Encapsulation in Networking

• Definition: The process of wrapping data with the necessary protocol information before
transmission.

• Analogy: Like putting a letter in an envelope before mailing it.

• De-encapsulation: The process of unwrapping the message at the receiving end (taking
the letter out of the envelope).

Ethernet Protocol & Frame Structure

• Purpose: Defines how data is packaged for transmission over Ethernet networks.

• Key Components of an Ethernet Frame:

o Preamble: Helps synchronize communication by providing timing information.

o Start Frame Delimiter (SFD): Marks the start of the actual frame data.

o Destination MAC Address: The intended recipient.

o Source MAC Address: The sender of the frame.

o Length/Type Field: Indicates either the length of the payload or the type of
protocol used (like IPv4, IPv6).

o Payload: The actual data being transported.

o Frame Check Sequence (FCS): Used for error detection (checks if data was
corrupted during transmission).

Access Layer Overview

• Purpose: Connects end-user devices (hosts) to the wired Ethernet network.

• Devices Used: Initially Ethernet hubs, now mostly switches.

• Connection: Each host connects to an access layer device using an Ethernet cable.

Ethernet Hubs (Obsolete)

• Function: Simple devices with multiple ports for connecting hosts.

• Limitation: Only one message can be transmitted at a time.

 • Collisions: Simultaneous messages cause collisions, leading to retransmissions and
network slowdowns.

Ethernet Switches

• Layer: Operate at Layer 2 (Data Link layer) of the OSI model.

• Function:

o Reads MAC addresses from Ethernet frames.

o Uses a MAC address table to map devices to ports.

o Establishes temporary circuits between sending and receiving devices.

• Learning: Dynamically learns MAC addresses by reading the source address of incoming
frames.

• Performance:

o Supports full-duplex communication (send/receive simultaneously).

o Eliminates collisions, improving speed and efficiency.

8.1.1 – The IPv4 Address

What is an IPv4 Address?

• A logical network address used to identify a host on a LAN or the internet.

• It must be:

o Unique on the local network for local communication.

o Globally unique for internet communication.

Where is it Assigned?

• Assigned to a network interface, typically a NIC (Network Interface Card).

• Devices like:

o Workstations

o Servers (can have multiple NICs/IPs)

o Network printers

o IP phones
 • Routers also have IPv4 addresses for each of their interfaces.

IPv4 in Packet Delivery

• Every packet on the internet includes:

o A source IPv4 address (where it came from)

o A destination IPv4 address (where it’s going)

• This ensures:

o Data reaches the right destination.

o Replies get back to the correct source.

Quick Concept Check:

1. What makes an IPv4 address valid?

→ It must be properly configured and unique.

2. Can a device have more than one IPv4 address?

→ Yes, if it has multiple network interfaces.

3. Why do routers need IPv4 addresses?

→ To connect to networks and route packets between them.

IPv4 Address Structure

• 32 bits in length: An IPv4 address consists of 32 binary digits.

o Example in binary: 11010001101001011100100000000001

• Grouped into octets: The 32 bits are split into 4 octets, with each octet being 8 bits long.

o Example: 11010001.10100101.11001000.00000001

• Converted to decimal: To make it more human-readable, each octet is converted to its

decimal value.

o Example in dotted-decimal: [Link]

Why Dotted-Decimal Notation?

• Binary: 11010001101001011100100000000001 (hard to read)

• Octets: 11010001.10100101.11001000.00000001 (still long, but structured)

• Dotted-Decimal: [Link] (easy to read, easy to configure)

 What You Need to Know Right Now:

• IPv4 addresses are 32 bits long, grouped into 4 octets (8 bits each).

• Dotted-decimal notation is the standard format used to represent these addresses.

8.2.2 – Networks and Hosts

IPv4 Address Hierarchy

• An IPv4 address is split into two parts:

1. Network portion: Identifies the network the host is on.

2. Host portion: Identifies the specific host (device) within that network.

• Subnet Mask: Defines the boundary between the network and host portions.

o Example: [Link] means the first 3 octets represent the network, and the
last octet represents the host.

Example Breakdown:

• IPv4 Address: [Link]

• Subnet Mask: [Link]

o Network portion: 192.168.5 (first 3 octets)

o Host portion: 11 (last octet)

So, the network address is [Link] (network), and the host on this network has the address
11.

Hierarchical Addressing in Action

• Why this hierarchy matters:

o Routers only need to know how to reach the network—not each individual host.

o This makes routing more efficient because the router handles data based on the
network portion.

