In this section we will:
- Learn the OSI and TCP/IP models
- Learn about the different networking devices
- Learn physical interfaces, cabling types, and various common network topologies
Introduction to Computer Networking
Just a side note before we get into it, this will be a longer study guide as it encompasses the basics of computer networking. It does assume some knowledge of the reader, i.e. difference between LAN and WAN, and knowing the ethernet types bandwidth and cabling type. Stick with it and feel free to reach out if I’ve missed anything or you have any questions.
In the beginning, there was little in the way of computer communications. There were lone computers doing what lone computers did, unaware of any other computers that might exist around them. No way to connect, no way to communicate. Eventually, we were able to connect a crossover cable between two computers, and suddenly, they could talk! From there, we connected a large cable between multiple computers, and now several computers could share information at a time. The progression in computer networking was slow but ever moving. As more devices were added, Routers, Hubs, Bridges, Switches, APs, etc., networks got larger and larger.
Today, there are two Networking Models that are used in the industry: TCP/IP (Transmission Control Protocol/Internet Protocol) model, and the OSI (Open Systems Interconnection) model. These models are used to help understand how communication occurs across a network. TCP/IP and OSI are called “Layered Models,” and are not vendor specific, meaning they can be, and are used, industry wide.
We’ll focus on the OSI model first. The OSI model primarily serves as a tool for explaining networking concepts and troubleshooting. It is the larger of the 2 models, with a total of 7 layers, listed from top to bottom:
- Application – Interfaces between the network and the application software. Examples include authentication services.
- Presentation – Defines the format and organization of data.
- Session – Establishes and maintains the end-to-end connection between endpoints.
- Transport – Connection establishment and termination, flow control, error recovery, and segmentation of large data into smaller blocks for transmission
- Network – Addressing, routing, and path determination
- Data Link – Formats data into frames for transmission over physical media (wired connections)
- Physical – Cabling, connections, and details for transmitting bits.
Best way to remember the 7 layers is through the mnemonic phrase from top to bottom:
[A]ll [P]eople [S]leeping [T]hrough [N]etworking [D]on’t [P]ass
Application data is passed down the model to be transmitted across the network or networks. The following steps outline how the OSI model transports data:
- Data is created at the application layer of the source host device
- Data gets passed down the stack in the device and is segmented and encapsulated
- Data is generated onto the media at the network layer
- Data is transported through the internetwork
- The destination host received the data at the network layer
- Data gets passed up the stack at the destination and is decapsulated and reassembled
- Data gets passed up to the application layer of destination device
The TCP/IP model focuses only on 5 of the OSI’s 7 layers. TCP/IP model is what is primarily used in todays networks. The layers from top to bottom:
5(7) – Application – Represents data to the user. Protocols used include DNS, DHCP, HTTP, FTP, etc.
4 – Transport – Supports communication between devices – TCP, UDP
3 – Network – Determines best path for packet forwarding – IP addressing
2 – Data Link – Contains the Address Resolution Protocol info – ARP
1 – Physical – The physical connections between devices – Ethernet, Wireless
There is also a simpler 4 layer model of the TCP/IP:
4 – Application – Represent data to the user
3 – Transport – Supports communication between devices
2 – Internet – Determines optimal path through network
1 – Network Access – Hardware devices that make up the network (Routers, Switches, etc.)
Looking at the TCP/IP model in the scope of requesting a web page, TCP model works in the following steps:
- Web client send HTTP request for a specific web server down to the transport layer
- TCP encapsulates the request with a TCP header which includes the destination port number for HTTP (port 80, port 443 for HTTPS)
- Lower layers process and send request to the web server
- Web server receives the HTTP requests and send a TCP ACK back to the to requesting web client
- Web server sends the HTTP response down to the transport layer (from the application layer)
- TCP encapsulates the HTTP data with a TCP header
- Lower layers process and send the response to the requesting web client
- The requesting web client sends an ACK back to the web server
This is a simplified version of how TCP handles web requests. These specific steps do not touch on the encapsulation process. Each layer of the model will add its own header information. A tailer is added at the Network layer. Below are the five steps of data encapsulation:
- Starts at the application layer. Data is encapsulated with a layer 7 header
- Layer 7 data gets encapsulated inside a transport layer header, which for user applications is typically either a TCP or UDP header.
