N E T W O R K I N G  W E E K  4

• Addressing scheme
  • (Fixed header [two addresses or 'fields'], physical address, hardware address, media address [MAC])
  • Packet (has to be a minimum size)
  • Speed is 23x10^8 m/s
  • EXAMPLE SCENARIO
    • Network Lab
    • If our minimum size packet is 512 bits, how long can our path be(so that the 1st bit can get back to the host before the message is done being sent.)?
    • Lab line = 10 MB per second.
      • 10^7 bits/sec.
    • How much time to send 1 bit?
      • 10^7 sec./bit so it travels....
        • 106^-7 sec./bit * 2.3 * 10^8 = 23 meters/sec.
      • 23 meters in a second.
    • So 256 bits or 5888 meters length.

• Checks the length (if falls between min. and max.).
• Checks the CRC.
• Discards the frames that contain errors.
• Performs functions of sending and receiving without the use of the computers CPU.

• Schemes are dependent on LAN technology used.
  • Static addressing scheme relies on the hardware manufacturer to assign a unique physical address to each network interface. The address doe not change unless the hardware is changed.
  • Configurable addressing scheme provides a mechanism that a user can set a physical address.
  • Dynamic addressing scheme has a mechanism that automatically assigns a physical address at boot up time. Usually uses a algorithm that tries random numbers until one is found that is not on the network.

• When a broadcast occurs only the named hardware will respond
• For efficiency todays technology includes a defined special reserved broadcast address. Computers are built to recognize this address along with its physical address and will respond to both.
• Very inefficient because of the extra processing.

• Fixes the broadcasting in-efficiency.
• instead of using the CPU the hardware interface must make the decision to respond or ignore.

• Uses an extension of the basic addressing scheme.
• Extends by reserving some addresses for multicast.

• Explicit
  • Frame type field in which the frame is self identifying.
• Implicit
  • Sender and receiver must agree on what is being sent.

64 bit preamble	48 bit Destination Address	48 bit Source Address	Frame Type	Data in Frame	CRC
0101...111	Header			Payload including the IP address	10

• Sender and receiver have already agreed on what is being sent .......OR
• Sender and receiver use a packet with a header that specifies the data.
  • IEEE LLC/SNAP header:

• Standard laptop with an ethernet card.
• Places its own hardware into promiscuous mode to access all frames.
• tcpdump is an example

• € = Efficiency of protocol
• ¥ = Size/Speed (Time to transmit)
• £ = Time for signal to propagate the entire length
• @ = £/¥ (Number of packets in line)

  € = 1/(1+5@)

• Only as fast as the hardware (CPU) can support.

• Network Interface Card is like an I/O device.

• Thick coaxial (needs transceiver and AUI cable)

• Electronic device that allows multiple computers to attach to a network through a single transceiver.

• Flexible coaxial cable that attaches directly to each computer without a separate transceiver.

• 10 base -T uses a hub and twisted pair wiring

• A maximum length specification is a fundamental part of LAN technology.
• LAN hardware is engineered for a fixed maximum length cable, and the hardware will not work correctly over wires that exceed the bound.

• A pair of fiber modems and optical fibers can be used to provide a connection between the network interface on a computer and a remote transceiver.

• The repeater connects directly to the cable
• A repeater is a hardware device used to extend a LAN.
• The repeater amplifies and sends all electrical signals that occur on one segment to the other segment.
• Computers can not tell if a repeater separates them

• An adaptive bridge examines physical addresses in the header of each frame it receives.
• The bridge uses the source address to automatically determine the location of the computer that sent a frame, and uses the destination address to determine whether to forward a frame.

• Propagation principle for bridged networks
  • In the steady state, a bridge forwards each frame only as far as necessary.

• Is like a repeater except a bridge handles complete frames and uses the same network interface as a conventional computer.
• Planning
  • Connect computers that talk frequently to the same segment so others can be busy on other segments

• Simple - which way has the less hops?

Each bridge decides the ports over which it is and is not willing to forward frames.

The bridges are prepared to reconfigure dynamically into a new spanning tree should some bridge fails.

Algorithm:

1. Each bridge contains a database:
   • Who is root
   • Which port to root
   • How far is root
   • Status of each port
   • best route to rout
   • designated for forwarding
   • disabled
2. Each bridge is assigned a unique identifier.
3. Bridge with the smallest ID is the root of the spanning tree.
4. The root always ports frames out over all its ports.
5. Each bridge computes the shortest path to the root and makes note of which of its ports is used(Proffered Path).
6. All the bridges in a LAN elect a single designated bridge that will be responsible for forwarding frames toward the root bridge( bridge that is closest to the root).
7. All this done with communication between bridges. Packet includes:
   • (Source) id for the bridge that is sending the message.
   • the distance in hops from the sending bridge to the root bridge.
   • (Root) id for what the sending bridge believes to be the root bridge.
8. Algorythm starts: Initially each bridge thinks that it is the root and sends a configuration message (me, 0 , me) out on each of its ports. When the bridge gets confirmation messages over a particular port:
   • identify a root with smaller id.
   • identify a root with an equal id but a shorter distance or...
   • if the root id and distance are equal, but the sending bridge has a smaller id
9. When the bridge receives a message that it is not the root it stops sending configuration packets.