Network Devices: Repeaters, Hubs, Bridges, Switches, and Routers

Switches and bridges excel at connecting devices within a local network, but they operate with a fundamental limitation: they understand only MAC addresses, which have no inherent geographical or organizational structure. A MAC address provides no information about where a device is located in the global network topology.

To communicate between different networks—between your home and Google's servers, between branch offices in different cities, or even between VLANs on the same physical switch—we need a device that understands logical addresses with hierarchical structure. This is the role of the router.

Routers operate at the Network Layer (Layer 3) of the OSI model, making forwarding decisions based on IP addresses. While switches ask 'Which port leads to this MAC address?', routers ask 'Which interface leads toward this destination network?' This seemingly simple distinction enables the global Internet and all inter-network communication.

A router is a network device that operates at the Network Layer (Layer 3) of the OSI model. Its primary function is inter-network communication—forwarding packets between different IP networks based on destination IP addresses.

Core Router Characteristics:

Router vs. Switch: Fundamental Comparison:

When Routing is Required:

• Communication between different IP subnets
• Traffic between VLANs (inter-VLAN routing)
• Connection to the Internet
• WAN links between sites
• Any path involving multiple Layer 3 domains

Router design has evolved significantly, from early software-based forwarding to modern distributed hardware architectures capable of terabit-per-second throughput.

Basic Router Architecture:

┌─────────────────────────────────────────────────────────────────────┐
│                         Router Architecture                          │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│  ┌──────────────────────────────────────────────────────────────┐   │
│  │                    Control Plane (CPU)                        │   │
│  │  ┌───────────────┐  ┌───────────────┐  ┌───────────────────┐ │   │
│  │  │ Routing       │  │ Management    │  │ Protocol Stack    │ │   │
│  │  │ Protocols     │  │ Interfaces    │  │ (TCP/IP, ARP...)  │ │   │
│  │  │ OSPF,BGP,RIP  │  │ CLI,SNMP,SSH  │  │                   │ │   │
│  │  └───────────────┘  └───────────────┘  └───────────────────┘ │   │
│  │                                                              │   │
│  │  ┌───────────────────────────────────────────────────────┐   │   │
│  │  │              Routing Information Base (RIB)            │   │   │
│  │  │         (All routes from all protocols)               │   │   │
│  │  └───────────────────────────────────────────────────────┘   │   │
│  └──────────────────────────────────────────────────────────────┘   │
│                              │                                       │
│                              │ Populate FIB                         │
│                              ▼                                       │
│  ┌──────────────────────────────────────────────────────────────┐   │
│  │                    Data Plane (Forwarding)                    │   │
│  │                                                              │   │
│  │  ┌───────────────────────────────────────────────────────┐   │   │
│  │  │           Forwarding Information Base (FIB)            │   │   │
│  │  │         (Optimized for fast lookup - TCAM)            │   │   │
│  │  └───────────────────────────────────────────────────────┘   │   │
│  │                                                              │   │
│  │  ┌─────────┐ ┌─────────┐ ┌─────────┐ ┌─────────┐             │   │
│  │  │Interface│ │Interface│ │Interface│ │Interface│ ...        │   │
│  │  │ GE0/0   │ │ GE0/1   │ │ GE0/2   │ │ GE0/3   │             │   │
│  │  │ [Port]  │ │ [Port]  │ │ [Port]  │ │ [Port]  │             │   │
│  │  └─────────┘ └─────────┘ └─────────┘ └─────────┘             │   │
│  └──────────────────────────────────────────────────────────────┘   │
│                                                                     │
│  [Power Supply]        [Memory/Storage]           [Console Port]    │
└─────────────────────────────────────────────────────────────────────┘

Control Plane vs. Data Plane:

Modern router design separates functions into two planes:

Control Plane:

• Runs routing protocols (OSPF, BGP, IS-IS, RIP)
• Builds and maintains the Routing Information Base (RIB)
• Handles management traffic (SSH, SNMP, console)
• Processes ARP, ICMP, and other control protocols
• Typically implemented in software on general-purpose CPU
• Not in the fast path for transit traffic

Data Plane (Forwarding Plane):

• Performs actual packet forwarding for transit traffic
• Consults Forwarding Information Base (FIB)—optimized subset of RIB
• In high-performance routers, implemented in specialized hardware (ASIC/NPU)
• Operations: IP lookup, TTL decrement, header checksum recalculation, Layer 2 rewrite
• Fast path: microsecond or nanosecond processing per packet

The routing table is the router's map of the network—specifying how to reach destination IP networks. Understanding routing table structure and lookup mechanisms is fundamental to understanding router operation.

