Subnetting Explained: A Beginner’s Guide with Practice Questions

Priya had aced every quiz in her networking class. Then the interviewer slid a whiteboard marker across the table and said, “Walk me through subnetting 192.168.1.0 for four equal networks.” She froze. The math she’d memorised the night before just evaporated.

That gap, between recognising subnetting and actually doing it under pressure, is where most beginners get stuck.

Here’s the good news. Subnetting isn’t hard. It feels hard because most guides bury you in binary before they tell you what you’re even trying to do. So let’s flip that around.

By the end of this guide, you’ll know what subnetting is, why it exists, and a shortcut method that lets you subnet in your head. You’ll also get a CIDR cheat sheet, a walkthrough of VLSM for when your subnets need different sizes, and 20 practice questions with full answers. No calculator needed. That’s the whole point.

Quick promise: if you can do basic arithmetic, you can subnet.

One thing before we start. Subnetting sticks through reps, not reading. So keep our CCNA subnetting practice workbook open and work the drills as you go. It’s built around the exact addressing questions the 200-301 throws at you, with full answer breakdowns. Now let’s get into it.

What Is Subnetting? (The Plain Answer)

Subnetting is splitting one big network into smaller networks.

That’s it. That’s the core idea.

Think of a large office building with one giant open floor. Everybody’s in the same room, shouting over each other, and you can’t control who hears what. Subnetting is putting up walls. You turn that one floor into separate rooms, each with its own address, its own space, and its own door.

In networking terms, you take a single block of IP addresses and carve it into smaller blocks called subnets. Each subnet is its own little broadcast domain. Traffic stays contained. Security improves. And you stop wasting addresses on devices that don’t need them.

Why does anyone care? Three reasons.

• Control: You decide which devices share a network and which stay separated.
• Efficiency: A point-to-point link between two routers needs 2 addresses, not 254. Subnetting stops the waste.
• Performance: Smaller broadcast domains mean less noise on the wire.

Subnetting shows up everywhere in the field, and it’s a graded skill on the Cisco CCNA. The current 200-301 exam still asks you to configure and verify IPv4 subnetting cold, and there’s no on-screen calculator during the test. If you want to pass, mental subnetting isn’t optional.

Want to practise this the way the exam tests it? Our CCNA workbook’s subnetting section gives you a full set of timed drills, so you build the speed before exam day, not during it.

The Building Blocks You Actually Need

Before you subnet, you need three things straight: the IP address, the subnet mask, and a tiny bit of binary. Don’t worry. We’ll keep the binary to a minimum.

An IP Address Is Just 32 Bits

An IPv4 address looks like 192.168.1.10. Four numbers, separated by dots. Each number is called an octet, and each octet is 8 bits. Four octets, 8 bits each. That’s 32 bits total.

Each octet can range from 0 to 255. Why 255? Because 8 bits can count up to 255 (and 0 makes 256 possible values). Hold that number in your head. It matters in a minute.

The Subnet Mask Draws the Line

Every IP address has two parts: a network part and a host part. The subnet mask tells you where the network part ends and the host part begins.

A common mask is 255.255.255.0. Read it like this. Where you see 255, that’s network. Where you see 0, that’s host. So in 192.168.1.10 with a 255.255.255.0 mask, the network is 192.168.1 and the host is 10.

Same building analogy. The network part is the building address. The host part is your specific desk inside.

Binary, the Short Version

Each octet is 8 bits, and each bit is worth a value. From left to right:

Bit position	1	2	3	4	5	6	7	8
Value	128	64	32	16	8	4	2	1

Add them all up: 128 + 64 + 32 + 16 + 8 + 4 + 2 + 1 = 255. There’s your 255 again.

A subnet mask is just a run of 1s followed by a run of 0s. The 1s are network bits. The 0s are host bits. That’s the only binary you truly need for now.

CIDR Notation: The Slash Number

You’ll see addresses written like 192.168.1.0/24. That /24 is CIDR notation, and it’s a shortcut.

The number after the slash counts how many bits are network bits. A /24 means 24 network bits. Since each octet is 8 bits, 24 bits fills the first three octets. That’s the same as a 255.255.255.0 mask. Same thing, shorter to write.

