CIDR (Classless Inter-Domain Routing)

“CIDR replaced the rigid class system with flexible, efficient address allocation.”

Overview

CIDR (Classless Inter-Domain Routing, pronounced “cider”) replaced the classful addressing system in 1993. It allows variable-length subnet masks (VLSM), enabling efficient IP address allocation and route summarization. CIDR notation uses a suffix (e.g., /24) to indicate the number of network bits.

Classful vs CIDR

graph TD
    subgraph "Classful (Legacy)"
        CF_A["Class A: /8 always<br/>16M hosts — too many!"]
        CF_B["Class B: /16 always<br/>65K hosts — still too many!"]
        CF_C["Class C: /24 always<br/>254 hosts — too few!"]
        CF_Waste["Problem: Huge address waste"]
    end
    
    subgraph "CIDR (Modern)"
        CIDR1["Any prefix length<br/>/8 to /30<br/>Exact fit for needs"]
        CIDR2["Route aggregation<br/>Smaller routing tables"]
        CIDR3["Efficient allocation<br/>No waste"]
    end

CIDR Notation

Format

IP_address/prefix_length

Example: 192.168.1.0/24
- 192.168.1.0 = Network address
- /24 = First 24 bits are the network portion
- Subnet mask: 255.255.255.0

Common CIDR Blocks

Route Aggregation (Supernetting)

CIDR enables combining multiple contiguous networks into a single route.

Example

Individual routes:
192.168.0.0/24
192.168.1.0/24
192.168.2.0/24
192.168.3.0/24

Aggregated (supernet):
192.168.0.0/22

4 routes → 1 route

How to Aggregate

graph TD
    subgraph "Finding Common Prefix"
        R1["192.168.0.0/24<br/>11000000.10101000.00000000.xxxxxxxx"]
        R2["192.168.1.0/24<br/>11000000.10101000.00000001.xxxxxxxx"]
        R3["192.168.2.0/24<br/>11000000.10101000.00000010.xxxxxxxx"]
        R4["192.168.3.0/24<br/>11000000.10101000.00000011.xxxxxxxx"]
        AGG["Common prefix: 22 bits<br/>11000000.10101000.000000xx<br/>= 192.168.0.0/22"]
    end
    
    R1 --> AGG
    R2 --> AGG
    R3 --> AGG
    R4 --> AGG

Rules for aggregation:

1. Networks must be contiguous (no gaps)
2. First address must be aligned on a power-of-2 boundary
3. Number of networks must be a power of 2
4. All share the same common prefix

Longest Prefix Match

When multiple routes match a destination, routers use longest prefix match (most specific route wins).

Routing table:
10.0.0.0/8       → via Router A
10.1.0.0/16      → via Router B
10.1.1.0/24      → via Router C

Destination: 10.1.1.100

All three routes match, but /24 is most specific → Router C

graph LR
    D["Destination<br/>10.1.1.100"] --> M1{"Match /8?"}
    M1 -->|Yes| R1["Router A"]
    M1 -->|Yes| M2{"Match /16?"}
    M2 -->|Yes| R2["Router B"]
    M2 -->|Yes| M3{"Match /24?"}
    M3 -->|Yes| R3["Router C ✓<br/>Most specific"]

CIDR and the Internet

Internet Routing Table Growth

graph LR
    subgraph "BGP Table Size (Approximate)"
        Y1994["1994: ~20,000"]
        Y2000["2000: ~100,000"]
        Y2010["2010: ~350,000"]
        Y2020["2020: ~850,000"]
        Y2024["2024: ~1,000,000+"]
    end
    
    Y1994 --> Y2000 --> Y2010 --> Y2020 --> Y2024

CIDR helps control routing table growth through aggregation, but:

• Deaggregation: Organizations announce more-specific routes for traffic engineering
• Multi-homing: Connecting to multiple ISPs requires separate route announcements
• Result: Table keeps growing despite CIDR

ISP Allocation Example

ISP receives: 203.0.112.0/20 (4096 addresses)

ISP allocates to customers:
Customer A: 203.0.112.0/24  (256 addresses)
Customer B: 203.0.113.0/24  (256 addresses)
Customer C: 203.0.114.0/23  (512 addresses)
Customer D: 203.0.116.0/22  (1024 addresses)
...

ISP announces single route: 203.0.112.0/20 to Internet

Interview Questions

Beginner

Q1: What is CIDR? CIDR (Classless Inter-Domain Routing) is a method for allocating IP addresses that replaced the old classful system. It uses a suffix (e.g., /24) to specify how many bits are the network portion, allowing flexible subnet sizes. For example, /24 means the first 24 bits are the network (254 hosts), while /28 means 28 bits (14 hosts).

