5-LAYER NETWORK MODEL MADE SIMPLIFIED!

Karthikayan Mailsamy
5 min readJun 23, 2020

In networking ,the most commonly used model is 5-layer model besides OSI model and 4 -layer model with some changes made to 5-layer model.

Credits: GOOGLE LLC

Spoiler Alert:

A basic overview of all the layers and its components are provided here. Don’t expect very detailed explanations over any layer.

5-Layers:

  • Physical Layer
  • Data-Link Layer
  • Network Layer
  • Transport Layer
  • Application Layer

Physical Layer:

This layer comprises of the Cat-6 cables(category 6,other variations are Cat-5 and Cat-5e) used to send or receive the encapsulated Ethernet frame that comprises of IP datagram and TCP segment and the Network ports to which the cables are connected to determine the connections between devices through LED’s(Link LED and Activity LED). This layer is like a carrying path.

Data Link Layer:

This is the layer where connection between nodes(devices) in a network is ensured through switches(similar to hubs(physical layer component,but prone to collision domain)but more reliable than them in aspect of collision domain as it uses Ethernet Protocol to efficiently send data packets i.e Ethernet frames in this layer).Ethernet protocol makes sure the node’s identity (i.e MAC address-Hardware address)to send the data packets. The correct node in a network intended as the destination is found by a protocol called ARP(Address Resolution Protocol) — sending broadcasting messages to all the nodes in the network and the MAC address present in the ARP response from the respected nodes is saved in the ARP cache(which gets restores often to ensure the changes made to the network) as IP-MAC pairs and the nodes lookup its respected ARP cache for future connections. Here CSMA/CD(Carrier Sense Multiple Access with Collision Domain), a part of Ethernet protocol makes the information(yes or no) about data transferring in the network available to all the nodes so that no other nodes commits to transfer when a node is transferring which prevents cross-talk/data miss-up is used.

MAC Address — Media Access Control is a permanent hard coded address of a node.This is known by other nodes in a network through broadcasting whenever needed by sending the IP of the respective nodes.

Network Layer:

This layer uses IP(Internet Protocol) predominantly to find the correct network where the destination node is present through Routers which connects . Networks are split up into subnets by subnetting process and are represented using CIDR ID(Class-less Inter-Domain Routing) to represent the networks.Routers use routing tables to find the destination network and send the data-grams to the appropriate routers that has shorter distance to the destination IP. It also uses couple of protocols out of which Link-Vector Protocol is commonly used nowadays which sends the routing information to all the routers in the ASN(Autonomous System of Networks) to update the shortest distance to all the routers.

Time-To-Live(TTL) — It is a number that is used to prevent the data to be passed infinitely which is in case of the destination IP is not found. It is decremented every time a router hops the data-gram to another one so that when TTL becomes zero ,the data is thrown out. It is 64 in general practice.

IP Address — Internet Protocol which is a 32 bit representation used in Network Layer to find the right node through the network in which the node is located. Each node will be assigned a specific IP which differs depending on the service provider or router used by the network in which the node is present. Subnetting and CIDR are used to increase the number of possibilities of IP addresses and also provides conventional way to find them easily. The most common one is IPv4 (Eg. 172.217.31.255) and now IPv6(Eg. 2001:db8:0:1234:0:567:8:1.) is gaining momentum which is a 128 bit representation.

Routing Table — This table is present in the routers to determine the destination network and the interface that has less passes. It primarily consists of 4 columns,

  • Destination Network
  • Next hop(next possible router)
  • Total hops(total passes) and
  • The interfaces(the connections of the router that connects to the destination IP).

Transport Layer:

This layer uses TCP(Transfer Control Protocol)/UDP(User Datagram Protocol). TCP makes connection to server through system ports and to clients through ephemeral ports . Multiplexing and De-multiplexing processes are made through these ports. Then it uses acknowledgements to make sure the connection is in proper state every time,a communication is made between the nodes(devices) which leads to extra traffic but necessary for important data transfers like phone calls. If the connection between devices is achieved on both the sides, Full-Duplex is achieved or vice versa called Simplex is achieved. But UDP does not makes any acknowledgements and it is suitable for faster transfer with a danger of data loss which does not matter a lot in case of video streaming,radio listening,etc. Acknowledgments are made through activity flags present in the segment fields. The Three-way handshake is used for initiating a connection and Four-way handshake is used to terminate a connection.

TCP Port — TCP Port is used to do multiplexing and de-multiplexing operations. IANA(Internet Assigned Numbers Authority) has limited this to 16 bit value i.e 0–65536 where 1–1023 are system ports assigned to servers, 1024–49151 are registered ports assigned for special operations like database listening,etc and finally 49152–65536 are ephemeral ports assigned to client’s programs used for proper delivery of data to the applications.

Application Layer:

This layer uses various protocols depending on the applications.For example, HTTP(Hyper Text Transfer Protocol) is used by the web-servers and the web-pages.In case of OSI model, this layer is split into Session layer , Presentation Layer and Application Layer.

Checksum Check:
It is used to check whether the data is corrupted or missing in Data-Link, Network and Transport Layers. It is calculated by doing Cyclical Redundancy Checks achieved by polynomial divisions over the entire header contents of all the three layers mentioned above. If the value doesn’t matches with the received value,the data is thrown away.Only the TCP decides when to send the corrupted data again.

Conclusion:

I have made a positive attempt in explaining the contents of a basis network model. This is how a data is transferred to us in a moment.These processes happen in milliseconds in order to keep us intact with the applications like I kept you all through this story😉. If you want to experience it and clear the doubts in the above contents, just copy and paste this address 172.217.31.255 into your web-browser(Chrome,Safari,Firefox or Internet Explorer).Give it a try folks!

See you all again after some time!

Don’t forget to drop your comments below!

Byeeeee!Happy Learning!

--

--