I was wondering if there is some sort of a freeware program to help with learning about networks... I have seen such programs for electronic circuits but couldn´t find anything. I am having problems understanding exactly how to use subnet masks and then calculate how many hosts you could have with the mask even though I have read quite a few articles which are supposed to explain it.

Any Ideas?
Adam

There are IP calculators...just google ip calc and you will find tons of them.

If you want to know how many hosts you can have with a given netmask then take the count the number of bit positions with corresponding zeros represented by 'n', raise two to that number and subtract '2':

255.255.255.224
=
11111111.11111111.11111111.11100000 (five zeros)
=
2^5 - 2 (raise 2 to the number of zeros - 2)
=
32 - 2
=
30

Thanks for that wicked explaination I have read lots of things similar but your explaination was what I need to understand how to calculate it easily!

Subtract 2 because the the last is the broardcast address??? and I can´t remember what they first number would be??


Do you have a similar explaination for number of Physical Segments?

Adam

To do the number of networks it is the exact opposite. Instead of using the number of zeros you use the number of ones. And you don't subtract two, but that depends on if you are doing classfull or classless ip addressing. I will use classless since that is what everyone uses nowadays. Read up on RFC1519.

Take class of address:
First octet rule -
0-127 Class A Default Netmask 255.0.0.0
128-171 Class B Default Netmask 255.255.0.0
127-223 Class C Default Netmask 255.255.255.0

Number of bits subnetted:
IPADDR=10.0.0.0
NETMASK=255.255.0.0

Default netmask since the IPADDR starts with 10 would be 255.0.0.0
Therefore, you borrowed 8 bits to get the 255.255.0.0.

2^8 =
256 subnets possible

Hey this second reply was a little less straigt forward for me but I am off to Read up on RFC1519!!

So impressed by your reply I have printed out this topic!

Thanks
Adam

Hi,

There are 2 different approach in IP adressing i think. One says that B Class is between 128-191 and other is as u say 128-171. Which one is right?

Class 'A' networks are 0-127 in the first octet defined bitwise:

0xxxxxxx

Class 'B' networks are 128-191 in the first octet defined bitwise:

10xxxxxx

Class 'C' networks are 192-223 in the first octet defined bitwise:

110xxxxx

Therefore, it must be this way. No other scheme.

Cheers,

Justin

Well actually i agree with this one already but;


u say "128-171 Class B" i didnt understand this one. Also i found this spesification in some other sources (very few). What is the point of this spesification?

Thanx again

I was drunk when I posted that. Obviously, I made a typo, but I don't remember as it was quite long ago.

In each subnet, there are two reserved and fixed IP addresses:
- the lowest IP address (all host bits are 0 (zero)) is the network address
- the highest IP address (all host bits are 1 (one)) is the broadcast address

Each subnet can (but that is not a requirement per se!) have one gateway address and this can be any IP address within the subnet, between the network address and the broadcast address. Good practice however is to use either the lowest or highest available address between network and broadcast address. Without a gateway address, you can't route any data out of this subnet.

Ralf

This is the best tutorial I've seen about subnetting. Don't be fooled by the HTML v. 3 look. http://www.ralphb.net/IPSubnet/index.html

When I first stumbled over it, it was at some dot-edu site. My guess is that, when the author retired or moved on, he took the website with him and put it up unchanged. I suspect the author may have been a teacher, so well does he put forward the concepts.