Subnetting Non Clear Idea

Hi

the subject of subnetting is very un clear to me, i mean if we have many
network segments and suppose we use class B IP addresses why we not just make
the first netork id 135.120.00 as an example and the second 135.121.00 or
136.122.0.0
with network mask 255.255.0.0 why we need to take from the 16 bits of the
hosts to create subnets with masks such as 255.255.224.0 ??!!

in my example if you know the class of ip address you will know the network
id and subnet mask without needing to chack bits of it so whats the reason of
complicating it ??!!!!!!!!

thank you

--
Ammar S. Mitoori
IT Head QIMCO Co.
Tel : +9744831199
Mobile : +9745378400
Fax : +9744831643

Pure Heart wrote:
> Hi
>
> the subject of subnetting is very un clear to me, i mean if we have many
> network segments and suppose we use class B IP addresses why we not just make
> the first netork id 135.120.00 as an example and the second 135.121.00 or
> 136.122.0.0
> with network mask 255.255.0.0 why we need to take from the 16 bits of the
> hosts to create subnets with masks such as 255.255.224.0 ??!!
>
> in my example if you know the class of ip address you will know the network
> id and subnet mask without needing to chack bits of it so whats the reason of
> complicating it ??!!!!!!!!

Do some Googling for VLSM (variable length subnet masks) and CIDR
(classless inter-domain routing).

IP address classes are old, and waste large chunks of addresses. They
are partly responsible for the impending shortage of IP addresses. Class
A networks were given out to companies that would never need that kind
of address space (a class A can have over 16 million hosts). And now
that unused address space can't be utilised unless the companies decide
to return them to IANA.

http://en.wikipedia.org/wiki/List_of...address_blocks


Talking about your example though, some reasons for subnetting
(especially a large address space like a Class B) are:

- Reduce the size of the broadcast domain, and therefore reduce
network traffic. A Class B network has over 65,000 potential hosts. Any
broadcast (ARP broadcasts, for example) would have to be sent to every
single one of those hosts.

- Enable you to segment your network with routers

Forget about IP address classes.

Cheers,

--
Chris M.

"Pure Heart" <(E-Mail Removed)> wrote in message
news:6B774063-206C-4B42-BA34-(E-Mail Removed)...
> Hi
>
> the subject of subnetting is very un clear to me, i mean if we have many
> network segments and suppose we use class B IP addresses why we not just
> make
> the first netork id 135.120.00 as an example and the second 135.121.00 or
> 136.122.0.0
> with network mask 255.255.0.0 why we need to take from the 16 bits of the
> hosts to create subnets with masks such as 255.255.224.0 ??!!
>
> in my example if you know the class of ip address you will know the
> network
> id and subnet mask without needing to chack bits of it so whats the reason
> of
> complicating it ??!!!!!!!!


Never anything less than /24 (255.255.255.0).

Ethernet effieciency drops off after around 300 hosts. A mask of /24 keeps
the host at 254 which is a perfect size. Lesser bit masks (like /8, /16,
etc) are for supernetting over backbones (which are "broken up" downstream)
and for keeping router's routing table efficient by combining many routes
into one (supernetting),...that is the method of how the Public Internet
routing functions,....those masks are not for creating Host Networks.

Like Chris said,...forget about "Classes",...they are meaningless for the
most part. Learn what CIDR and VLSM mean and what to do with them.

--
Phillip Windell
www.wandtv.com

The views expressed, are my own and not those of my employer, or Microsoft,
or anyone else associated with me, including my cats.
-----------------------------------------------------

Hi,
even your 255.255.0.0 nets are subnets, although they are bigger ones.
these are "class" (-ified) based nets;
More or less they are rough structurised.

if in your ".0.0" portion, you can realise a more granular
portionising of the whole net with setting some "1"'s into it.

so if you have to manage a 255.255.255.0 net and it is appropriate for you,
why then
not having two 255.255.255.128 (/25) nets?

this used, no one could really figure out the net out of the ip, without the
subnet mask.
these adresses then are not "class" based.

jk