- Transport layer data get encapsulated in an IP header. IP is the only available protocol in the TCP/IP network model at the internet layer.
- Internet layer data gets encapsulated in a network access layer header and trailer. Only the Network access layers makes use of trailers.
- The frame gets transmitted by the physical layer.
Common Networking Devices
For most home networks, your ISP (internet service provider) will provide you with an all-in-one modem, router, switch, AP. For home use, this is perfectly fine in most cases. However, once you determine the need to scale your network and routing capabilities, and security needs, this all-in-one method becomes less applicable and more cumbersome. Small Office/Home Office (SOHO) networks, for instance, may make use of an added switch to have more wired devices. Below is a breakdown of common devices and their definitions, as well as whether they are legacy or still currently used in most set ups:
- Router – These typically connect your LAN to the Internet. A router will route the data from your network to another network, and the other way around. The Internet is basically comprised of a bunch of routers. Routers are Layer 3 (network layer) devices
- Switch – Switches most commonly act as your access layer device in hierarchical network topologies. End user devices would connect directly to the switch when wireless isn’t possible or wanted. Switches are Layer 2 (data link Layer) devices. There are L3 switches, with routing capabilities, but we will touch on those later.
- Hub – These are mostly legacy devices, and were replaced by switches. If a switch is a ‘smart device’ that can think about it’s actions, a hub is a dumb device that only forwards traffic.
- Bridge – Bridges were replaced by hubs, which then were replaced by switches. Almost never used in todays networks. They again simply forwarded traffic, with no ability to make decisions as whether the packets should be forwarded to that specific host.
- Firewalls – These can be hardware or software based, Firewalls control access to the network. Acting essentially as a gate keeper, firewalls will protect data and resources within a network from security threats. There are two different variations on the firewall that should be knowm:
- Stateless Firewall – Do not inspect traffic, does not examine the entire packet, instead makes decisions based on predetermined security rules. These are however cost-effective, fast, and suitable for heavy traffic
- Stateful Firewall – Can be easier to compromise if not updated properly. Better at detecting unauthorized traffic or forged messaging. Retains key attributes of network connections. Extensive logging capabilities.
- IDS and IPS – Intrusion Detection System and Intrusion Prevention System, respectively. Both of these devices recognize network attacks, however, the detection system will monitor and alert of an attack, the prevention system will actively prevent attacks.
- AP/WAP/WLC – Access Points (AP)/Wireless Access Points (WAP. No, not that WAP) provide wireless capabilities to otherwise wired networks. Access points are often controlled by Wireless LAN Controllers (WLC), which allow configuration of the AP. Most WLCs can monitor, maintain, and configure multiple APs in a network. APs and WLC are touched on in the study guide for Wireless Concepts (link here)
The CCNA does expect you to know which cables connect which types of devices in a network topology. Below is a breakdown:
UTP (Unshielded twisted pair) Straight-Through Cable:
- Switch to Router (ethernet port)
- Computer to Switch
- Computer to Hub
UTP Crossover Cable:
- Switch to Switch
- Switch to Hub
- Hub to Hub
- Router to Router (ethernet port)
- Computer to Computer
- Computer to Router
You may be asked a multiple choice, multiple selection question showing you a network diagram and you will need to be able to know whether a cable connecting a switch to a router is crossover or straight-through, or computer to switch is straight-through or crossover. Your best bet is to remember the three straight through connections, and know everything else is a crossover connection.
Common Network Topologies
There are various common network topologies that the CCNA will ask you to identify, usually as a fill-in-the-blank, drag and drop type question. Below are a few examples and simple diagrams:
Extended Star –
Partial Mesh –
Full Mesh –
Additionally, there are what are called hierarchical campus network topologies and designs. These are, as suggested in the name, not flat but hierarchical, and consist of L2 and L3 switches (remember L3 switches have routing capabilities). Typically these are 3 -tier designs with a Core Layer acting as the redundant backbone, Distribution Layer controlling the data between the access and the core layers, and the Access Layer where local and remote users connect to the network. Below is an example of a campus network design: