SlideShare a Scribd company logo

Ethernet

•
•
S
selvakumar_b1985

Ethernet uses CSMA/CD access method where nodes can sense carrier and detect collisions. It was first defined in 1978 and formed basis for IEEE 802.3 standard. It uses exponential backoff to retry transmission after collisions and is limited to 2500m to ensure collisions can be detected. Ethernet addresses are unique to each adapter and frames contain fields for source, destination, data and error checking.

1 of 20
Ethernet IEEE 802.3
Ethernet IEEE 802.3 – Ethernet Uses CSMA/CD technology. So Ethernet is a multiple access network. i.e., set of nodes send and receive frames over a shared link.  CSMA/CD (Carrier sense Multiple Access / Collision Detect),  Carrier sense – all nodes can distinguish between an idle and a busy line.  Collision Detect – node listens as it transmit and can therefore detect when a frame it is transmitting has collided with a frame transmitted by another node. Uses ALOHA as the root protocol. Digital Equipment Corporation and Intel joined Xerox to define 10Mbps Ethernet standard in 1978. This standard formed the basis for IEEE standard 802.3
Ethernet Physical Mediums, It includes 100Mbps fast Ethernet & 1000Mbps or 1Gbps Gigabit Ethernet. Others are 10Base5, 10Base2, 10BaseT, 100BaseT, 1Base5, 10Broad36. Fast Ethernet 100 Base T4 –4 pairs of UTP 100 Base X 100 Base Tx –2 pairs of UTP or STP 100 Base Tx –2 optical fibers
Ethernet Physical Properties,  Initially Ethernet was implemented on a coaxial cable up to 500m.  This cable is similar to the type used for cableTV except that it typically has an impedance of 50 ohms instead of cable TV’s 75 ohms.  Hosts connect to an Ethernet segment by tapping into it. Transceiver,  Small device directly attached to the tap that detects when the line is idle and drives signal when the host is transmitting.  It also receives incoming signal.  The transceiver is connected to an Ethernet adaptor which is plugged into the host.  Protocol is implemented on the adaptor.
Ethernet
Ethernet Repeater,  Multiple Ethernet segments can be joined together by a repeater.  It forwards a digital signal.  No more than four repeaters may be positioned between any pair of hosts.  Ethernet has a total reach of only 2500m.  Any signal placed on the Ethernet by a host is broadcast over the entire network.  Signal is propagated in both directions.  Repeaters forward the signal on all outgoing segments.  Terminators attached to the end of each segment absorb the signals.  Ethernet uses Manchester encoding scheme.
Ethernet In 10 Base 2 (thin Ethernet), the cable is thin, so T-joint spliced into the cable instead of tap the cable. Therefore 10 Base 2 is used in daisy chain configuration. With 10 Base T, the common configuration is to have several point segments coming out of a multiway repeater called Hub.
Ethernet Access Protocol, The algorithm is commonly called Ethernet’s Media Access Control (MAC).  MAC is implemented on the network adaptor.
Ethernet Preamble – allows the receiver to synchronize with signal (sequence of alternating 0s & 1s) Src & dest addr – unique address belongs to the adaptor in the source and destination host. Type – packet type acts as demux key to identify the higher level protocol. Data – original data up to 1500 bytes. Minimally a frame must contain at least 46 bytes of data, because frame must be long enough to detect collision. CRC – error detection code.
Ethernet Ethernet Addresses  Every Ethernet host in the world has a unique Ethernet addresses.  This address belongs to the adaptor on the host.  It is usually burnt in to the ROM.  Ethernet addresses are typically printed in a human readable format.  Sequence of six numbers separated by colons.  1 byte of the 6 bytes address is given by a pair of hexadecimal digits, one for each of the 4 bit nibbles in the byte. e.g, 8:0:2b:e4:b1:2 – 00001000 00000000 00101011 11100100 10110001 00000010  To ensure that every adaptor gets a unique address, each manufacturer of Ethernet device is allocated a different prefix that must be prepended to the address every adaptor they build. e.g, AMD has assigned the 24bit prefix 8:0:20