Here are the three defaults beginners see most:

CIDR	Subnet mask	Network bits	Host bits
/8	255.0.0.0	8	24
/16	255.255.0.0	16	16
/24	255.255.255.0	24	8

When you subnet, you push that slash number higher. You borrow bits from the host side and hand them to the network side. Borrow 1 bit from a /24 and you get a /25. Borrow 2 and you get a /26. Each borrowed bit doubles your number of subnets and halves your hosts per subnet.

That trade-off is the heart of subnetting. More networks, fewer hosts each. Your job is finding the right balance.

How to Subnet in Your Head (The Block Size Method)

Forget converting every address to binary. Pros use a faster trick called the block size method, sometimes called the magic number. Here’s the whole thing.

Step 1: Find the block size.
Block size = 256 minus the mask value in the interesting octet.

Step 2: Count subnets in jumps of that block size.
Start at 0 and keep adding the block size.

Step 3: Read off the network, hosts, and broadcast.

Let me show you with a real example.

Worked Example: 192.168.1.0 /26

A /26 mask is 255.255.255.192. The interesting octet is the fourth one, and its value is 192.

Block size = 256 – 192 = 64.

Now count your subnets in jumps of 64: 0, 64, 128, 192. Four subnets. Done.

Take the second subnet, the one starting at 64. Here’s how to read it:

• Network address: 192.168.1.64 (the start of the block)
• First usable host: 192.168.1.65 (one above the network)
• Last usable host: 192.168.1.126 (one below the broadcast)
• Broadcast address: 192.168.1.127 (one below the next block, which starts at 128)

That’s the full pattern. The network address and the broadcast address can’t be assigned to devices, so usable hosts always sit between them.

How many usable hosts per /26? You’ve got 6 host bits left (32 minus 26). So 2 to the power of 6 is 64, minus 2 for the network and broadcast addresses. That’s 62 usable hosts.

Remember Priya from the start? This is the exact question she froze on. Four equal networks from a /24 means borrowing 2 bits, which gives you a /26, block size 64, subnets at 0, 64, 128, 192. Thirty seconds, no binary, no calculator.

If you can run those three steps, you can subnet most CCNA-level questions. The trick is reps. Speed comes from doing 20 of these, not reading about one. The graded subnetting drills in our CCNA workbook build that muscle memory the way the exam demands, from easy warm-ups to full VLSM scenarios.

The Subnet Mask Cheat Sheet

Print this. Tape it to your monitor. These are the masks you’ll meet inside a /24, where most subnetting questions live.

CIDR	Subnet mask	Block size	Subnets in a /24	Usable hosts
/24	255.255.255.0	256	1	254
/25	255.255.255.128	128	2	126
/26	255.255.255.192	64	4	62
/27	255.255.255.224	32	8	30
/28	255.255.255.240	16	16	14
/29	255.255.255.248	8	32	6
/30	255.255.255.252	4	64	2

Notice the pattern. As the CIDR number climbs, the block size halves and so do your usable hosts. The /30 at the bottom gives you exactly 2 hosts, which is perfect for a link between two routers. Nothing wasted.

You don’t have to memorise all of this. If you remember the powers of 2 (2, 4, 8, 16, 32, 64, 128, 256) and the block size formula, you can rebuild the whole table from scratch in under a minute.

Worked Example #2: Subnetting in the Third Octet

Beginners get comfortable subnetting the fourth octet, then panic when the interesting octet moves. Let’s defuse that now.

Take 172.16.0.0 /22.

A /22 mask is 255.255.252.0. The interesting octet here is the third one, and its value is 252.

Block size = 256 – 252 = 4.

So your subnets jump by 4 in the third octet: 172.16.0.0, 172.16.4.0, 172.16.8.0, and so on.

Look at the 172.16.4.0 subnet:

• Network: 172.16.4.0
• First host: 172.16.4.1
• Last host: 172.16.7.254
• Broadcast: 172.16.7.255

The broadcast sits one below the next block, which starts at 172.16.8.0. Same logic as before, just in a different octet. The method never changes. Only the column you’re working in changes.

This is the moment subnetting clicks for most people. Once you trust the block size method works in any octet, the fear drops away.

