What is CIDR and Why It Is Important

In short, what is CIDR is a modern way to describe networks and subnets without strict reference to outdated “classes” A/B/C. The transcript will be below, but for now — the essence in simple words.
The CIDR says: “Let’s take an IP address and add a fraction with the number of network bits after it,” for example, 192.168.10.0/24. This is the “mask” in compact notation: /24 means that the first 24 bits are the network, and the remaining 8 are the hosts.

Why did CIDR appear after classes A/B/C? The class system was crude: class A — giant networks, B — large, C — small. I often had to take the network “to grow” and waste thousands of addresses. The Internet was growing, IPv4 was rapidly running out, and routing tables were swelling. Flexibility was needed.

CIDR gave it to us: instead of ready—made A/B/C “bricks”, we began to “cut” the network exactly for the task – at least /30 for point-to-point, at least /27 for the office, at least /19 for the data center. Plus— route aggregation, which has made life easier for backbone routers.

Where is CIDR used?

• In Internet routing (BGP, prefix aggregation).
• In the network settings of servers, routers, and clouds.
• In hosting and cloud infrastructure (VPC/VNet, subnets, ACLs).
• In the daily work of the admins: “should I assign subnet /26 to a new department? Not a question.”

Spoiler to the following sections: when you ask what is a CIDR block, we are talking about just such a “segment of the address space” written in CIDR notation (for example, 10.0.0.0/12).

What Does CIDR Mean?

What does CIDR mean? Literally— “Classless Inter-Domain Routing”. The keyword is classless: we stopped thinking in terms of A/B/C and switched to an accurate marking of the network by the number of bits.

How CIDR changed the rules of the game:

• Flexible addressing. Instead of coarse—grained classes, use any prefix from /0 to /32. Do you need 256 addresses? We take /24. Do we need 32 addresses? /27.
• Saving IPv4. There are few addresses, and CIDR minimizes the “air” between networks, giving out exactly as much as needed.
• Route aggregation. Several “small” networks can be represented by one large prefix. Routers see fewer rows in the table — the Internet breathes more freely.

The math is simple: block size = 2^(32 − prefix) for IPv4. For example, /24 gives 256 addresses, /29 – 8 addresses, /16 — 65 536 addresses. In practical terms, this means precisely fitting the network to a department, project, VLAN, or cluster.

What Does CIDR Stand For?

To the direct question of what does CIDR stand for, we answer: Classless Inter-Domain Routing. However, it is important to distinguish between a concept and a notation:

• CIDR as a concept — abandoning class addressing and switching to a variable prefix length (VLSM). This is about the architecture and principles of address/route allocation.
• CIDR notation is a practical form of writing “IP + /prefix”, for example 203.0.113.0/25. This is short, unambiguous, and understandable to any networker.

Why is it important to understand both sides? Because in real work, you plan the address space (concept) and write specific prefixes in configurations (notation). When you ask what is a CIDR block, you are actually talking about the “address allocation/accounting unit” in this system — exactly the same prefix that can be routed, divided into subnets, or aggregated back.

A simple example of the difference:

• The concept will tell you that it is better to “cut” /16 into several /24 for different departments, so as not to generate wide broadcasts and simplify the access policy.
• Notation is already specific notation: 10.12.0.0/24, 10.12.1.0/24, 10.12.2.0/24…

Total: knowing what is CIDR, what does CIDR mean and what does CIDR stand for, you speak the language in which the modern Internet “communicates” — from home routers to backbone BGP routers. In order to understand the topic in the better way you should understand what an IP address is and how it works.

Historical Background: From IP Classes to CIDR

To understand what CIDR is, you need to go back in time. Until 1993, the Internet lived according to the system of “IP classes”:

• Class A – networks with ~16 million addresses (/8).
• Class B — networks with ~65 thousand addresses (/16).
• Class C — a total of 254 addresses (/24).

The system was convenient on paper, but absolutely wasteful. Imagine a startup that needs only 500 addresses. Class C (254 addresses) is too small, but the closest option is Class B (65,536 addresses). The result is tens of thousands of “lost” IP addresses.