Routing Table Entry Components:

Typical Routing Table Entry:
──────────────────────────────

Destination: 192.168.10.0/24      # Network prefix to match
Next-Hop:    10.0.0.2             # IP of next router toward destination
Interface:   GigabitEthernet0/1   # Outgoing interface
Metric:      110                  # Cost/distance (lower = preferred)
Admin Dist:  110                  # Protocol preference (lower = preferred)
Origin:      OSPF                 # How this route was learned
Age:         01:23:45             # Time since route learned

Example Routing Table:

Router# show ip route
Codes: C - connected, S - static, R - RIP, O - OSPF, B - BGP

Gateway of last resort: 10.0.0.1 to network 0.0.0.0

     10.0.0.0/24 is subnetted, 2 subnets
C       10.0.0.0/24 is directly connected, GigabitEthernet0/0
C       10.0.1.0/24 is directly connected, GigabitEthernet0/1
     192.168.0.0/24 is subnetted, 3 subnets
O       192.168.10.0/24 [110/20] via 10.0.0.2, GigabitEthernet0/0
O       192.168.20.0/24 [110/30] via 10.0.0.2, GigabitEthernet0/0  
S       192.168.30.0/24 [1/0] via 10.0.1.5
S*   0.0.0.0/0 [1/0] via 10.0.0.1                    # Default route

The Forwarding Decision Process:

When a packet arrives, the router performs these steps:

Step 1: Receive and Parse

Packet arrives on interface GE0/0
Extract destination IP: 192.168.10.47

Step 2: Longest Prefix Match (LPM)

Search routing table for matching entries:
  192.168.10.0/24  - Matches! (24 bits match)
  192.168.0.0/16   - Matches! (16 bits match)
  0.0.0.0/0        - Matches! (0 bits match - default)

Select longest match: 192.168.10.0/24 (24 bits)

Step 3: Determine Next-Hop and Interface

Route: 192.168.10.0/24 via 10.0.0.2, GE0/0
Next-hop IP: 10.0.0.2
Outgoing interface: GE0/0

Step 4: Layer 2 Resolution

Need MAC address of next-hop (10.0.0.2) for GE0/0
Check ARP cache or send ARP request
Next-hop MAC: AA:BB:CC:DD:EE:FF

Step 5: Header Modification and Forwarding

- Decrement TTL: 64 → 63 (drop if TTL reaches 0)
- Recalculate IP header checksum
- Rewrite L2 header:
    Source MAC: Router's GE0/0 MAC
    Destination MAC: AA:BB:CC:DD:EE:FF
- Forward out interface GE0/0

Routing tables are populated through three primary methods, each appropriate for different scenarios.

1. Connected Networks:

Routes to networks directly attached to router interfaces are automatically added when the interface is configured and activated:

interface GigabitEthernet0/0
  ip address 10.0.0.1 255.255.255.0
  no shutdown

→ Automatic routing table entry:
  C   10.0.0.0/24 is directly connected, GigabitEthernet0/0

2. Static Routes:

Administrator-configured routes for destinations that don't change or for backup paths:

ip route 192.168.30.0 255.255.255.0 10.0.1.5
ip route 0.0.0.0 0.0.0.0 10.0.0.1          # Default route

→ Routing table entries:
  S   192.168.30.0/24 [1/0] via 10.0.1.5
  S*  0.0.0.0/0 [1/0] via 10.0.0.1

3. Dynamic Routing Protocols:

Protocols that automatically discover network topology and exchange routing information between routers:

Administrative Distance:

When multiple sources (static, OSPF, BGP, etc.) provide routes to the same destination, Administrative Distance determines preference:

Lower AD = more trusted. A static route (AD 1) overrides OSPF (AD 110) for the same prefix.

Example:

Both sources provide route to 10.20.0.0/24:
  Static:  10.20.0.0/24 via 192.168.1.1 [AD: 1]
  OSPF:    10.20.0.0/24 via 172.16.1.1 [AD: 110]
  
Router uses static route (AD 1 < AD 110)

One of the router's most critical functions is broadcast domain segmentation—isolating broadcast traffic within individual subnets. This behavior has profound implications for network design and scalability.

Why Broadcasts Don't Cross Routers:

IP broadcasts (destination 255.255.255.255 or directed broadcast like 192.168.1.255) are Layer 3 constructs. When a router receives such traffic:

1. It recognizes the destination as a broadcast
2. By default, it does NOT forward the packet to other interfaces
3. The broadcast is contained within the originating subnet

                    ┌─────────────┐
                    │   Router    │
                    │             │
      Subnet A      │   GE0/0     │      Subnet B
   192.168.1.0/24 ──┤             ├── 192.168.2.0/24
                    │   GE0/1     │
                    └─────────────┘

Host A (192.168.1.10) sends broadcast to 255.255.255.255:

  → Packet received on GE0/0
  → Router recognizes broadcast destination
  → NOT forwarded to GE0/1
  → Broadcast contained in Subnet A only
  
Result: Hosts in Subnet B do NOT receive broadcast

Broadcast Domain Size Implications:

Types of Broadcast Traffic:

Broadcasts are essential for certain protocols but become problematic at scale:

Calculating Broadcast Overhead:

Scenario: 500 devices, ARP every 5 minutes, DHCP every 30 minutes

ARP broadcasts: 500 devices × (1 / 300 sec) = 1.67 broadcasts/sec
DHCP broadcasts: 500 devices × (1 / 1800 sec) = 0.28 broadcasts/sec

Total: ~2 broadcasts/sec × 500 receiving hosts = 1000 packets/sec
processed across the network for broadcast handling alone

With 50 devices per broadcast domain (10 subnets):
50 × 2/sec × 50 receiving = 5000 packets/sec per subnet
But only 50 devices process each broadcast instead of 500

Overall network: 10× reduction in broadcast processing

Most routers connecting private networks to the Internet perform Network Address Translation (NAT)—modifying IP addresses in packet headers to enable communication between private and public address spaces.

Why NAT is Necessary:

The IPv4 address space (approximately 4.3 billion addresses) is insufficient for all connected devices. RFC 1918 defined private address ranges that can be reused:

Private Address Ranges (RFC 1918):
─────────────────────────────────
10.0.0.0/8        (10.0.0.0 - 10.255.255.255)     ~16 million addresses
172.16.0.0/12     (172.16.0.0 - 172.31.255.255)   ~1 million addresses
192.168.0.0/16    (192.168.0.0 - 192.168.255.255) ~65,000 addresses

These addresses are not routable on the public Internet.
NAT translates between private internal addresses and public external addresses.

NAT Operation:

                         ┌─────────────────────┐
                         │      Router         │
    Private Network      │    with NAT         │     Internet
    ─────────────────    │                     │    ──────────────
                         │ Inside    Outside   │
    192.168.1.10   ─────►│ 192.168.1.1        │    
    192.168.1.20         │         203.0.113.5│─────► Web Server
    192.168.1.30         │                    │       93.184.216.34
                         └─────────────────────┘

Outbound packet from 192.168.1.10 to 93.184.216.34:

  Original:     Src: 192.168.1.10    Dst: 93.184.216.34
  After NAT:    Src: 203.0.113.5     Dst: 93.184.216.34
                      ▲
                      └── Router's public IP replaces private IP

Return packet:

  Arriving:     Src: 93.184.216.34   Dst: 203.0.113.5
  After NAT:    Src: 93.184.216.34   Dst: 192.168.1.10
                                          ▲
                                          └── NAT table lookup restores
                                              original destination

Port Address Translation (PAT):

PAT is the most common form, allowing thousands of internal devices to share a single public IP:

NAT Table (PAT Example):
─────────────────────────────────────────────────────────────────────
Inside Local        Inside Global      Outside Global   Outside Local
192.168.1.10:49152  203.0.113.5:1024   93.184.216.34:80  93.184.216.34:80
192.168.1.20:52301  203.0.113.5:1025   93.184.216.34:80  93.184.216.34:80  
192.168.1.10:49153  203.0.113.5:1026   8.8.8.8:53        8.8.8.8:53
─────────────────────────────────────────────────────────────────────

Multiple internal IPs share 203.0.113.5 by using different port numbers.
Router maintains translation table to route return traffic correctly.

The strict separation between bridges/switches (Layer 2) and routers (Layer 3) became problematic as networks grew. The need for devices that could do both led to the development of brouters and, more successfully, multilayer switches.

Brouter (Bridge-Router):

A brouter is a device that can operate as either a bridge or a router depending on the protocol:

Brouter Operation:
─────────────────

Receive frame with:
  - Ethernet header (Layer 2)
  - IP payload (Layer 3, routable)
  → Examine destination IP, route packet between interfaces

Receive frame with:
  - Ethernet header (Layer 2)  
  - NetBIOS payload (Layer 3, non-routable)
  → Bridge based on MAC address (no IP routing possible)

The brouter bridges protocols it cannot route
and routes protocols it understands.

