💾 Archived View for gmi.noulin.net › rfc › rfc1924.gmi captured on 2024-08-25 at 03:32:04. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2021-12-05)
-=-=-=-=-=-=-
Keywords: encoding
Network Working Group R. Elz
Request for Comments: 1924 University of Melbourne
Category: Informational 1 April 1996
A Compact Representation of IPv6 Addresses
Status of this Memo
This memo provides information for the Internet community. This memo
does not specify an Internet standard of any kind. Distribution of
this memo is unlimited.
1. Abstract
IPv6 addresses, being 128 bits long, need 32 characters to write in
the general case, if standard hex representation, is used, plus more
for any punctuation inserted (typically about another 7 characters,
or 39 characters total). This document specifies a more compact
representation of IPv6 addresses, which permits encoding in a mere 20
bytes.
2. Introduction
It is always necessary to be able to write in characters the form of
an address, though in actual use it is always carried in binary. For
IP version 4 (IP Classic) the well known dotted quad format is used.
That is, 10.1.0.23 is one such address. Each decimal integer
represents a one octet of the 4 octet address, and consequently has a
value between 0 and 255 (inclusive). The written length of the
address varies between 7 and 15 bytes.
For IPv6 however, addresses are 16 octets long [IPv6], if the old
standard form were to be used, addresses would be anywhere between 31
and 63 bytes, which is, of course, untenable.
Because of that, IPv6 had chosen to represent addresses using hex
digits, and use only half as many punctuation characters, which will
mean addresses of between 15 and 39 bytes, which is still quite long.
Further, in an attempt to save more bytes, a special format was
invented, in which a single run of zero octets can be dropped, the
two adjacent punctuation characters indicate this has happened, the
number of missing zeroes can be deduced from the fixed size of the
address.
In most cases, using genuine IPv6 addresses, one may expect the
address as written to tend toward the upper limit of 39 octets, as
long strings of zeroes are likely to be rare, and most of the other
Elz Informational [Page 1]
RFC 1924 A Compact Representation of IPv6 Addresses 1 April 1996
groups of 4 hex digits are likely to be longer than a single non-zero
digit (just as MAC addresses typically have digits spread throughout
their length).
This document specifies a new encoding, which can always represent
any IPv6 address in 20 octets. While longer than the shortest
possible representation of an IPv6 address, this is barely longer
than half the longest representation, and will typically be shorter
than the representation of most IPv6 addresses.
3. Current formats
[AddrSpec] specifies that the preferred text representation of IPv6
addresses is in one of three conventional forms.
The preferred form is x:x:x:x:x:x:x:x, where the 'x's are the
hexadecimal values of the eight 16-bit pieces of the address.
Examples:
FEDC:BA98:7654:3210:FEDC:BA98:7654:3210 (39 characters)
1080:0:0:0:8:800:200C:417A (25 characters)
The second, or zero suppressed, form allows "::" to indicate multiple
groups of suppressed zeroes, hence:
1080:0:0:0:8:800:200C:417A
may be represented as
1080::8:800:200C:417A
a saving of just 5 characters from this typical address form, and
still leaving 21 characters.
In other cases the saving is more dramatic, in the extreme case, the
address:
0:0:0:0:0:0:0:0
that is, the unspecified address, can be written as
::
This is just 2 characters, which is a considerable saving. However
such cases will rarely be encountered.
Elz Informational [Page 2]
RFC 1924 A Compact Representation of IPv6 Addresses 1 April 1996
The third possible form mixes the new IPv6 form with the old IPv4
form, and is intended mostly for transition, when IPv4 addresses are
embedded into IPv6 addresses. These can be considerably longer than
the longest normal IPv6 representation, and will eventually be phased
out. Consequently they will not be considered further here.
4. The New Encoding Format
The new standard way of writing IPv6 addresses is to treat them as a
128 bit integer, encode that in base 85 notation, then encode that
using 85 ASCII characters.
4.1. Why 85?
2^128 is 340282366920938463463374607431768211456. 85^20 is
387595310845143558731231784820556640625, and thus in 20 digits of
base 85 representation all possible 2^128 IPv6 addresses can clearly
be encoded.
84^20 is 305904398238499908683087849324518834176, clearly not
sufficient, 21 characters would be needed to encode using base 84,
this wastage of notational space cannot be tolerated.
On the other hand, 94^19 is just
30862366077815087592879016454695419904, also insufficient to encode
all 2^128 different IPv6 addresses, so 20 characters would be needed
even with base 94 encoding. As there are just 94 ASCII characters
(excluding control characters, space, and del) base 94 is the largest
reasonable value that can be used. Even if space were allowed, base
95 would still require 20 characters.
Thus, any value between 85 and 94 inclusive could reasonably be
chosen. Selecting 85 allows the use of the smallest possible subset
of the ASCII characters, enabling more characters to be retained for
other uses, eg, to delimit the address.
4.2. The Character Set
The character set to encode the 85 base85 digits, is defined to be,
in ascending order:
'0'..'9', 'A'..'Z', 'a'..'z', '!', '#', '