- Most enterprise computer networks can be divided into two general types of technologies: LAN and WAN
- LANs connect nearby devices
- WANs connect devices that are far apart
- Two types of LANs: Ethernet LANs and Wireless LANs
- Ethernet LANs use cables (copper).
- Ethernet LANs are also called wired LANs
- Wireless LANs use radio waves for links between nodes.
- Term Ethernet refers to a family of LAN standards that define physical and data link layers of LAN technology.
- LANs need an Ethernet switch
- WLANs do not use Ethernet but AP does to connect to LAN
Ethernet physical layer standards
- Ethernet cabling starts with 802.3
- Ethernet includes many standards for different kinds of optical and copper cabling and different speeds.
- Unshielded Twisted Pair (UTP) saves money compared to optical fibers. (discutable)
- suffix T indicates UTP
- suffix X indicates optic fibers
Transmitting data using twisted pair UTP cable
To understand how Ethernet sends data using electricity, break down the idea in two parts:
- how to create an electrical circuit
- how to make that electrical signal communicate 1s and 0s
- Creates by defining how to use two wires inside a single twisted pair of wires. An electrical circuit requires a complete loop, allowing electricity to flow
- To send data, two devices follow rules called encoding schemes. The idea is the same when two people talk the same language, speaker speaks and the listener understands.
- In twisted pair, wires will be twisted to cancel out most of the electromagnetic interference EMI which interferes with electrical signals in nearby wires including the wires inside the cable, therefore they are twisted to remove that effect.
- EMI in the same cable is called crosstalk
- The term Ethernet link refers to any physical cable between two Ethernet nodes.
- Ethernet link consists of the cable itself, connectors on the ends, and matching ports where connectors are placed.
- 10BASE-T and 100BASE-T require 2 pairs of wires
- 1000BASE-T requires 4 pairs of wires.
- Cable connectors have 8 placeholders for those wires, called pins or pin positions
- Those pinpoints touch ends of wires and create an electrical loop.
- Small form-factor pluggable SFP+ transceiver, runs at 10Gbps.
- UTP Cabling Pinouts for 10Base-T and 100Base-T
- As a rule, Ethernet decided to use pin positions:
- Transmitting: 1 and 2
- Receiving: 3 and 6
- Transmitting 3 and 6
- Receiving: 1 and 2
- NIC and Switch communicate with straight-through cable pinout
When connecting two switches we use crossover cable UTP
Cisco switches have a feature called auto-mdix that notices when the wrong cable is used and automatically changes its logic to make the link work.
Sending data in Ethernet networks
- MAC 6 byte long, 48 bit address.
- 12 digit hexadecimal number
- Called unicast ethernet address. Unicast means that this MAC address represents only one interface
- Error Detection with FCS
- Sender sends complex math formula and stores it together with the result in the frame.
- Receiver solves formula and compares the results.
- If they are not same, Ethernet discards frame
- Ethernet will not try to recover failed frames. Other protocols, TCP for example, will handle error recovery and lost segments.
- HALF DUPLEX, FULL DUPLEX
- Hubs are layer 1 devices
- When electrical signal comes in hub, it will repeat that signal to all other ports except the port from which signal came