Multiple Logical Networks on One Physical Network

• Multiple logical networks can share the same physical network if they have different
network portions.

o Example: Two different logical networks:

 ▪ 192.168.18.x (one network)

▪ 192.168.5.x (another network)

• Communication between networks:

o Hosts on the same network portion (e.g., 192.168.5.x) can communicate directly.

o Different network portions require routing to communicate.

Analogy: Telephone System

• Telephone Hierarchy:

o The country code, area code, and exchange = network portion.

o The local number = host portion.

Just like with IP addresses, a call can be routed based on the network (country/area/exchange
codes), and the final number identifies the specific person (host) within that network.

Key Takeaways:

• IPv4 Address: A 32-bit address split into network and host portions.

• Subnet Mask: Defines how much of the address is the network portion.

• Routing: Routers handle communication between different networks, not individual

hosts.

Unicast Transmission

• Definition: A one-to-one communication between a source and a destination.

• Key Characteristics:

o A unicast packet has a destination IP address that is specifically for one device.

o The source IP is also a unicast address, meaning it’s coming from one specific
source.

o Address Range: IPv4 unicast addresses fall within [Link] to [Link].

• Example: A device sending a message to a specific server.

Broadcast Transmission

• Definition: A one-to-all communication, where a message is sent to all devices on a

network.

• Key Characteristics:

o The destination IP address in a broadcast packet has all ones (1s) in the host
portion (i.e., [Link]).

o Broadcasts are processed by every device in the same broadcast domain.

• Types of Broadcasts:

1. Directed Broadcast: Sent to all hosts on a specific network. For example, [Link]
might broadcast to all devices in the 192.168.1.x network.

2. Limited Broadcast: Sent to [Link] and targets all devices within a local
network or subnet. Routers typically do not forward broadcast packets to other networks.

• Example: A device sending a message to all devices on a local network (e.g., ARP
requests).

Multicast Transmission

• Definition: A one-to-many or many-to-many communication, where a message is sent

to a group of devices (multicast group).

• Key Characteristics:

o The destination IP address is a multicast address within the range [Link] to

[Link].

o Multicast groups allow devices to subscribe to a particular group and only

process the packets sent to that group’s multicast address.

o This reduces network traffic because a single message is sent to all group
members without being broadcast to everyone on the network.

• Example: Streaming video to multiple devices (e.g., a live broadcast to several

subscribers).
Summary Comparison:

Transmission
Communication Type Destination Address Example
Type

Sending a message to a
Unicast One-to-One Specific IP (e.g., [Link])
specific server

All ones in the host portion ARP requests or local

Broadcast One-to-All
(e.g., [Link]) network-wide messages

Video streaming to
One-to-Many / Many- Reserved range ([Link] -
Multicast multiple devices in a
to-Many [Link])
group

Key Points to Remember:

1. Unicast is a direct message between two devices.

2. Broadcast sends messages to all devices on a network.

3. Multicast sends messages to a group of devices, reducing unnecessary network traffic.

Types of IPv4 Addresses

1. Public IPv4 Addresses

• Definition: Globally unique IP addresses that are routable on the internet.

• Routing: Public IPv4 addresses are routed between ISP routers and must be unique to
ensure they can be reached globally.

• Example: [Link] (Google's public DNS server).

2. Private IPv4 Addresses

• Definition: IP addresses used internally within an organization (intranet) and not

routable on the internet.

• Purpose: Often assigned to internal hosts, such as computers, printers, and routers,
within a private network.

Private IP Address Ranges:

o Class A: [Link] - [Link]

 o Class B: [Link] - [Link]

o Class C: [Link] - [Link]

• NAT (Network Address Translation):

o Before a private address can communicate on the internet, NAT is used to

translate it into a public IP address.

o Example: When an internal device with a private IP ([Link]) accesses a

website, NAT translates the private IP into a public one, like [Link], so the
traffic can be routed.

3. Loopback Addresses

• Definition: Special addresses that allow a device to send traffic to itself.

• Range: [Link]/8 (typically [Link] is used).

• Purpose: Used for testing and troubleshooting.

o Example: ping [Link] checks if the TCP/IP stack is working on the device.

4. Link-Local Addresses

• Definition: Automatically assigned addresses used when a device cannot obtain an

address from a DHCP server.

• Range: [Link]/16.

• Common Use: Known as APIPA (Automatic Private IP Addressing) in Windows, these

addresses allow a device to communicate with others on the same local network.

o Example: When a Windows device cannot reach a DHCP server, it assigns itself
an IP like [Link].

