# Finding the number of subnets. Which method is correct?

Apr 15th, 2007

Hi there! I have a rather important question which I hope someone knowledgeable will be able to help me with.

It's a rather important subnetting question. And I'm asking, how do you find the number of subnets?

Ok at this stage I'd like to point out that I do actually know how to find the number of subnets but I'm rather confused which method is acceptable for the CCNA exam.

In all my literature it is stated that in order to find the number of subnets one must do the follwing calculation:

2 to the power of n, minus 2 (where n is the number of borrowed bits).

However, in an official Cisco video made by Yash Bajpai, and incidentally, one can view it in the CCNA Prep Centre, he mentions in order to calculate the number of subnets one must do the following calculation:

2 to the power of n (where n is the number of borrowed bits).

Whilst I understand that on a Cisco router it is possible to create subnets using Yash Bajpai's method, is it now the correct answer for the exam?

I look forward to hearing from you.

Best wishes,

Ashraf.

Overall Rating: 5 (2 ratings)

## Replies

pciaccio Sun, 04/15/2007 - 08:54

IP Subnetting in itself is very difficult if you do not understand binary to decimal conversion. It took me a long time to get it straight in my 1k RAM in my head. Now that RAM is full... However I have taught many people on how to subnet and others have learned it from other techniques. From my experience if you know and understand IP subnetting and you can regergitate it on an exam, then use it and do not try to learn another technique. If you try to do it another way you may confuse yourself and have difficulties all together in getting results... If you know it use that techniquethat works best for you. The tests do not care how you came up with the answer, just as long as it is correct...Good Luck...

dgahm Sun, 04/15/2007 - 12:49

If you understand subnet zero you will be fine on the test. The questions will either let you know if the zero subnets can or can't be used, or the answers will not include both possibilities.

Early routing protocols like RIP version 1 do not send any subnet mask information with route advertisements. This means that all non classfull networks are assumed to have the same subnet mask. All classfull networks are assumed to have the natural mask. So a classfull network of 172.16.0.0 could never be used as a 8 bit subnet (255.255.255.0). The 172.16.255.0 could also not be used because packets with a destination address of 172.16.255.255 would be handled as classfull network broadcasts. That leaves you with 1-254 as valid subnets or 2 to the power of N minus 2.

Newer routing protocols like EIGRP or OSPF include mask information. If the networks in this example are advertised as 172.16.0.0/16 or 172.16.0.0/24 there is no longer a problem with using all 2 to the power of N subnets. You do need to have the IP subnet zero comand in the router.

Dave

ashraf_can2 Sun, 04/15/2007 - 19:21

Dear Dave,

Thank you very much again.

Kind regards,

Ashraf.

platignum Sun, 04/15/2007 - 12:56

Hi Ashraf,

I remember doing this last year - CCNA exam it was. The lecturer at university where I took the exam advised using the 2 to the power of n -2 for the cisco exam, although the use of all-ones subnet and subnet zero has been explicity allowed since the release of IOS v12.0.

The following RFC link will give you more background if you're interested.

http://www.ietf.org/rfc/rfc1878.txt

Hope this is helpful to you

good luck with the exam

Janet

ashraf_can2 Sun, 04/15/2007 - 19:23

Dear Janet,

Thank you very much for your reply. I appreciate very much the link you gave me. It's really helped.

Wish you all the best,

Ashraf.