Q2: How is CIDR different from classful addressing? Classful addressing had fixed boundaries: Class A (/8), Class B (/16), Class C (/24). If you needed 300 hosts, you had to get a Class B (65,534 addresses) — wasting 65,000+ addresses. CIDR lets you use /23 (510 hosts) — a perfect fit. CIDR also enables route summarization, reducing routing table sizes.

Q3: What does /24 mean? /24 means the first 24 bits of the 32-bit IP address are the network portion, and the remaining 8 bits are for hosts. The subnet mask is 255.255.255.0. This gives 2^8 - 2 = 254 usable host addresses.

Intermediate

Q4: Explain route aggregation and its benefits. Route aggregation (supernetting) combines multiple contiguous routes into one. For example, four /24 routes (192.168.0.0/24 through 192.168.3.0/24) can be summarized as 192.168.0.0/22. Benefits: (1) Smaller routing tables → faster lookups, (2) Less memory in routers, (3) Reduced BGP update traffic, (4) Simpler management.

Q5: How does longest prefix match work? When a router has multiple routes that match a destination, it selects the one with the longest prefix (most specific). For example, if routes 10.0.0.0/8, 10.1.0.0/16, and 10.1.1.0/24 all exist, and the destination is 10.1.1.5, the router chooses 10.1.1.0/24 because /24 is longer than /16 or /8. This ensures more specific routes take precedence.

Q6: What is a /31 subnet and when is it used? A /31 subnet has only 2 addresses (no usable hosts in traditional sense). RFC 3021 defines /31 for point-to-point links between routers. Both addresses are usable (no network ID or broadcast needed on a link with exactly two endpoints). This saves addresses on WAN links.

Advanced / FAANG-Level

Q7: How would you design CIDR allocation for a new ISP with a /16 block? Design:

1. Total space: /16 = 65,536 addresses
2. Customer allocations:
   • Small customers: /28 (14 hosts) or /27 (30 hosts)
   • Medium customers: /24 (254 hosts)
   • Large customers: /22 (1,022 hosts) or /21 (2,046 hosts)
3. Infrastructure: /24 for backbone links, /30 or /31 for P2P WAN links
4. Growth reserve: Keep 20% unallocated for future
5. Announce: Single /16 to upstream ISPs (aggregation)
6. Deaggregation: May need to announce more-specifics for multi-homed customers
7. Documentation: Maintain IPAM (IP Address Management) database

Q8: Explain the implications of BGP route deaggregation. When an organization announces more-specific routes (e.g., splitting /22 into four /24s):

• Pros: More granular traffic engineering, multi-homing flexibility
• Cons: Larger global routing table (currently 1M+ entries), more memory/CPU in routers, more BGP updates
• Mitigation: RPKI for route validation, BGP communities for policy, prefix filtering by ISPs
• Policy: Many ISPs filter announcements longer than /24 (IPv4) or /48 (IPv6) to prevent table bloat

Q9: How does CIDR interact with BGP in the Internet’s routing architecture? CIDR and BGP work together:

1. ISPs receive large CIDR blocks from RIRs (e.g., /14)
2. ISPs allocate smaller CIDR blocks to customers (e.g., /24)
3. ISPs aggregate customer routes and announce the summary (e.g., /14) to peers
4. Multi-homed customers may announce their own block to multiple ISPs
5. BGP selects best path based on AS path, local preference, MED, etc.
6. Longest prefix match ensures traffic goes to the most specific route

The combination enables the Internet to scale: millions of hosts, but only ~1M routing table entries.

Common Mistakes

1. ❌ Forgetting alignment rules when aggregating — routes must be on power-of-2 boundaries
2. ❌ Confusing CIDR notation with subnet mask — they’re equivalent but written differently
3. ❌ Assuming CIDR eliminates all waste — it reduces waste but doesn’t eliminate it
4. ❌ Not understanding longest prefix match — critical for routing behavior
5. ❌ Mixing up route aggregation with subnetting — aggregation combines, subnetting divides

Summary

• CIDR replaced classful addressing with flexible prefix lengths (/8 to /30)
• Enables efficient allocation — right-sized blocks for actual needs
• Route aggregation combines contiguous routes, reducing routing table size
• Longest prefix match ensures most specific route is chosen
• CIDR is essential for Internet scalability — without it, routing tables would be unmanageable
• Still the foundation of Internet routing today (with BGP)

Cross-References

• IPv4 — IPv4 addressing basics
• Subnetting — Dividing CIDR blocks
• NAT — Using private CIDR blocks
• IP — Internet Protocol overview

Cross References

• Subnetting
• IPv4
• Routing Tables