Brouters were significant in the 1980s-90s when multiple network protocols coexisted (IP, IPX, AppleTalk, NetBIOS). As IP became dominant, the brouter concept faded.

Multilayer Switch (Layer 3 Switch):

The multilayer switch represents the modern fusion of switching and routing:

┌─────────────────────────────────────────────────────────────────────┐
│                     Multi-Layer Switch                               │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│   ┌───────────────────────────────────────────────────────────┐     │
│   │               Layer 3 Routing Engine                       │     │
│   │   Routing protocols (OSPF, BGP), Routing table (RIB)      │     │
│   │   Access Control Lists, Policy-Based Routing              │     │
│   └─────────────────────────────┬─────────────────────────────┘     │
│                                 │                                   │
│                                 ▼                                   │
│   ┌───────────────────────────────────────────────────────────┐     │
│   │      Unified Hardware Forwarding Engine (ASIC)            │     │
│   │                                                           │     │
│   │   ┌─────────────────┐     ┌─────────────────┐            │     │
│   │   │  MAC Table      │     │     FIB         │            │     │
│   │   │  (L2 Forwarding)│     │  (L3 Forwarding)│            │     │
│   │   └─────────────────┘     └─────────────────┘            │     │
│   │                                                           │     │
│   │   Decision at wire speed:                                 │     │
│   │   - L2: Same VLAN → switch based on MAC                  │     │
│   │   - L3: Different VLAN → route based on IP               │     │
│   └───────────────────────────────────────────────────────────┘     │
│                                                                     │
│   ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐            │
│   │Port 1│ │Port 2│ │Port 3│ │Port 4│ │Port 5│ │Port N│            │
│   │VLAN10│ │VLAN10│ │VLAN20│ │VLAN20│ │VLAN30│ │Trunk │            │
│   └──────┘ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘            │
└─────────────────────────────────────────────────────────────────────┘

Router vs. Multilayer Switch:

When to Use Each:

• Multilayer Switch: Inter-VLAN routing, campus/data center core, most enterprise LAN scenarios
• Traditional Router: WAN connectivity, Internet edge, MPLS, complex policy routing, locations requiring advanced NAT/firewall

Having explored all the major network devices across Layers 1-3, let's consolidate our understanding with a comprehensive comparison.

Device Hierarchy by OSI Layer:

                                OSI Layer Operation
                                
    Device          │ Layer 1  │ Layer 2  │ Layer 3
    ────────────────┼──────────┼──────────┼──────────
    Repeater        │    ●     │          │
    Hub             │    ●     │          │
    Bridge          │    ●     │    ●     │
    Switch (L2)     │    ●     │    ●     │
    Switch (L3)     │    ●     │    ●     │    ●
    Router          │    ●     │    ●     │    ●
    
    ● = Operates at this layer

*Switches extend broadcast domains unless VLANs are configured; L3 switches segment broadcast domains through inter-VLAN routing.

Decision Framework for Device Selection:

┌─────────────────────────────────────────────────────────────────┐
│                Device Selection Decision Tree                    │
└─────────────────────────────────────────────────────────────────┘

Need to extend physical distance?
├── Yes → Consider fiber, media converter, or wireless
│         (Standalone repeaters rarely used)
│
└── No → Continue

Need to connect multiple devices locally?
├── Yes → Use a Switch (L2)
│         Calculate ports needed, PoE requirements
│         
└── No → Continue  

Need to connect different IP subnets/VLANs?
├── Yes → Use L3 Switch or Router
│         ├── Campus/Data Center: L3 Switch
│         ├── WAN/Internet Edge: Router
│         └── Complex routing (BGP, MPLS): Router
│
└── No → Continue

Need Internet/WAN connectivity?
├── Yes → Router (with NAT, firewall features)
│
└── No → Reassess requirements

We have completed our comprehensive exploration of network devices from Layer 1 through Layer 3. Let's consolidate the key concepts for routers and the overall device hierarchy:

Module Complete:

You have now mastered the fundamental network devices that form the building blocks of all computer networks. From the simple signal regeneration of repeaters through MAC-based switching to IP-based routing, you understand how devices at each OSI layer contribute to network connectivity.

This foundation prepares you for the subsequent modules in this chapter, where we'll explore:

• Collision and broadcast domains in greater detail
• Learning bridges and MAC table management
• Spanning Tree Protocol for loop prevention
• VLANs for logical network segmentation
• Switch vs. Router detailed comparison and selection criteria