With the growth of the Internet, this became a disaster-scale problem: routing tables were swollen, IPv4 was running out. The solution was the introduction of CIDR (Classless Inter-Domain Routing) in 1993. It made it possible to divide addresses into blocks of any length — flexibly, economically and scalably.

What is a CIDR Block?

Now to the specifics. What is a CIDR block? This is a unit that describes the range of IP addresses in the CIDR format.

Example: 192.168.0.0/24 is a CIDR block. It means that the first 24 bits are fixed for the network, and the last 8 can be used for hosts. The total is 256 addresses, from 192.168.0.0 to 192.168.0.255.

What is the difference between a CIDR block and an IP range?

• The range (192.168.0.0 – 192.168.0.255) describes addresses “from below and above”.
• The CIDR block (192.168.0.0/24) describes the same thing, but in the “mathematical form” of a prefix.

Thus, a CIDR block is not just a list of addresses, but a structured unit that routers, BGP, and administrators.

Understanding CIDR Notation

To read blocks, you need CIDR notation, which is the format for writing an IP address + a slash + the number of prefix bits.

Examples:

• /8 → 16.7 million addresses (for example, 10.0.0.0/8).
• /16 → 65 536 addresses (for example, 172.16.0.0/16).
• /24 → 256 addresses (for example, 192.168.1.0/24).
• /32 → exactly 1 address (203.0.113.45/32).

How can I calculate the number of addresses in a CIDR block?
The formula is simple: 2^(32 − prefix) for IPv4.
For example:

• /24 → 2^(32-24) = 256.
• /29 → 2^(32-29) = 8.

That is why when you ask the question “what is a CIDR block”, the correct answer always includes CIDR notation. Without this format, the block remains just a “smeared range”, and with it — an understandable building block for the network.

CIDR Prefix	Number of IP Addresses	Typical Use Case
/8	16,777,216	Very large networks, ISPs
/16	65,536	Corporate networks, big sites
/24	256	Standard LAN, home/office use
/29	8	Small subnets, point-to-point
/32	1	Single host address

How CIDR Improves IP Address Allocation

Saving IPv4 addresses. CIDR solves the main pain of the old class scheme: we stop taking fixed-size “bricks” and give out exactly as many addresses as needed. Do you need ~300 hosts? Instead of “class B” for 65,536 addresses, we take /23 (512 addresses) and that’s it. Do you need 28 cameras? Suitable for /27 (32 addresses). The formula is simple for IPv4: number of addresses = 2^(32−prefix).
Examples: /24 = 256, /25 = 128, /26 = 64, /29 = 8.

Route aggregation. CIDR allows you to “glue” many small networks into one large announcement. Instead of four routes 203.0.113.0/24, .1/24, .2/24, .3/24, the provider announces one 203.0.112.0/22, and the routing tables lose weight dramatically. This is critical for backbone BGP: fewer records means faster convergence and less memory for routers.

Convenience of network planning (VLSM). CIDR = variable mask length (VLSM). We can “slice” the space hierarchically:

• we issue 10.20.0.0/16 to the company;
• each branch has /20 (4096 addresses);
• there are several /24s for VLANs inside the branch (office, telephony, printers, IoT), and /26s for guests.

This design is easy to scale, clearly documented, and simplifies ACL and routing (you can summarize it back to the branch).

CIDR in IPv6: Does It Work the Same Way?

Yes, the principle is the same. CIDR for IPv6 works the same way: address + “/prefix”, only the space is orders of magnitude larger. The formula is 2^(128 is the prefix). But there are practical rules.

The difference in scale. In IPv6, it is customary to give:

• home users /56 (this is 256 subnets /64),
• organizations — /48 (this is 65,536 subnets /64),
• one L2 network/ VLAN is almost always /64 (needed for SLAAC).
• /64 is a huge pool (≈1.8×101⁹ addresses), but this is the norm. For point-to-point links, they use /127 so as not to “scatter”.

Why the principle has remained the same. We still:

• use prefixes for flexible address allocation (saving is no longer the goal, structuring is more important),
• summarize routes,
• we are designing a hierarchy (for example, 2001:db8:1234::/48 → departments by /56 → VLANs by /64).