VLSM: When Subnets Need Different Sizes

Every example so far splits a network into equal pieces. Real networks rarely need that. Your Sales floor might have 50 people. Your server room might have 4 devices. A router-to-router link needs exactly 2 addresses. Handing every one of those a same-sized subnet wastes hundreds of addresses on the link alone.

Variable Length Subnet Masking, VLSM for short, just means: subnet the same network more than once, at different mask lengths, so each piece is sized for what it actually needs. It’s the same block size method you already know. You just run it more than once.

The one rule that matters: always size the biggest requirement first, then work down to the smallest. If you start small, you’ll carve up address space that a bigger chunk needed later, and you’ll have to start over.

Worked Example: VLSM on 192.168.20.0 /24

Say you’re handed 192.168.20.0/24 and told to build four subnets for these needs:

• Sales: 50 hosts
• Engineering: 25 hosts
• Guest Wi-Fi: 10 hosts
• Router link between two branch offices: 2 hosts

Step 1: Sort biggest to smallest. Sales (50), Engineering (25), Guest Wi-Fi (10), router link (2).

Step 2: Size Sales first. You need 2 to the power of n, minus 2, to be at least 50. Six host bits gives 62 usable, which is the smallest mask that covers 50. Six host bits means a /26. Block size 64. Sales gets 192.168.20.0/26, network .0, usable hosts .1 to .62, broadcast .63.

Step 3: Size Engineering next, starting right after Sales ends. The next available address is 192.168.20.64. You need at least 25 hosts, so 5 host bits (30 usable) is the smallest fit. Five host bits means a /27, block size 32. Engineering gets 192.168.20.64/27, network .64, usable hosts .65 to .94, broadcast .95.

Step 4: Size Guest Wi-Fi next, starting at the next free address. That’s 192.168.20.96. You need 10 hosts, so 4 host bits (14 usable) fits. Four host bits means a /28, block size 16. Guest Wi-Fi gets 192.168.20.96/28, network .96, usable hosts .97 to .110, broadcast .111.

Step 5: Size the router link last. Next free address is 192.168.20.112. A point-to-point link only ever needs 2 usable hosts, so 2 host bits (2 usable) fits exactly. Two host bits means a /30, block size 4. The link gets 192.168.20.112/30, network .112, usable hosts .113 to .114, broadcast .115.

Add it up and you’ve used addresses .0 through .115, roughly 45% of the original /24, with every subnet sized to what it actually needs instead of one wasteful blanket mask. That’s the entire payoff of VLSM: no wasted addresses, and every subnet is exactly as big as its job requires.

Want more VLSM reps before exam day? The CCNA workbook includes a full set of VLSM design scenarios, not just the equal-split kind.

20 Subnetting Practice Questions (With Answers)

Cover the answers. Work each one with pen and paper. No calculator, just like the real exam. Answers and full explanations follow the list.

1. How many usable host addresses are on a /29 network?
2. What is the subnet mask for a /27 in dotted-decimal form?
3. Which subnet does the host 192.168.10.70 belong to if the mask is /26? Give the network and broadcast addresses.
4. What is the broadcast address for 172.16.4.0 /22?
5. You need at least 50 hosts per subnet. What is the smallest mask (largest CIDR number) that works?
6. You borrow 3 host bits from a /24. How many subnets do you create?
7. On the network 192.168.1.0 /26, is 192.168.1.63 a valid host address? Why or why not?
8. Convert the mask 255.255.255.240 to CIDR notation.

Answers and Explanations

1. Six usable hosts.
A /29 leaves 3 host bits. 2 to the power of 3 is 8, minus 2 for the network and broadcast addresses. 8 – 2 = 6.

2. 255.255.255.224.
A /27 means 27 network bits. The first three octets use 24, leaving 3 bits in the fourth octet: 128 + 64 + 32 = 224.

3. The 192.168.10.64 subnet. Network 192.168.10.64, broadcast 192.168.10.127.
Block size for a /26 is 256 – 192 = 64. Subnets fall at .0, .64, .128, .192. The host .70 sits in the .64 block (which runs 64 to 127). So the network is .64 and the broadcast is .127.

4. 172.16.7.255.
Block size for a /22 is 256 – 252 = 4, in the third octet. The 4.0 subnet covers 172.16.4.0 through 172.16.7.255. The broadcast is the top of that range: 172.16.7.255.