Classful Addressing (Historical)

In 1981, IPv4 addressing was based on classful addressing. The classes were designed to
support different types and sizes of networks.

IPv4 Classes:
 1. Class A ([Link] to [Link]):

o Designed for: Extremely large networks.

o Hosts per Network: Over 16 million.

o Default Subnet Mask: [Link].

o Example: [Link] (Private address in Class A).

2. Class B ([Link] to [Link]):

o Designed for: Medium to large networks.

o Hosts per Network: Up to 65,000.

o Default Subnet Mask: [Link].

o Example: [Link] (Private address in Class B).

3. Class C ([Link] to [Link]):

o Designed for: Small networks.

o Hosts per Network: Up to 254.

o Default Subnet Mask: [Link].

o Example: [Link] (Private address in Class C).

4. Class D ([Link] to [Link]):

o Multicast Addressing: Used for multicast transmissions (one-to-many).

o Example: [Link] (used for network-wide multicast).

5. Class E ([Link] to [Link]):

o Experimental Addressing: Reserved for research and future use, not used for
public or private addressing.

IPv4 Address Allocation

• IANA (Internet Assigned Numbers Authority): Manages global allocation of IP address

blocks.

• RIRs (Regional Internet Registries): Allocate IP address blocks from IANA to ISPs and
organizations.

• ISPs: Assign addresses to end-users or smaller ISPs.

Key Takeaways:

• Public IPs: Globally routable on the internet.

• Private IPs: Used for internal networks; require NAT to communicate on the internet.

• Loopback IP: [Link] allows a device to communicate with itself.

• Link-Local IP: Used when there’s no DHCP; 169.254.x.x (APIPA).

• Classful Addressing: Historically, addresses were divided into Classes A, B, C, D, and E.

Broadcasts and ARP in Ethernet LANs

ARP (Address Resolution Protocol)

• Function: ARP helps devices on an Ethernet LAN map IPv4 addresses to MAC addresses
(Layer 2 addresses).

• How it Works:

1. When a device wants to send data to another device on the same network, it
first needs to know the destination's MAC address.

2. If it doesn't already know the MAC address, it sends an ARP broadcast to all
devices in the local network.

3. The device with the matching IPv4 address responds with its MAC address.

4. This process allows the source device to send data to the correct MAC address.

Broadcasts in Ethernet

• Definition: A broadcast is a message sent to all devices on the network.

• Propagation: Ethernet switches forward broadcasts out all interfaces except the one on
which they were received. This means every device connected to the network (within
the same broadcast domain) will process the broadcast.

• Examples:

o ARP requests to resolve IP addresses.

o DHCP discovery messages to locate a DHCP server.

Challenges with Large Broadcast Domains

• A broadcast domain is a network segment where broadcasts can reach all devices.
 • Problem: When there are many hosts in a large broadcast domain, the volume of
broadcasts can become excessive, leading to:

o Network congestion.

o Performance degradation.

o Security concerns from potentially malicious broadcast traffic.

Solution: Subnetting

What is Subnetting?

• Subnetting involves dividing a large network (broadcast domain) into smaller subnets.

• Each subnet has a smaller range of IP addresses and can be isolated from others to
reduce unnecessary broadcast traffic.

• Subnetting involves borrowing bits from the host portion of the IP address to create
additional network bits.

Benefits of Subnetting:

1. Reduced Broadcast Traffic: By dividing a large network into smaller subnets, broadcast
traffic is confined to that subnet, reducing network congestion and improving
performance.

2. Better Network Performance: Smaller broadcast domains mean fewer devices are
affected by broadcasts and other network traffic.

3. Improved Security: Administrators can apply more granular security policies for
different subnets. For example, one subnet might be allowed to access a specific server,
while another subnet is restricted.

4. Network Management: With smaller subnets, it's easier to manage IP addresses,

devices, and services.

Subnetting Example:

Imagine a network with the address [Link]/24 (Class C network):

• The default subnet mask is [Link], which allows for 254 host addresses.

• If you split this network into two smaller subnets using [Link] (subnet mask),
you get:

o Subnet 1: [Link]/25 (Addresses from [Link] to [Link]).

 o Subnet 2: [Link]/25 (Addresses from [Link] to [Link]).

Now, broadcast traffic from [Link] will only affect devices in Subnet 1, not those in
Subnet 2.

Conclusion:

• Broadcasts and ARP are crucial for device discovery in Ethernet LANs.

• Large broadcast domains can create network performance issues, which can be
mitigated by subnetting.

• Subnetting reduces broadcast traffic, improves security, and provides better network
management by creating smaller, more manageable subnets.