1 / 40

Chapter 11

Chapter 11. Ethernet Evolution: Fast and Gigabit Ethernet. Revolution : From a LAN to a bridged LAN From a bridged LAN to a switched LAN From a switched LAN to a full-duplex switched LAN From a 10 Mbps Ethernet to a Fast Ethernet (100 Mbps)

Download Presentation

Chapter 11

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Chapter 11 Ethernet Evolution: Fast and Gigabit Ethernet Revolution: From a LAN to a bridged LAN From a bridged LAN to a switched LAN From a switched LAN to a full-duplex switched LAN From a 10 Mbps Ethernet to a Fast Ethernet (100 Mbps) From a Fast Ethernet to a Gigabit Ethernet (1000 Mbps)

  2. Figure 11-1 Sharing Bandwidth In an unbridged network, the total capacity (10Mbps) is shared among all stations with a frame to send.

  3. Figure 11-2 11.1 BRIDGED ETHERNET A Network with and without a Bridge A bridge divides the network into two or more segments. Bandwidth-wise, each segment is independent. Each station is theoretically offered 10/12 Mbps in a network with a heavy load. Each station is now offered 6/12 Mbps if the traffic is not going through the bridge. This is the first advantage of using a bridge.

  4. Figure 11-3 Collision Domains in a Non-bridged and Bridged Network Second advantage of a bridge: separation of collision domain 12 stations contend for access to the medium (4-port bridge) Only 3 stations contend for access to the medium

  5. Figure 11-4 Switched Ethernet (Sec. 11.2) A layer-2 switch is an N-port bridge with additional sophistication that allows faster handling of the packets.

  6. Figure 11-5 11.3 FULL-DUPLEX ETHERNET Full-Duplex Switched Ethernet No need for carrier sensing or collision detection. No need for CSMA/CD access method. The CS and CD functionality of the MAC sublayer can be turned off.

  7. 11.4 MAC CONTROL Traditional Ethernet : Connectionless at the MAC sublayer, using CSMA/CD. There is no explicit flow control or error control to inform the sender. Receiver does not send any positive or negative acknowledgement. • In traditional Ethernet, there are two sources of errors: • The frame is corrupted during transmission. • Probability of bit error rate (BER) is very low: 10 ^ (-10) • If it happens, the receiver discards the frame by checking FCS and • the upper layer (LLC) notifies the sender for retransmission • The frame is lost due to collision. • Sender can detect this situation (using CD protocols) and resend the frame. • In full-duplex switched Ethernet, there are also two sources of errors: • The frame is corrupted during transmission. • This is the same situation as in traditional Ethernet. • The frame is lost because the switch buffer is full • The switch discards the frame. • But, there is no CSMA/CD mechanism to inform the sender that the packet • is lost.  There is a need for an explicit ACK in full-duplex switched Ethernet.

  8. Figure 11-6 MAC Control Layer (optional) Why MAC control sublayer: for flow and error control. How: by inserting special control packets between data packets coming from the upper layers. The MAC control packet is encapsulated in a MAC frame in the same way as the data packets. For efficiency, a frame carrying a MAC control packet should be as smallas possible  the size of a MAC control packet <=46 bytes

  9. Figure 11-7 Encapsulation of a MAC Control Packet in a MAC Frame DA: device (station or switch) at the other end of the link, not the final destination of the data frame. A special multicast address 01-08-C2-00-00-01 for three reasons: 1. The sender does not need to know the address of the device at the other end. 2. This address is blocked by all bridges and switches. 3. This address is ignored by all stations that do not use MAC control option. SA: address of the device that sends the MAC control packets Type/length: indicates the type of the frame, not the length which is of fixed size. FCS: CRC error detection field Code: MAC control packet identifier, 0001 (based 16) for PAUSE packets.

  10. Figure 11-8 Interleaving of MAC Control Frames

  11. Figure 11-9 Format of the PAUSE Packet PAUSE packet: Currently, thisis the only packet defined by the MAC control sublayer. Purpose: used to temporarily slow down the flow of frames between two device (switch or station) connected at the ends of a full-duplex link. It provides a very simple flow control (called stop-start) Code parameters Code: a 2-byte field with the value 0001 (base 16) Parameters: the only parameter defined is the parameter p for pause-time actual pause-time = p * slot time, where slot-time = the duration of 512 bits Rule: current PAUSE packet overwrites previous PAUSE packet

