You are designing a subnet mask for the 10.0.0.0 network. You want 2100 subnets with up to 3200 hosts on each subnet. What subnet mask should you use?

*How do I use this chart when the hosts and networks it is asking for are larger than the numbers within the chart? Can someone please help me with this?*

Thanks.

Sorry, I couldn't do superscripts in this post.

**Use the formula - (2[to the]n) - 2**

**Here, 'n' is the number of bits assigned to the host portion of the subnet mask.**

For instance a \24 would have (2 [to the]8) - 2 or

256 - 2 or

254 usable addresses.

For this example, as has been pointed out, you would need to expand the chart to do a \18 or whatever give you your correct number of hosts.

By the same token, you can use the same formula to figure out the number of subnets by using the network portion of your mask as 'n'.

## 23 Replies

No, I got it from here

http://subnet19.blogspot.com/2013/09/subnetting-based-on-host-and-network.html

This is similar to a question from last summer. Here is the explination of how subnet bits work.

https://community.spiceworks.com/topic/post/4718010

To aid in answering the question above, you don't have the right section of the chart to answer the question. A /25 subnet will only give you 253 available hosts. If you want 3200 hosts you will need to shift the subnet mask to the left a few positions.

The problem with your chart is it doesn't extend far enough to the left. If you notice the # of hosts is basically doubling every time you go left (but you subtract 2, 1 for the network 1 for the broadcast).

Take a look at this image:

http://oldsite.cis131.com/cis131old/images/chapter15/IPsheet.jpg

In the real world, I have only seen /22 thru /30. For testing, I suggest you learn how to do the math. Unless you can memorize things easily, then just memorize it all.So do I need to have it memorized for every CIDR position?

Then when you get old like me you just use the online subnet calculators http://www.subnet-calculator.com/ so you can focus on the results and not the journey. But in the beginning understanding how ip addresses mix with the subnet masks to produce a device address IS important. Being able to do all of this in your head isn't. If you can first remember the old class based addresses that will help 10.0.0.0/8 (256 networks, 2M hosts), 172.16.0.0/16 (64K networks, 64K hosts), 192.168.1.0/24 (2M networks, 256 hosts) and then just bit shift by either doubling or halving the number of hosts and networks as you move the subnet mask to the left or the right (ie /24 -> /23)

I'm not in school. I'm studying for the CCENT.

The problem is (why your are getting so much hate on this thread) is you posted a question that was out of a text book (or a web site) that nobody would ever dream of doing in the real world. You didn't explain it, just wanted help. To us, it looked like a homework question (the fact it came from a web site asking you to do something very odd is similar to a homework question). They want you to figure it out because you understand subnetting.

You should have posted that you are trying to understand subnetting. You should let us know what you understand, what you don't. If you know up to what the chart you posted but don't know how to get larger than a /24, then you need to state that.

There are threads here (I have posted in them) that help people learn how to subnet and how to break down the various parts of the IP address to determine network size and number of hosts.

One of the first thing you should do when taking tests like this is write this chart out on the sheet of paper you brought with you(they still let you do that, right?) and you can use it for quick reference throughout your exam. Above posters are correct, you will need to write out another 8 bits back to a /17 to get what you are looking for.

However, knowing this info off the top of your head comes in handy in the real world. You may not see private networks bigger than a /22, but the Internet is much bigger than that, and some IT professionals deal with that all the time for route summerization, supernetting, etc.

Like any task, the more you do it the easier it becomes. I used to do an hour a day subnetting when I first started networking. I kept this up for three years. I was sick of it by the end. BUT... I can do almost any subnetting or summarisation task in my head before colleagues have even found and on-line calculator and typed in the numbers.

In the exam you will be given a dry wipe marker pen and a piece of plasticised paper for notes and stuff. Draw out that chart as it is all you need, other than adding another octet or maybe two. Whilst on the subject of the exam, subnetting questions are what they are. The key thing is the longer it takes you the less time you have for the lab questions (when you get there) and they will test your knowledge of a LOT more than just subnetting. You can really burn some time there, especially if your subnetting isn't second nature and you CANNOT pass the CCNA (or CCENT part 2 or whatever) if you don't get the lab questions close to perfect.

Final thought... I wouldn't bother with the CCENT. Go straight for a CCNA. The CCENT part two is pretty much the same as the CCNA anyway so all you're doing is spending double the amount of time and effort to get to the same destination (CCNA) which is just crazy. You won't learn any more by taking the two exams, it's just lost time and more stress. My provenance here is I used to teach the CCNA and CCNP.

Sorry, I couldn't do superscripts in this post.

**Use the formula - (2[to the]n) - 2**

**Here, 'n' is the number of bits assigned to the host portion of the subnet mask.**

For instance a \24 would have (2 [to the]8) - 2 or

256 - 2 or

254 usable addresses.

For this example, as has been pointed out, you would need to expand the chart to do a \18 or whatever give you your correct number of hosts.

By the same token, you can use the same formula to figure out the number of subnets by using the network portion of your mask as 'n'.

In the OP's defense he is just starting to learn about networking, hence the 'studying for CCENT' and he is just trying to understand how to accomplish this. It really doesn't matter that no one would put 3200 hosts on a subnet. I think it is good to understand how to do the math and calculate when starting out. Use the subnet calculator when you are in the real wprld or after a thorough understanding. Good luck and keep studying.

In the OP's defense he is just starting to learn about networking, hence the 'studying for CCENT' and he is just trying to understand how to accomplish this. It really doesn't matter that no one would put 3200 hosts on a subnet. I think it is good to understand how to do the math and calculate when starting out. Use the subnet calculator when you are in the real wprld or after a thorough understanding. Good luck and keep studying.

I agree. When I was a kid, I would ask my dad some pretty outlandish questions 1) to gain further understanding and 2) test my current understanding. A hypothetical situation is just that......hypothetical. The principles applied to said hypothetical apply to the outlandish and the realistic.

This is a question that could have been answered with 30 seconds or less of Googling ip subnets. That's something that one needs to know how to do if you're going to be in this field.

Google ip subnets and the first link would have answered it. (It's down right now, but click the drop down and select cached.)

A better question to post in a technical forum is something along the lines of, I'm struggling with the concept of netmasks and subnetting. Can someone point me towards a good resource that explains it in an easy to understand manner?

OP basically posted "Here's my homework assignment".