Ethernet EthernetTransmitting Algorithm When the adaptor has a frame to send & line is idle, it transmits the frame immediately. The upper bound of 1500bytes in the message means that the adaptor can occupy the line for a fixed length of time. When the adaptor has a frame to send and the line is busy, it waits for the line to go idle then transmits immediately. Ethernet is said to be 1 – persistent protocol, because an adaptor with the frame to send transmits with probability 1 whenever a busy line goes idle.
Ethernet Since there is no centralized control, it is possible for two or more adaptors to begin transmitting at the same time. Either because both found the line to be idle. or, both been waiting for a busy line to become idle. When this happens, the two or more frames are said to be collide on the network.
Ethernet Collision Detection Since Ethernet supports collision detection, so each sender is able to determine that a collision in progress. At the moment an adaptor detects that, its frame is colliding with another, Transmitter will send a 32 bit jamming sequence with 64-bit preamble (totally 96 bits) This 96 bit frame is called as runt frame.
Ethernet Worst cast scenario, (two nodes are at opposite ends of the Ethernet) To know, the frame it just sent did not collide with another frame, the transmitter may need to send as many as 512 bits. 512 bits (64 bytes) -> 14 bytes of header + 46 bytes of data + 4 bytes of CRC Why 512 bits & why its length is limited to 2500m? The farther apart two nodes are, the longer it takes for a frame sent by one to reach the other and the network is vulnerable to collision during this time.
Ethernet Fig. (a) A sends a frame at time t; (b) A’s frame arrives at B at time t + d; (c) B begins transmitting at time t + d and collides with A’s frame; (d) B’s runt (32-bit) frame arrives at A at time t + 2d.
Ethernet ‘A’ begins transmitting frame at time ‘t’  The first bit of A’s frame arrives at ‘B’ at time t + d, where d is one link latency.  Suppose an instant before host A’s frame arrives, host B begins to transmit its own frame. B’s frame will immediately collide with A’s frame and this frame collision will be detected by host B.
Ethernet Host B will send the 32 bit jamming sequence.  Host A will not know that the collision occurred until B’s frame reaches it, which will happen at t + (2*d) Host A must continue to transmit until this time in order to detect collision. Determined delay in the maximally configured Ethernet (2500m long) is 51.2 s. i.e., delay on 10Mbps Ethernet corresponds to 512 bits. By limiting the maximum length of the Ethernet to 2500m we can limit the Ethernet’s maximum latency to 51.2 s (fairly small value). So the access algorithm works efficiently.
Ethernet Exponential Back off, Once an adaptor has detected a collision & stopped its transmission, it waits a certain amount of time and tries again. Each time the adaptor tries to transmit but fails, it doubles the amount of time it waits before trying again. This strategy of doubling the delay interval between each retransmission attempt is known as exponential back off.
Ethernet The adaptor first delays either 0 or 51.2 s selected at random. If this effort fails, it then waits 0, 51.2s, 102.4s, 153.6s… before trying again. i.e., k * 51.2s, k= 0,1,2,3. After the third collision, it waits k * 51.2s, k=0… 23 – 1 In general, the algorithm randomly selects a ‘k’ between 0 and 2n – 1 & waits for k * 51.2s where ‘n’ is the number of collisions experienced so far.
Ethernet Advantages  Ethernets are easy to administer & maintain.  There are no switches that can fail & no routing and configuration tables that have to be kept up to date.  Easy to add a new host to the network.  In expensive (i.e., cable is cheap) Disadvantages  Ethernets work best under lightly loaded conditions.  Most Ethernets are used in conservative way.  Most Ethernets are far shorter than 2500m with a round trip delay of closer to 5s than 51.2s.