  12. Figure 11-10 11.5 FAST ETHERNET Layers in Fast Ethernet LLC is the same as discussed in Chap. 9

  13. 11.6 FAST ETHERNET MAC SUBLAYER Main idea: keep the MAC sublayer for 10 Mbps Ethernet untouched. Frame format, min. and max. frame length, and addressing are the same for both 10- and 100-Mbps Ethernet. Access method: CSMA/CD (no need for full-duplex Fast Ethernet) The implementation keeps CSMA/CD for backward compatibility Slot time: duration of 512 bits = 51.2 microseconds for 10-Mbps Ethernet = 5.12 microseconds for 100-Mbps Ethernet Slot time and max. network length: Max. length = propagation speed * (slot time / 2) = 5120 m for 10-Mbps Ethernet (= (2*10^8)(51.2*10^(-6) / 2)) (2500 m specified) = 512 m for 100-Mbps Ethernet (= (2*10^8)(5.12*10^(-6) / 2)) (250 m specified) Auto negotiation (a new feature added): 1. to allow two devices to negotiate the mode (half- or full-duplex) or data rate of operation (10- or 100-Mbps) 2. to allow one device to have multiple capabilities 3. to allow a station to check a hub’s capabilities

  14. Figure 11-11 Example of Auto Negotiation Notes on negotiation: 1. Covers only the link, not the whole network (btw. a station & a hub or btw. two hubs) 2. Can only occur during link initialization, based on common capabilities 3. Each device at the end of the link advertises its capabilities to the other 3. A hierarchy of common capabilities is defined to facilitate the decision 4. Use a separate frame format and signaling system

  15. Figure 11-12 Auto Negotiation Message Format Selector field: defines the type of the LAN technology (=10000 for Ethernet) Ability field: for advising the sender’s capabilities, when set A0: 10Base-T A1: 10Base-T full duplex A2: 100Base-TX A3: 100Base-T dull-duplex A4: 100base-T4 A5: Pause operation A6: Reserved A7: Reserved Fault bit: announces that a fault has occurred when set Ack bit: announces that a message was successfully received when set Next page bit: another message coming (called next page) when set

  16. Figure 11-13 11.7 Fast Ethernet Physical Layer passing data in 4-bit format (nibble) to MII MII is needed only for external transceiver Can be used with 10– and 100-Mbps data rate Transceiver (encoding/decoding) medium dependent

  17. Figure 11-14 MII MII operates at both 10– and 100-Mbps Backward compatible

  18. Figure 11-15 Signals in MII 4-bits parallel data path (TX data) 4-bits parallel data path (RX data) Inside NIC

  19. Figure 11-16 MII Connector 40-pin D-connector 20 twisted-pair cables for synchronization btw PHY & Reconciliation sublayer

  20. Figure 11-17 MII Cable

  21. Figure 11-18 Fast Ethernet Implementations 2-wire implementation 4-wire implementation uses 2 pairs of C-5 UTP or STP uses 2 strands of fiber- optical cables uses 4 pairs of C-3 UTP

  22. Figure 11-19 100Base-TX Implementation 2 pairs: one for TX, one for RV can operate in full-duplex mode NIC with internal transceiver In full-duplex mode, frames are buffered and switched internally within the hub (like a switch)

  23. Figure 11-20 Encoding and Decoding in 100Base-TX 3-levels, multiline Transmission, see next slide 4 parallel bits from NIC Inside the transceiver

  24. Figure 11-21 MLT-3 Signal Similar to NRZ-I, but uses three levels Transition at the beginning of a 1 bit No transition at the beginning of a 0 bit

  25. Figure 11-22 100Base-FX Implementation

  26. Figure 11-23 Encoding and Decoding in 100Base-FX 4 parallel bits from NIC

  27. Figure 11-24 NRZ-I Encoding

  28. Figure 11-25 100Base-T4 Implementation 4 pairs of C-3 UTP Encoding/decoding is more complex here Uses 8B/6T (8 binary bits  6 ternary signals) See Appendix K

  29. Figure 11-26 Example of 8B/6T Encoding - + 0 + - 0 8 binary bits 6 ternary signals as a single unit (each unit represents 8 bits)

  30. Figure 11-27 Using Four Wires in 100Base-T4 Mbaud 33.3 Mbps

  31. Figure 11-28 Layers in Gigabit Ethernet

  32. Figure 11-29 Two Approaches in Gigabit Ethernet Medium Access

  33. Figure 11-30 A Frame Using Carrier Extension Method = =

  34. Figure 11-31 Frame Bursting Approach

  35. Figure 11-32 Gigabit Ethernet Physical Layer

  36. Figure 11-33 Gigabit Ethernet Implementations

  37. Figure 11-34 1000Base-X Implementation

  38. Figure 11-35 Encoding in 1000Base-X

  39. Figure 11-36 1000Base-T Implementation

  40. Figure 11-37 Encoding in 1000Base-T

More Related