- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-16-2012 08:24 AM - edited 03-01-2019 05:32 PM
Hi everybody.
How is everyone doing?
My book says " the dynamic forwarding logic requires more work per packet as compared to point-point tunnels which is one of the main reason multipoint tunnels are best used for less frequent traffic while point-to-point tunnels are best suited for more frequent traffic"
In case of multipoint ipv6 tunnel, router has to drive the tunnel destination ip which is encoded in ipv6 address . What other work is performed on a packet when it comes to multipoint ipv6 tunnel which is not performed in case of static ipv6 over ip tunnel ?
thanks and have a great day
Solved! Go to Solution.
- Labels:
-
IPv6 Configuration
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-16-2012 11:21 PM
Hi Sarah,
Apart from "extracting" the embedded IPv4 address from the IPv6 address and placing it into a newly constructed IPv4 encapsulating header, I do not think there is any more significant work involved for multipoint tunnels. It is true that with static point-to-point tunnels, you can already have that header prepared beforehand in memory for all packets - you just use it again and again. With dynamic tunnels, you have first to derive the destination IPv4 address and then place it into a new IPv4 header but even this can be done in software so that the difference in the amount of work is negligible.
Best regards,
Peter
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-16-2012 11:21 PM
Hi Sarah,
Apart from "extracting" the embedded IPv4 address from the IPv6 address and placing it into a newly constructed IPv4 encapsulating header, I do not think there is any more significant work involved for multipoint tunnels. It is true that with static point-to-point tunnels, you can already have that header prepared beforehand in memory for all packets - you just use it again and again. With dynamic tunnels, you have first to derive the destination IPv4 address and then place it into a new IPv4 header but even this can be done in software so that the difference in the amount of work is negligible.
Best regards,
Peter
