Chapter: 1 Review of Networking Basics

1.1Sub-netting
⮚ An IP address is 32 bits long. One portion of the address indicates a network
(net_id), and other portion indicates the host_id.
⮚ For example imagine an organization with class B address.with this addressing
scheme, the organization is limited to two level of hierarchy.
⮚ The hosts cannot be organized into single groups.
⮚ All the hosts are at the equal level. The organization has one network with many
hosts.
⮚ One solution to this problem is sub netting the further division of the network into
smaller network called sub networks.
⮚ The rest of the Internet is not aware that the network is divided into two physical
sub networks. The two sub network still appear as single network to the rest of
the Internet.
⮚ A packet for host 144.28.16.20 still reaches router. This IP address is still a class
B address where 144.28 defines the net_id and 16.20 define the host id.
⮚ Router knows that network 144.28 is physically divided into two sub network so it
define 144.28 is net id , 16 is subnet id and 20 is host id.
144.28. 16.20
Net id host id
(without subnet)

144.28. 16. 20
Net id subnet id host id
(with subnet)
1.2 Masking

Masking is process that extracts address of the network from an IP address.

⮚ Masking can be done whether we have subnetting or not.
⮚ If we have no subnetted network then masking is use to identified the network
address.
⮚ If we have subnetted then masking is used to define the network address from an
IP address.
⮚ Mask without subnetting
⮚ Router uses a mask even if there is no subnetting.

Table 1.2 Without Subnet

 Class Mask IP Address Network Address

A 255.0.0.0 15.32.56.7 15.0.0.0

B 255.255.0.0 135.67.13.9 135.67.0.0

C 255.255.255.0 201.34.12.72 201.34.12.0

D NA NA NA

E NA NA NA

⮚ Mask with subnetting

⮚ When network is divided into subnet ,the mask address can vary
Table 4.3 With Subnet
Class Mask IP Address Network Address

A 255.255.0.0 15.32.56.7 15.32.0.0

B 255.255.255.0 135.67.13.9 135.67.13.0

C 255.255.255.192 201.34.12.72 201.34.12.64

D NA NA NA

E NA NA NA

Finding the network address

There are two way to find network address from IP address.
⮚ Boundary-level Masking
There are two rules.
1. The bytes in the IP address which is correspond to 255 in the mask address will
remain as it.
2. The bytes in the IP address that correspond to 0 in the mask it will change to 0.

Example
IP: 15. 32.56.7
Mask : 255.255. 0. 0
Network: 15. 32. 0. 0
⮚ Non – Boundary level Masking
 It follow these three rules.

2 The bytes in the IP address which is correspond to 255 in the mask address will
remain as it.
3 The bytes in the IP address that correspond to 0 in the mask it will change to 0.
4 For other bytes use the bit-wise AND operation.

Example
IP : 201. 34. 12. 72
Mask : 255.255.255.192
Network : 201. 34. 12. 64

72 - 01001000
192 - 11000000
01000000

1. 198.25.69.14
IP : 198. 25. 69. 14
Mask : 255.255.255.192
Network : 198. 25. 69. 0

14 - 00001110
192 - 11000000
00000000

2. 206.25.24.24
IP : 206. 25. 24. 24
Mask : 255.255.255.192
Network : 206. 25. 24. 0

24 - 00011000
192 - 11000000
00000000

1.3 Multicasting
 ⮚ Multicast is a technique of group communication in which the sender
simultaneously distributes data to a number of network nodes or recipients.
⮚ One-to-many and many-to-many communication are both forms of multicasting.
⮚ While Multicasting functions similarly to Broadcasting, it sends information to
targeted or particular network participants.
⮚ Similar to when a number of employees all receive business emails addressed to
the same email.

1.4 IGMP (Internet Group Management Protocol)

⮚ IGMP is a network layer protocol that enables several devices to share a single
IP address so they can all access the same data and is used to set up
multicasting on networks using IPv4.
⮚ When they want to join a multicast group, computers and other networked
devices use IGMP. The devices that are a part of each multicast group are
identified by the router that implements IGMP by listening to their IGMP
messages.
⮚ IGMP multicast groups are always subject to change. An IGMP "join group" or
"leave group" message can be sent at any time by a device.
1.4.1 Working:

Devices that support multicast groups and dynamic multicasting can use IGMP. These
gadgets enable the host to join or depart the multicast group membership. Additionally,
these tools enable adding and removing clients from the group. The local multicast
router and the host implement this communication mechanism. When a multicast group
is created, its address, which is relayed as the packet's destination IP address and falls
inside the class D (224-239) IP address range, is created.