5. A /26.
You need 2 to the power of n, minus 2, to be at least 50. With 5 host bits you get 30 (too few). With 6 host bits you get 62 (enough). Six host bits means 32 – 6 = 26 network bits, so /26.

6. Eight subnets.
Borrowing 3 bits gives you 2 to the power of 3 combinations. 2 cubed is 8.

7. No. It’s the broadcast address.
Block size for a /26 is 64. The first subnet runs 192.168.1.0 through 192.168.1.63. The .63 is the broadcast for that block, so it can’t be assigned to a device.

8. /28.
The 240 in the fourth octet is 128 + 64 + 32 + 16, which is 4 bits. 24 + 4 = 28.

How did you do? If you nailed 6 or more, your foundation is solid. If a few tripped you up, that’s normal at this stage. The fix is more reps, not more reading. Not sure which cert path fits you? Our CCNA vs Network+ comparison breaks down both. Our Network+ practice test includes addressing questions if you’re prepping for CompTIA instead of Cisco, and the CCNA workbook goes deeper on Cisco-style subnetting scenarios.

Intermediate

9. A /24 network needs to support 6 equal-sized subnets. What’s the smallest mask that provides at least 6 subnets, and how many hosts does each get?
10. What is the network address for the host 10.10.55.130 with a /25 mask?
11. You’re given 172.20.0.0/20. What’s the block size, and in which octet does it apply?
12. Convert /29 to a dotted-decimal subnet mask.
13. A device has the address 192.168.5.29/28. What’s the valid host range for its subnet?
14. How many total subnets can you create by borrowing 4 bits from a /24?
15. What’s the broadcast address for the network 10.0.16.0/22?

Answers and Explanations

9. A /27, 30 usable hosts per subnet.
You need at least 6 subnets. Borrowing 2 bits gives 4 subnets (too few). Borrowing 3 bits gives 8 subnets (enough). Three borrowed bits from a /24 gives a /27, leaving 5 host bits, so 2 to the power of 5 minus 2 equals 30 usable hosts.

10. 10.10.55.128.
A /25 mask is 255.255.255.128, block size 128 in the fourth octet. Subnets fall at .0 and .128. The host .130 sits in the .128 block, so the network address is 10.10.55.128.

11. Block size 16, in the third octet.
A /20 mask is 255.255.240.0. The interesting octet is the third one, with a value of 240. Block size = 256 – 240 = 16.

12. 255.255.255.248.
A /29 means 29 network bits, leaving 3 in the fourth octet: 128 + 64 + 32 = 248.

13. 192.168.5.17 to 192.168.5.30.
A /28 mask gives a block size of 16. Subnets fall at .0, .16, .32, and so on. The address .29 sits in the .16 block, which runs .16 to .31. The network is .16, broadcast is .31, so usable hosts run .17 to .30.

14. 16 subnets.
Borrowing 4 bits gives 2 to the power of 4 combinations. 2 to the fourth power is 16.

15. 10.0.19.255.
A /22 mask is 255.255.252.0, block size 4 in the third octet. Subnets fall at .16, .20, and so on. The address 10.0.16.0 starts a block that runs through 10.0.19.255, which is the broadcast.

Advanced (VLSM and design)

16. You’re handed 192.168.30.0/24 and must build 3 subnets sized for 100 hosts, 40 hosts, and 20 hosts, in that order. What CIDR mask does each subnet get?
17. Using the same 192.168.30.0/24 network from question 16, what is the network address of the third subnet (the 20-host one), assuming the 100-host and 40-host subnets are allocated first?
18. Why does VLSM require you to size the largest subnet first?
19. A network engineer needs two subnets of exactly 2 usable hosts each, for two separate router links, carved out of 10.5.5.0/29. Is this possible? Why or why not?
20. You’ve been given 172.16.8.0/22 and need one subnet supporting 200 hosts and four subnets supporting 2 hosts each (for point-to-point links). List all five subnets with their CIDR masks.

