Ethernet Switching Overview

Ethernet Switching

• Operates in data link and physical layers.
• Defined by the IEEE 802.2 and 802.3 standards.
• LLC and MAC are two sublayers of the data link layer.
• LLC (Logical Link Control) places information in the frame to identify the network layer protocol being used.
• MAC (Media Access Control) handles data encapsulation, media access control, and provides data link layer addressing.
• Data Encapsulation includes:
  • Ethernet frame: internal structure of the frame.
  • Ethernet Addressing: uses source and destination MAC addresses to deliver frames between NICs on the same LAN.
  • Ethernet Error Detection: uses a frame check sequence (FCS) trailer.
• The MAC sublayer defines specifications for Ethernet communication standards over various media (copper, fiber).
• Legacy Ethernet uses a shared, half-duplex medium with a bus topology or hubs.
• Half-duplex communication uses CSMA/CD (Carrier Sense Multiple Access/Collision Detection).
• Modern Ethernet LANs use switches that operate in full-duplex.
• Full-duplex communication does not require CSMA/CD for access control.
• The minimum Ethernet frame size is 64 bytes and the maximum is 1518 bytes. The preamble field is not included in the frame size.
• Frames less than 64 bytes are considered collision fragments and are discarded.
• Frames larger than 1500 bytes are considered jumbo or baby giant frames.
• If the size of a transmitted frame is outside the limits, the receiving device drops the frame.
• Dropped frames are likely caused by collisions or unwanted signals and are considered invalid.
• Jumbo frames are often supported by Fast Ethernet and Gigabit Ethernet switches and NICs.
• An Ethernet MAC address is a 48-bit binary value, expressed using 12 hexadecimal values (00 to FF).
• Leading zeroes are always displayed in hexadecimal representation to complete the 8-bit representation.
• Hexadecimal values are often represented with a '0x' prefix (e.g., 0x73) or a subscript '16' (e.g., 73_16) or an 'H' suffix (e.g., 73H).

MAC Addressing

• All network devices on an Ethernet LAN are connected to the same, shared media.
• MAC addressing identifies devices at the data link layer.
• An Ethernet MAC address is a 48-bit address consisting of 12 hexadecimal digits (6 bytes in length).
• All MAC addresses must be unique.
• Vendors selling Ethernet devices must register with the IEEE to obtain a unique 6 hexadecimal (24-bit) code called an OUI (Organizationally Unique Identifier).
• An Ethernet MAC address consists of a 6 hexadecimal vendor code followed by a 6 hexadecimal NIC-assigned value.

Switch Operation

• When forwarding a message, an Ethernet NIC includes a Source MAC address and a Destination MAC address.
• When receiving a frame, an NIC checks the Destination MAC address against its own physical MAC address. If a match is found, the frame is passed to higher OSI layers. If there is no match, the frame is discarded.
• NICs accept frames with unicast or multicast destination MAC addresses if the host is a member of that group.
• Every network device with an Ethernet NIC has a MAC address.
• Switches make forwarding decisions based solely on the Layer 2 Ethernet MAC addresses.
• When a switch is turned on, its MAC address table is empty.
• Switches use Content Addressable Memory (CAM) to store MAC address table entries.
• Each frame entering a switch is checked for new information. The Source MAC address and port number are used to create or update the MAC address table.
• Entries in the MAC address table are typically kept for 5 minutes.
• If a Source MAC address is found in the table but on a different port, it is treated as a new entry. The entry is replaced with the most current port number.
• If the Destination MAC address is a unicast address, the switch looks for a match in its MAC address table. If found, the frame is forwarded out the specified port. Otherwise, the frame is flooded out all ports except the incoming port (unknown unicast).
• Broadcast and multicast frames are flooded out all ports except the incoming port.

Duplex Modes

• Full-duplex: both ends of the connection can send and receive simultaneously.
• Half-duplex: only one end of the connection can send at a time.

Autonegotiation

• Autonegotiation is an optional function on Ethernet switches and NICs. It allows devices to automatically negotiate the best speed and duplex capabilities.
• Gigabit Ethernet ports operate in full-duplex only.

Duplex Mismatch

• Duplex mismatch is a common cause of performance issues on 10/100 Mbps Ethernet links.
• It occurs when one port on the link operates in half-duplex and the other operates in full-duplex.
• This can occur when ports are reset and autonegotiation does not result in consistent configurations.
• Best Practice: Configure both Ethernet switch ports to full-duplex.

Auto-MDIX (Automatic Medium-Dependent Interface Crossover)

• Most switches support auto-MDIX.
• It automatically detects the type of cable connected to the port and configures the interfaces accordingly.
• Auto-MDIX is enabled by default on switches running Cisco IOS Release 12.2(18)SE or later but can be disabled.
• Best Practice: Use the correct cable type and do not rely on auto-MDIX.
• The 'mdix auto' command can re-enable auto-MDIX.

Network Layer

• The Network Layer provides services for end devices to exchange data.

• Primary Network Layer communication protocols are IP version 4 (IPv4) and IP version 6 (IPv6).

• The four basic operations of the network Layer are:

  • Addressing
  • Routing
  • Packet forwarding
  • Fragmentation

• Routers provide services for packet forwarding and fragmentation.

• Routing is the process of determining the path that data will travel between source and destination networks.

• Routes that routers do not have a direct connection to can be learned either manually or dynamically.

• Manually configuring routes is done with static routes.

• Dynamically learning routes utilizes routing protocols to share information between routers.

• A default route forwards all traffic to a specific direction if a matching route is not found in the routing table.

• Default routes must be configured manually.

• Dynamic routing automatically discovers remote networks, maintains updated information, chooses the best path, and finds new paths when the topology changes.

• show ip route command displays route sources and types:

  • L: directly connected local interface IP address.
  • C: directly connected network.
  • S: static route configured manually.
  • O: OSPF
  • D: EIGRP

• Directly Connected routes: C and L

• **Remote Routes: ** O, D, etc

• Default Routes: S

Explore the fundamental concepts of Ethernet switching in this quiz, focusing on its operations within the data link and physical layers. Understand the key components such as LLC and MAC sublayers, data encapsulation, and error detection methods. Test your knowledge on Ethernet standards and communication techniques.