For IGMP spying, L2 or Level-2 devices, like as switches, are placed between the host
and multicast router. A controlled method of listening to the IGMP network
communication is known as IGMP snooping. Following receipt of the message from the
host, the switch transmits the membership report to the neighbourhood multicast router.
Local multicast routers use PIM (Protocol Independent Multicast) to further forward the
multicast traffic to distant routers so that clients can get the message/data packets. The
switch intercepts the join message sent by clients desiring to join the network and adds
the ports of the clients to its multicast routing table.
Advantages:

⮚ The IGMP communication protocol effectively transmits the multicast data to the
receivers, resulting in optimal performance because no trash packets are
forwarded to the host.
⮚ Because all of the shared links are connected, bandwidth is completely used.
⮚ Hosts have the option to join and quit multicast groups.
Disadvantages:

⮚ Due to the absence of TCP, network congestion may occur.

⮚ It does not offer good efficiency in filtering and security.
⮚ Some exploits, such as a DOS attack, can exploit IGMP (Denial-Of-Service).
Applications

⮚ Streaming - Multicast routing protocols, such as one-to-many or many-to-many,

are used for audio and video streaming over networks.
⮚ Gamming: Online games and other simulation games with several users over a
network frequently employ the Internet group management protocol.
⮚ Tools for Web Conferencing: Video conferencing is a novel way to meet people
at your convenience, and IGMP connects to users for conferencing and
effectively transmits message/data packets.
Types of IGMP messages

⮚ Membership statistics
⮚ "Leave group" notifications
⮚ Questions about general membership
⮚ Membership inquiries that are specific to groups

1.5 PIM
⮚ PIM (System Independent Multicast) is a multicast routing protocol used to
distribute traffic over a network from a single source to numerous destinations.
⮚ PIM is a grouping of three protocols, including PIM Bi-directional, PIM Dense
Mode, and PIM Sparse Mode.
⮚ PIM is said to be protocol-independent since it does not have a topology
discovery process of its own and instead makes use of routing data from other
routing protocols.
⮚ Shared trees are employed.
⮚ Finding the address of an RP for a multicast group via a shared tree is one of the
key prerequisites for PIM Sparse Mode and Bi-directional PIM.
1.5.1 PIM Sparse Mode

⮚ It is a multicast routing protocol that is built on the premise that subscribers to

any given multicast group will be sporadically dispersed over the network.
⮚ In other words, it is presumptive that the majority of network subnets will reject
any particular multicast packet.
⮚ Routers must actively inform their upstream peers of their interest in specific
groups and sources in order to receive multicast data.
⮚ PIM-SM employs shared trees by default, which are multicast distribution trees
that are rooted at a particular node (in PIM, this router is known as the
Rendezvous Point, or RP), and are utilised by all sources sending to the
multicast.
⮚ Sources must package data in PIM control messages and deliver it by unicast to
the RP in order to send it to the RP.
1.5.2 PIM Dense Mode

⮚ In contrast to PIM-SM, PIM Dense Mode (PIM-DM) is a multicast routing

technology created with the premise that receivers for any multicast group are
dispersed widely across the network.
⮚ Multicast data is first broadcast to all hosts in the network; it is anticipated that
most subnets in the network will request any particular multicast packet.
⮚ When a source first begins to send data, each router on the source's LAN
receives the data and forwards it to all of the neighbouring sources (PIMs) as
well as to any links with directly associated sources.
⮚ Each router that gets a packet that has been forwarded does the same thing, but
only after making sure the packet actually arrived on the upstream interface. The
packet is dropped if not. By using this approach, forwarding loops are avoided.
The network as a whole receives a flood of data in this way.
1.5.3 Bi-directional PIM

⮚ The third PIM protocol, Bi-directional PIM (BIDIR-PIM), is based on PIM-SM. The
primary reason BIDIR-PIM and PIM-SM are different from one another is in how
data is sent from a source to the RP. In contrast to PIM-SM, where data is
transferred using either encapsulation or a source-based tree, BIDIR-PIM uses a
shared tree that is bi-directional, meaning that data can travel in either direction
along any given branch.
BIDIR-PIM's major differences from PIM-SM are as follows
 ⮚ There aren't any source-based trees, and there isn't even a single (S,G) state. As
a result, source-specific multicast is not supported and routers cannot move from
a shared tree to a source-based tree.
⮚ For each RP, one router on each link is chosen as the Designated Forwarder in
order to prevent forwarding loops (DF). The DF election message is used for this
at RP discovery time.
⮚ The idea of a designated router does not exist.
⮚ Encapsulation is not applied.
⮚ There are absolutely no data-driven events in the control plane, and the
forwarding rules are substantially simpler than they were in PIM-SM.

1.6 DVMRP

⮚ Protocol for Distance Vector Multicast Routing

⮚ The source address of the packet is used as the basis for the routing choice in
the distance vector multicast routing protocol.
⮚ This method creates the network's routing tree.
⮚ Depending on the source address of the packet, a router passes each packet it
receives to one or more of its ports.
⮚ Downstream routers construct the remaining portions of the routing tree.
⮚ In this manner, a routing tree is built from source to destination.
The following objectives of the protocol must be met:

1. It must stop loops from forming in the network.

2. It must stop duplicate packets from forming.
3. It must guarantee that a packet takes the quickest route possible from its source
to the router.
4. It ought to support dynamic membership.