Answers and Explanations

16. /25 for 100 hosts, /26 for 40 hosts, /27 for 20 hosts.
For 100 hosts you need 2 to the power of n minus 2 to be at least 100; 7 host bits gives 126, so a /25. For 40 hosts, 6 host bits gives 62, so a /26. For 20 hosts, 5 host bits gives 30, so a /27.

17. 192.168.30.192.
The 100-host subnet (/25) takes 192.168.30.0 through .127. The 40-host subnet (/26) takes the next block, 192.168.30.128 through .191. The 20-host subnet starts right after, at 192.168.30.192.

18. Because VLSM subnets are carved out sequentially, in order, from the remaining address space.
If you size a small subnet first, it can land in the middle of the address block that a larger, later subnet needs as one continuous range. Sizing biggest to smallest guarantees every later subnet still has enough contiguous room.

19. No, not as written.
A /29 only has 8 addresses total and 6 usable hosts as a single subnet; it can’t be split further into two usable point-to-point subnets without borrowing more bits, and there’s no room left to borrow inside a /29. You’d need to start from at least a /28 to carve out two /30 links (4 usable… 2 each) alongside anything else.

20. 172.16.8.0/24 for 200 hosts, then 172.16.9.0/30, 172.16.9.4/30, 172.16.9.8/30, and 172.16.9.12/30 for the four links.
The 200-host requirement needs 2 to the power of n minus 2 of at least 200, which takes 8 host bits (254 usable), a /24. That consumes the entire first half of the /22 (172.16.8.0 to 172.16.8.255). The four point-to-point links each need only 2 usable hosts, a /30 with block size 4, and are carved from the next available space starting at 172.16.9.0.

How did you do? If you nailed 14 or more, your foundation is solid, VLSM included. If a few tripped you up, that’s normal at this stage. The fix is more reps, not more reading. Not sure which cert path fits you? Our CCNA vs Network+ comparison breaks down both. Our Network+ practice test includes addressing questions if you’re prepping for CompTIA instead of Cisco, and the CCNA workbook goes deeper on Cisco-style subnetting scenarios. Want even more reps? SMEnode Academy’s 50-question subnetting practice set picks up where these 20 leave off, timed and tiered for exam-day speed.

Common Subnetting Mistakes Beginners Make

Marcus studied subnetting for a month before his CCNA attempt. He could solve every problem at home. Then he sat the exam, leaned on the calculator habit he’d never broken, and ran out of time on the addressing questions. He missed the cut by a few points. The math was never his problem. Speed was.

Here’s what trips people up most.

Forgetting the minus 2. Total addresses in a subnet and usable hosts are not the same number. Always subtract the network and broadcast addresses.

Assigning the network or broadcast address to a device. The first address in a block is the network ID. The last is the broadcast. Neither goes on a computer.

Mixing up subnets and hosts. Borrowing bits adds subnets and removes hosts. If a question asks for more networks, you push the CIDR number up. If it asks for more hosts per network, you pull it down.

Working in the wrong octet. Find the interesting octet first, the one where the mask isn’t 255 or 0. That’s where your block size lives.

Leaning on a calculator. Tools are fine while you learn the concept. But the CCNA gives you no calculator, so wean yourself off early. Drill until block sizes are automatic.

The career payoff is real, by the way. Network and systems administration roles in the US sit around a $95K median wage per the Bureau of Labor Statistics, and subnetting fluency is one of the gatekeeper skills hiring managers screen for. Get it solid and a lot of doors open.

Frequently Asked Questions

Bottom Line

Subnetting is splitting one network into smaller ones, and the block size method makes it something you can do in your head.

Here’s what to take away:

• A subnet mask separates the network part of an address from the host part.
• CIDR (the slash number) counts network bits. Higher number, smaller subnet.
• Block size = 256 minus the mask value in the interesting octet.
• Usable hosts = 2 to the power of host bits, minus 2.
• Speed beats theory. Drill, don’t just read.

The learners who pass don’t have special brains. They just put in 20 minutes a day until the patterns became automatic. You can get there in a couple of weeks.

Ready to build that speed? Our CCNA practice workbook gives you timed subnetting drills modelled on the real 200-301 exam, with full answer breakdowns. And if you’d rather learn live with an instructor walking you through it, SMEnode Academy’s CCNA training covers subnetting hands-on. Pick your path and start drilling today.