Part 6: The IP protocol (and UDP) are called \"connectionless\" and \"unreliable
ID: 3727110 • Letter: P
Question
Part 6: The IP protocol (and UDP) are called "connectionless" and "unreliable." Describe what those two words mean in the context of data communications.
Part 7: What is an Ethernet collision? Explain the process by which Ethernet recovers from a collision.
Part 8: Explain how RAID can protect against disk hardware failure. (You need about a paragraph of detail; a single sentence will not do.)
Part 9: Explain how collisions are avoided in modern Ethernet network design. (You need about a paragraph of detail; a single sentence will not do.)
Part 10: Explain how an Ethernet switch determines to which port to forward an incoming packet. Your answer must cover both the case of a destination that has previously made a transmission and one that has not.
Explanation / Answer
Part 6
Connectionless : It does not require to create a session between sender and reciever before transporting the packets.Neither Sender and reciever doesn't maintain any state information to send or to recieve data.They just send data or recieve data from each other.
Unreliable : The sender just keep sending the packets to destination without regard to whether they are actually reach the reciever.That means there is no acknowledgment to guarantee the transmission.
Part 7
In a network if two or more than two devices attempt to transmit data i.e. want to use the medium channel at the same time,then,all of the device sense that some of the other devices is also trying to access the channel.This incident is called 'Ethernet Collision'.
How ethernet recovers from collision can be described in the following way :
1.If frame ready for transmission then,transmit it.If collision occurs,then a jam signal will be sent to this node and that node will detect collision and goto 2.If not collision detected then continue transmission.
2.Continue transmission for minimum packet time so that,all the devices that were trying to access the channel will recieve the jam signal meanwhile.
3.This was a failed attempt to transmit.So,increase the retransmission counter by 1.
4.Check whether maximum attempt reched?If yes then abort transmission.If no then,goto 5.
5.Calculate the back-off time using the number of collision and contention period.And wait for back-off time.
6.Goto 1.
Part8
RAID(redundant array of independent disks) is a way of storing a particular data on multiple hard disks to protect data in the case of a disk drive failure.
RAID place data on multiple disks and allow the input/output task to be operated in a balanced way,this way performance will be increased,as the use of multiple disk drive increases the mean time between failures (MTBF).It also increases fault tolerance by storing data redundantly.RAID is located generally in OS as a single logical hard disk. RAID involves the techniques of disk mirroring or disk striping. Disk mirroring does what,it copies identical data onto more than one drive.Disk striping does what,it partitions each drive's storage space into unit spaces with different sectors of different MB's.The addresses of all the disk drives are interleaved and addressed in order.
Performance, flexibility and cost are among the major benefits of RAID. By combining multiple hard drives together,RAID reduces load on single hard drive.Depending on how this is set up,it can increase computer speed and reliability after a crash.
Part9
1.Carrier Sense : Before transmitting data it first sense the shared medium channel to determine whether any other node is transmitting or not.
2.Collision Avoidance : If it detects that any other node is transmitting,it stops to listen to the channel for a amount of time before transmitting again.
RTS/CTS(Request to send/clear to send),it can request the base station to mediate access to the channel and base station can send clear to send.
If no node was found transmitting,then it sends data.If acknowledgment do not come then ,again wait for back-off time(calculate this back-off time).