More Related Content

Ethernet

  • 2. Ethernet IEEE 802.3 – Ethernet Uses CSMA/CD technology. So Ethernet is a multiple access network. i.e., set of nodes send and receive frames over a shared link.  CSMA/CD (Carrier sense Multiple Access / Collision Detect),  Carrier sense – all nodes can distinguish between an idle and a busy line.  Collision Detect – node listens as it transmit and can therefore detect when a frame it is transmitting has collided with a frame transmitted by another node. Uses ALOHA as the root protocol. Digital Equipment Corporation and Intel joined Xerox to define 10Mbps Ethernet standard in 1978. This standard formed the basis for IEEE standard 802.3
  • 3. Ethernet Physical Mediums, It includes 100Mbps fast Ethernet & 1000Mbps or 1Gbps Gigabit Ethernet. Others are 10Base5, 10Base2, 10BaseT, 100BaseT, 1Base5, 10Broad36. Fast Ethernet 100 Base T4 –4 pairs of UTP 100 Base X 100 Base Tx –2 pairs of UTP or STP 100 Base Tx –2 optical fibers
  • 4. Ethernet Physical Properties,  Initially Ethernet was implemented on a coaxial cable up to 500m.  This cable is similar to the type used for cableTV except that it typically has an impedance of 50 ohms instead of cable TV’s 75 ohms.  Hosts connect to an Ethernet segment by tapping into it. Transceiver,  Small device directly attached to the tap that detects when the line is idle and drives signal when the host is transmitting.  It also receives incoming signal.  The transceiver is connected to an Ethernet adaptor which is plugged into the host.  Protocol is implemented on the adaptor.
  • 6. Ethernet Repeater,  Multiple Ethernet segments can be joined together by a repeater.  It forwards a digital signal.  No more than four repeaters may be positioned between any pair of hosts.  Ethernet has a total reach of only 2500m.  Any signal placed on the Ethernet by a host is broadcast over the entire network.  Signal is propagated in both directions.  Repeaters forward the signal on all outgoing segments.  Terminators attached to the end of each segment absorb the signals.  Ethernet uses Manchester encoding scheme.
  • 7. Ethernet In 10 Base 2 (thin Ethernet), the cable is thin, so T-joint spliced into the cable instead of tap the cable. Therefore 10 Base 2 is used in daisy chain configuration. With 10 Base T, the common configuration is to have several point segments coming out of a multiway repeater called Hub.
  • 8. Ethernet Access Protocol, The algorithm is commonly called Ethernet’s Media Access Control (MAC).  MAC is implemented on the network adaptor.
  • 9. Ethernet Preamble – allows the receiver to synchronize with signal (sequence of alternating 0s & 1s) Src & dest addr – unique address belongs to the adaptor in the source and destination host. Type – packet type acts as demux key to identify the higher level protocol. Data – original data up to 1500 bytes. Minimally a frame must contain at least 46 bytes of data, because frame must be long enough to detect collision. CRC – error detection code.
  • 10. Ethernet Ethernet Addresses  Every Ethernet host in the world has a unique Ethernet addresses.  This address belongs to the adaptor on the host.  It is usually burnt in to the ROM.  Ethernet addresses are typically printed in a human readable format.  Sequence of six numbers separated by colons.  1 byte of the 6 bytes address is given by a pair of hexadecimal digits, one for each of the 4 bit nibbles in the byte. e.g, 8:0:2b:e4:b1:2 – 00001000 00000000 00101011 11100100 10110001 00000010  To ensure that every adaptor gets a unique address, each manufacturer of Ethernet device is allocated a different prefix that must be prepended to the address every adaptor they build. e.g, AMD has assigned the 24bit prefix 8:0:20
  • 11. Ethernet EthernetTransmitting Algorithm When the adaptor has a frame to send & line is idle, it transmits the frame immediately. The upper bound of 1500bytes in the message means that the adaptor can occupy the line for a fixed length of time. When the adaptor has a frame to send and the line is busy, it waits for the line to go idle then transmits immediately. Ethernet is said to be 1 – persistent protocol, because an adaptor with the frame to send transmits with probability 1 whenever a busy line goes idle.
  • 12. Ethernet Since there is no centralized control, it is possible for two or more adaptors to begin transmitting at the same time. Either because both found the line to be idle. or, both been waiting for a busy line to become idle. When this happens, the two or more frames are said to be collide on the network.
  • 13. Ethernet Collision Detection Since Ethernet supports collision detection, so each sender is able to determine that a collision in progress. At the moment an adaptor detects that, its frame is colliding with another, Transmitter will send a 32 bit jamming sequence with 64-bit preamble (totally 96 bits) This 96 bit frame is called as runt frame.
  • 14. Ethernet Worst cast scenario, (two nodes are at opposite ends of the Ethernet) To know, the frame it just sent did not collide with another frame, the transmitter may need to send as many as 512 bits. 512 bits (64 bytes) -> 14 bytes of header + 46 bytes of data + 4 bytes of CRC Why 512 bits & why its length is limited to 2500m? The farther apart two nodes are, the longer it takes for a frame sent by one to reach the other and the network is vulnerable to collision during this time.
  • 15. Ethernet Fig. (a) A sends a frame at time t; (b) A’s frame arrives at B at time t + d; (c) B begins transmitting at time t + d and collides with A’s frame; (d) B’s runt (32-bit) frame arrives at A at time t + 2d.
  • 16. Ethernet ‘A’ begins transmitting frame at time ‘t’  The first bit of A’s frame arrives at ‘B’ at time t + d, where d is one link latency.  Suppose an instant before host A’s frame arrives, host B begins to transmit its own frame. B’s frame will immediately collide with A’s frame and this frame collision will be detected by host B.
  • 17. Ethernet Host B will send the 32 bit jamming sequence.  Host A will not know that the collision occurred until B’s frame reaches it, which will happen at t + (2*d) Host A must continue to transmit until this time in order to detect collision. Determined delay in the maximally configured Ethernet (2500m long) is 51.2 s. i.e., delay on 10Mbps Ethernet corresponds to 512 bits. By limiting the maximum length of the Ethernet to 2500m we can limit the Ethernet’s maximum latency to 51.2 s (fairly small value). So the access algorithm works efficiently.
  • 18. Ethernet Exponential Back off, Once an adaptor has detected a collision & stopped its transmission, it waits a certain amount of time and tries again. Each time the adaptor tries to transmit but fails, it doubles the amount of time it waits before trying again. This strategy of doubling the delay interval between each retransmission attempt is known as exponential back off.
  • 19. Ethernet The adaptor first delays either 0 or 51.2 s selected at random. If this effort fails, it then waits 0, 51.2s, 102.4s, 153.6s… before trying again. i.e., k * 51.2s, k= 0,1,2,3. After the third collision, it waits k * 51.2s, k=0… 23 – 1 In general, the algorithm randomly selects a ‘k’ between 0 and 2n – 1 & waits for k * 51.2s where ‘n’ is the number of collisions experienced so far.
  • 20. Ethernet Advantages  Ethernets are easy to administer & maintain.  There are no switches that can fail & no routing and configuration tables that have to be kept up to date.  Easy to add a new host to the network.  In expensive (i.e., cable is cheap) Disadvantages  Ethernets work best under lightly loaded conditions.  Most Ethernets are used in conservative way.  Most Ethernets are far shorter than 2500m with a round trip delay of closer to 5s than 51.2s.