Ethernet Architecture Facts

The following table shows specifics of the Ethernet architecture.

Specification

Description

Topology

The physical topology is the mapping of the nodes of a network and the physical connections between them, such as the layout of wiring, cables, the locations of nodes, and the interconnections between the nodes and the cabling or wiring system. The logical topology is the way messages are sent through the network connections. Ethernet supports the following topologies:

Physical bus, logical bus
Physical star, logical bus
Physical star, logical star

Media access

Ethernet uses Carrier Sense, Multiple Access/Collision Detection (CSMA/CD) to control access to the transmission medium. Devices use the following process to send data:

  1. Because all devices have equal access to the transmission media (multiple access), a device with data to send first listens to the transmission medium to determine if it is free (carrier sense).
  2. If it is not free, the device waits a random time and listens again to the transmission medium. When it is free, the device transmits its message.
  3. If two devices transmit at the same time, a collision occurs. The sending devices detect the collision (collision detection) and send a jam signal.
  4. Both devices wait a random length of time before attempting to resend the original message (called a backoff).

Transmission media

Ethernet supports the following cable types:

  • Unshielded twisted-pair cables (UTP) with RJ-45 connectors. This is the most common transmission medium used for Ethernet. Each cable consists of eight wires, twisted into four pairs. UTP cables are classified by categories:
    • Cat3, rated up to 10 Mbps
    • Cat4, rated up to 16 Mbps
    • Cat5, rated up to 100 Mbps
    • Cat5e, rated up to 1,000 Mbps (gigabit)
  • Fiber optic, most commonly used in high-speed applications such as servers or streaming media. Fiber optic cables have ST, SC, LC, and MT-RJ connectors.
  • Coaxial for older Ethernet implementations (often called thinnet or thicknet networks). Coaxial cables have F-Type and BNC connectors.

Frame type

The Ethernet frame size is 64 to 1518 bytes (this is the same for all Ethernet standards). Four frame types are supported:

  • Ethernet 802.3 is the original Ethernet frame type.
  • Ethernet 802.2 is the frame type that accommodates standards set by the IEEE 802.2 committee related to the logical link control (LLC) sublayer. It is a more current frame type than 802.3.
  • Ethernet II is a frame type that provides the ability to use TCP/IP as a transport/network layer protocol. Other Ethernet frame types operate strictly with IPX/SPX as a transport/network layer protocol.
  • Ethernet SNAP (SubNetwork Address Protocol) is an enhanced version of Ethernet 802.2 that allows for greater compatibility with other network architectures such as Token Ring. This frame type also supports TCP/IP.

Physical address

The MAC address (also called the burned-in address) is the Data Link layer physical device address.

  • The MAC address is a 12-digit hexadecimal number (each number ranges from 0-9 or A-F).
  • The address is often written as 00-B0-D0-06-BC-AC or 00B0.D006.BCAC, although dashes, periods, and colons can be used to divide the MAC address parts.
  • The MAC address is guaranteed unique through design. The first half (first 6 digits) of the MAC address is assigned to each manufacturer. The manufacturer determines the rest of the address, assigning a unique value which identifies the host address. A manufacturer that uses all the addresses in the original assignment can apply for a new MAC address assignment.

Note: Some network cards allow you to change (logically assigned address) the MAC address through jumpers, switches, or software. However, there is little practical reason for doing so.