those of small mpls-based networks) often implement "non-standard" solutions in order to provide value-added services to their customers. in doing so, they implement route-leaking, vrf importation, security measures, etc that aren't supported fully in ipv6, or its not supported *in hardware* (meaning the ipv6 packets must be punted to the supervisor/route-processor for software handling, rather than being able to be switching using something similar to cisco express forwarding (cef) in hardware). as i see it, comcast makes no mention of going "fully ipv6" in the core, but offering ipv6 to the customer endpoints via dhcpv6 and then providing a typical 6-to-4 tunnel at some point before it hits the core - allowing existing infrastructure to be used. will this affect a few things - sure. cpe has to support ipv6. comcast has to work out the bugs on each of their respective markets to make sure that everything works as promised. this is the point of a "trial". its not like their going cold-turkey. q. Excellent points.
FWIW - the Comcast core network fully supports IPv6 and is native, dual-stack.
John | |
tubbynetreminds me of the danse russePremium,MVM
join:2008-01-16
Chandler, AZ | said by jjmb:FWIW - the Comcast core network fully supports IPv6 and is native, dual-stack. i assume in transit, when you're just passing packets, its of no concern. i've just read a lot about the troubles that service providers are experiencing on their pe routers as they import native ipv6 into vrfs/coloring mpls labels. there is a lot left to be desired on brand c's ipv6 manipulation and mangling, however, it will come in due time. it just means that service provider (smaller, mpls-based) will not be able to maintain the same value-added services/one-offs for customers in ipv6 space as they have in ipv4.
q.
--
"...if I in my north room dance naked, grotesquely before my mirror waving my shirt round my head and singing softly to myself..." | |
|
