About address allocating
Miles Nordin
carton@Ivy.NET
Sat, 20 May 2000 13:40:50 -0600
On Sat, May 20, 2000 at 03:17:57PM +0200, Francis Dupont wrote:
> >is to introduce "small site" which get
> >/56 (on byte boundary, large enough for up to 256 subnetworks or
> >a few levels of hierarchy).
> => the idea is that it is easier for someone which needs a /48
> to deal with its ISP than for a common customer to fight in order to
> get a /48 because /64 is not enough:
w.t.f.? I thought the point if IPv6 was to get rid of all this stingy
political nonsense. I should have enough addresses to represent my
internal topology. I should not have to pay or fight for them. PERIOD.
Why is RIPE backpedaling from this?
I do not want a repeat of our current situation, where American fools
drinking cheap beer and checking their Hotmail accounts get service
for $40/mo, and a small oppressed minority of technically creative
people, who actually have such a thing as ``internal topology'' have
to pay $140/mo for a connection of the exact same speed, just because
we're not using the sealed web-box that The Corporation gave us.
Who is proposing giving out /64's in exchange for money? I thought
the (rather well-considered?) standard specifically recommend the
Site Level Aggregator for the use of an ISP's _subscribers_. that
part of the point of this ``standard'' was that it solved political
disputes by dividing up the address bits and assigning them to each
of the interested parties. no?
Is there really an expected shortage of IPv6 addresses? Or is this
all about people who say, ``but that seems like an awful _waste_ of
addresses, 16 bits on someone with one subnet. We shouldn't give
them out like that--addresses are precious.'' NO! Addresses
_were_ precious.
The IPv6 paper I read makes a great case for ``wasting'' addresses in
terms of improved routing performance (addresses are more aggregateable)
and elimination of special cases (NAT, proxy ARP). so, there are
technical reasons subscribers should get the SLA, and no technical
reasons that they shouldn't.
I am confused as to where this strange address-mongering way-of-thinking
is coming from. It sounds like IPv4-Think to me, and we don't do that
here in the Brave New World.
The only _political_ reason not to give subscribers the SLA is so that
you can later sell it back to them for more money. It sounds to me
like that's what's going on here. ISP's are attempting to design a
profitable future pricing structure into the IPv6 standard, and
protect their substantial existing revenues for selling hotly-contested
IPv4 space to rich corporations. The standard as it stands takes a
big chunk out of their pie--``yes, by all means, let's create a `small
subscriber'--that way, we can charge for numbers, which are free, and
make pure revenue. Any standard which throws us back to the old days
of simply charging for bandwidth & rtt is unacceptable to us.''
>From The Case for IPv6:
http://www.6bone.net/misc/case-for-ipv6.html
Next level aggregators can divide the NLA address field to create
their own hierarchy, one that maps well to the current ISP industry,
in which smaller ISPs subscribe to higher level ISPs, and so on.
This is accomplished by the further subdivision of the 32-bit
NLA field (see Figure 15). Following the NLA ID are for
<------------ 32 bits -----------> <--16 bits-> <---- 64 bits
+-------+-------------------------+------------+-------------------+
| NLA 1 | Site | SLA | Interface ID |
+-------+-------------------------+------------+-------------------+
+-------+-----------------+------------+-------------------+
| NLA 2 | Site | SLA | Interface ID |
+-------+-----------------+------------+-------------------+
+-----------------+------------+-------------------+
| NLA 3 | Site | SLA | Interface ID |
+-----------------+------------+-------------------+
Figure 15: Subdividing the NLA Address Space
subscriber site networking information: Site Level Aggregator (SLA)
and Interface ID. Typically, service providers supply subscribers
with blocks of contiguous addresses, which are then used by
individual organizations to create their own local address hierarchy
and identify subnets and hosts. The 16-bit SLA field supports up to
65,535 individual subnets.
--
Miles Nordin / v:+1 720 841-8308 fax:+1 530 579-8680
555 Bryant Street PMB 182 / Palo Alto, CA 94301-1700 / US
<IMG SRC="C:\CON\CON" HEIGHT="0" WIDTH="0">