Switched Networks and VLANs
A plain switch keeps every port in one giant broadcast domain — efficient for four desks, unworkable for four hundred. A VLAN splits that one domain into many virtual LANs at Layer 2, and almost every switch question on the CCNA turns on this one idea: the switch segments collisions, but only a VLAN (or a router) segments broadcasts.
- Explain what a VLAN is and the Layer 2 broadcast domain it creates.
- Name the reasons an organisation uses VLANs: security, performance, and flexibility.
- Describe the normal range (1–1005) and extended range (1006–4094) of VLAN numbers.
- Identify the default VLAN 1 and the reserved legacy VLANs 1002–1005.
- Distinguish a data VLAN from a voice VLAN.
- Predict broadcast behaviour when two VLANs exist on one switch.
The flat broadcast domain
You saw in Lesson 11 that a switch increases the number of collision domains (it gives every port its own) but leaves the broadcast domain untouched. Legally, all ports of an unconfigured switch belong to the same Layer 2 group: every broadcast frame that arrives on any port is flooded out of every other port, and every host hears it. In a small office with half a dozen machines this is harmless. Add hundreds of hosts, and it is a problem.
Each broadcast frame interrupts every CPU on the segment for the instant it takes the NIC to discover the frame is not for it. Layer 2 discovery protocols, ARP requests, DHCP DISCOVERs, and service announcements all generate broadcasts constantly. The result is a slow, noisy, insecure segment where any host can sniff every neighbour’s conversations and one misconfigured machine can disturb the whole floor. This single undivided segment is what a VLAN is designed to cut into pieces.
What a VLAN is
When you create VLAN 10 on a switch, you are telling the switch: ports in VLAN 10 form one broadcast domain, and ports not in VLAN 10 are invisible to them. The switch enforces this in its forwarding logic. A frame learned from a VLAN 10 port may be forwarded only to other VLAN 10 ports. VLAN 20 frames are handled by the same hardware but treated as a separate universe — a separate broadcast domain, a separate MAC address table region, a separate Layer 2 network.
This is the phrase to remember: a VLAN is a broadcast domain. One VLAN — one broadcast domain. If a question says two PCs are in the same VLAN, they can reach each other at Layer 2; if they are in different VLANs, they cannot — something at Layer 3 must join them (Lesson 15).
Why VLANs exist: security, performance, flexibility
The exam will ask you to justify a VLAN design. The three accepted reasons are security, performance, and flexibility (or organisational grouping). Learn the one-line argument for each.
- Security — By isolating departments on separate VLANs, you ensure that traffic from one group never reaches the ports of another at Layer 2. Human resources, finance, and guests cannot sniff each other’s frames on the shared segment, and any inter-VLAN access must pass through a router or firewall where you can apply policies.
- Performance — Broadcasts, multicast, and unknown unicast frames are flooded only within the VLAN, so a chatty group in VLAN 20 no longer interrupts VLAN 10. Smaller broadcast domains mean less noise, fewer interrupts, and more predictable throughput for every host.
- Flexibility — Users no longer have to sit next to their server or their teammates. Because a VLAN can span many switches (via trunks, Lesson 14), you can regroup people logically without rewiring. Moving a user to another department is a configuration change, not a cable pull.
VLAN ranges and reserved numbers
Cisco switches divide the VLAN ID space in two, and the boundary is a favourite fact:
| Range | IDs | How they can be used |
|---|---|---|
| Normal range | 1–1005 | Configurable on any platform; stored in the VLAN database (vlan.dat for IOS); advertised by VTP. |
| Extended range | 1006–4094 | Available on most modern switches but not stored in vlan.dat; must be configured globally, not per-port, and VTP will not carry them (switch must be in VTP transparent mode). |
| Reserved legacy | 1002–1005 | Token Ring and FDDI VLANs; present by default and cannot be deleted. |
| Reserved system | 0 and 4095 | Never usable for user traffic; reserved for 802.1Q system use. |
Note the trap: 1002–1005 sit inside the normal range but are reserved for the legacy Token Ring and FDDI translations, so in practice the VLANs you will ever create sit in 2–1001 (normal) or 1006–4094 (extended). The extended range tops out at 4094, not 4096 — the 12-bit VLAN ID field gives 4096 values, but 0 and 4095 are reserved, leaving 4094 usable IDs.
Default and native VLAN 1
Out of the box, every port on a Cisco switch is an access port in VLAN 1, and VLAN 1 is the default VLAN. It cannot be renamed or deleted. VLAN 1 is also the native VLAN by default — the VLAN whose frames travel across a trunk without a tag (Lesson 14).
Because VLAN 1 is where everything lands until you say otherwise, it is a security concern: any untagged traffic that arrives on a trunk enters VLAN 1. Best practice in production is to move management traffic out of VLAN 1 and to make the native VLAN something other than VLAN 1. The exam tests the facts: VLAN 1 is default, is native by default, and cannot be deleted.
Data VLAN vs voice VLAN
IP phones sit between the switch and the PC and carry two kinds of traffic: the phone’s itself (signalling and RTP audio) and the PC’s data that passes through the phone’s built-in switch. Mixing both on one VLAN would let voice traffic compete with noisy PC traffic, so Cisco design separates them:
- Data VLAN — the VLAN the PC belongs to, configured as the port’s access VLAN.
- Voice VLAN — a separate VLAN for the phone. Usually the phone negotiates it using CDP (Lesson 16) and applies QoS marking (DSCP EF 46 for audio).
Conceptual exam questions ask: “Where should IP phones go?” Answer: a dedicated
voice VLAN, so that phones are grouped, QoS can be applied cleanly, and a PC’s broadcast storm
cannot choke call control traffic. Configuration — the switchport voice vlan
command — is Lesson 13.
The cost: inter-VLAN traffic needs a router
Segmentation is not free. Two devices in VLAN 10 and VLAN 20 are Layer 2 strangers: their frames never cross the VLAN boundary, and broadcasts do not either. For them to exchange data, the traffic must be routed — which means it must be sent to a Layer 3 device (a router, or a Layer 3 switch) that holds an interface in each VLAN. This is inter-VLAN routing, and it is the whole of Lesson 15. For now, hold the one-line cost: VLANs demand a router for cross-VLAN communication.
Put the two lessons together. A switch partitions broadcast domains by VLAN; a VLAN is a broadcast domain; therefore moving a host into a new VLAN both protects and isolates it. Every design answer that involves VLANs is really an answer about where the Layer 3 boundary will be.
The vocabulary you need
Every term below appears in the lesson text above. Select one to read its definition, then use the quiz at the end to check what stuck.
Review
Check what you have learned with these review questions.
Sources & references
- Cisco. CCNA (200-301) Exam Topics. cisco.com.
- IETF. Request for Comments (RFCs). rfc-editor.org.
- IANA. Internet Assigned Numbers Authority registries. iana.org.
Target exams: CCNA 200-301 v1.1 · CyberOps Associate 200-201. Last content verification: 2026-10-11.