IPv6 20-bit End-to-End Field (was Flow Label?)

1. When only 64-bit Address Fields are needed, spare bits are found in the
right-most Source and Destination 128-bit fields.

2. Up to 16 bytes of DATA can be stealthed in those un-used Address bits.
VOIP samples and other streams are ideal.

3. To support variable-length 16-bit fields, 4-bits are needed to encode 1
to 16.

4. If the 4-bits are stored in the 20-bit "Flow" field, then 16 bits remain
and can make a handy checksum or per-DATA-byte parity bits.

5. To make the 4-bit LLLL field more reliable it could be stored several
times in the "Flow" field. Three copies would only use 12 of the 20 bits,
leaving 8 bits for a small checksum of only the DATA.

6. Even with the LLLL field, the DATA can be repeated as often as possible
in the spare Address bits. The copies can be compared to ensure a reliable
transmission.
The fields are fixed length. Any usage is better than 0s.

7. The 20-bits are sent End-to-End without change. The Edge devices know
what
algorithm to apply. When the LLLLCCCCLLLLCCCCLLLL format is used a MTM
may be able to guess the algorithm. The CCCC fields could also select a
compression table for the DATA.

8. The DATA appears as rapidly changing Address bits making it appear that
packets are being routed to many nodes. Only 64-bits are needed/used (66 for
IEEE
802.1Q VLAN 3q tags) for addressing.

9. The 320-bit IPv6 header is all that is needed, it has Addresses and DATA.
When combined with a 160-bit Routing Header (IPv4) one has 480-bits. Those
480-bits become natural DHT - Distributed Hash Table - Key values.
put(Key,Data,Time) The Keys may contain several "IP Address" values, plus
all the rest of the bits in both headers, including DATA.

10. The illusion at the Edge of the (Fringe(Edge(Core))) 3 target
architecture is
that the DHT is part of the .NET. It just works, like the cloud. It is like
a virtual
disk drive with a 480-bit sector index. That is sort of large.

IPv6 20-bit End-to-End Field (was Flow Label?)
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to