In other words, “what is CIDR” in IPv6 is the same planning language, only with a much more spacious vocabulary.

Practical Examples: How to Use CIDR in Networks

Home network: 192.168.1.0/24.
The classic router distributes one subnet /24 to everything: laptops, phones, TV. But CIDR allows you to quickly restore order:

• 192.168.1.0/26 (hosts) — 62 addresses;
• 192.168.1.64/27 (IoT) — 30 addresses;
• 192.168.1.96/28 (guests) — 14 addresses;
• 192.168.1.112/28 (printers/NAS) — 14 addresses;
  — and we leave the rest in reserve. This makes it easier to write firewall rules (“guests can’t see the NAS”) and look for problems.

Corporate network: partitioning into subnets. We allocate the company 10.50.0.0/16. For each office — /20, so that there are enough addresses with a margin. In the office:

• 10.50.16.0/24 — users,
• 10.50.17.0/24 — telephony/VoIP,
• 10.50.18.0/26 — printers,
• 10.50.19.0/25 — the server room.

Next, we summarize the office /24 into the parent /20 and declare it in the core — the tables are cleaner, the access policy is shorter.

Providers and hosting. The data center received 203.0.112.0/21. Clients are given a /24 or /23, and an aggregated /21 is announced outside in BGP. Everything is the same in the clouds.:

• VPC on AWS: 10.0.0.0/16,
• Subnets on AZ — /24,
• An important rule is not to cross CIDR with on—prem (otherwise IPsec/Direct Connect will suffer).

Where “what is a CIDR block” is literally an entity in settings (VPC CIDR, Subnet CIDR), the convenience of CIDR is felt by the hands every day.

Common Mistakes When Understanding What is CIDR

1. Confusion between the IP address and the block. A common mistake is to think that 192.168.1.10/24 “is one address”. In fact, after the slash is the size of the network, not the “part of the address”. The question of what is a CIDR block is just about the range described by the prefix: 192.168.1.0/24 is 256 addresses (from .0 to .255), not one host.
2. Errors in calculating the network size. Classic: confusing “total number of addresses” and “number of hosts”. The formula for IPv4 is: addresses = 2^(32 is the prefix). There are only 8 addresses in /29, but there are usually 6 hosts (if the network uses network/broadcast). Point-to-point can use /31 (2 addresses without broadcast) — many people forget about this optimization. There is another trap in IPv6: assigning end segments to something already /64. For SLAAC and normal compatibility, /64 is the norm; attempts to give /120 break the auto-configuration.
3. A mix of CIDR and the old class system. Some people still think in “classes” A/B/C and try to “fit” the mask to the class. CIDR was created precisely to get away from class rigidity. 10.0.0.0/8 does not mean “class A”, it’s just the prefix /8. Another trap is to equate the dotted—decimal mask to the prefix and be one step wrong: 255.255.255.192 is /26, not /25. Check using the formula and binary representation.
4. Incorrect summation and intersections. When aggregating (“gluing”) prefixes, it is easy to assemble a “beautiful /22” from two non—adjacent /23 – routers will not understand this. Even worse, you can set overlapping subnets in the cloud or on-prem: routing will become unpredictable. Remember: CIDR blocks must be adjacent and aligned along the prefix boundary.
5. Ignoring the IPv6 context. In IPv6, many people ask what is CIDR and adopt IPv4 habits. The main rule is: subnet = /64, and aggregation is done higher (for example, /56, /48). Prefixes are already /64 — an exception for special links (for example, /127 for p2p).

Mistake	Why It Happens	Correct Approach
Mixing up IP vs. CIDR block	Confusing one host with a subnet	Remember: CIDR block = range, not a single
Wrong subnet size calculation	Miscounting usable addresses	Use formula: 2^(32−prefix)
Sticking to old IP classes	Thinking in A/B/C terms	CIDR is classless, no fixed categories
Overlapping or misaligned subnets	Creating ranges that conflict	Align prefixes properly, avoid overlaps
Misusing IPv6 prefixes	Applying IPv4 logic to IPv6	Use /64 for subnets, larger for summary

Frequently Asked Questions