This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
All assignments or projects will have a compulsory presentation requirement which carries 10. Communication skills, visual aids, and personal grooming are the criteria for obtaining the 10%.
Answer both questions: (2Q * 15% = 30%)
(Presentation = 10%)
A group of business employees would like to set up a small networking office. Explain the meaning of topology. Discuss and draw the different types of network topology that are available.
The purpose of data link control is to provide functions like flow control, error detection and error control. Explain each function in detail.
To set up a small networking office, there will be multiple computers connected together.
When there are multiple computers connected together, there will be a network topology. Network topology refers to the shape of how to computers and other network components are connected with each other. The purpose of having network topology is a major design consideration for reliability and cost. There are several types of network topology available and each of them has their own advantages and disadvantages
To decide which type of network topology is the most suitable for a particular networking office,
we must study everything regarding network topology.
First of all I would like to explain 2 terms which is important and related to network topology.
Node is a device that is connected to a network and its purposes is same with a computer. Network topology deals with how the nodes of a network are connected with each other.
Packet is a message that is sent over to a network from one node to the other node. Packet includes addresses of the node and data.
There are 5 types of network topology that is widely used in a networking office which is:
In a networking office which has multi-point connection and each machine is connected to a single cable is called bus topology. Bus topology is one of the most popular topologies used in a networking office.
In bus topology, a terminator is required at each end of the bus cable to prevent the signal from reflecting back and forth. The signal from source will travel in both directions until it finds the
IP addresses on the network that is the intended receiver. If the machine addresses node does not match, it will then ignores the data while if it does match the machine addresses node, the data will be accepted.
The advantages of bus topology are low cost, because it only requires one cable and easy expansion of the network. But however, the disadvantage is since there's only one cable that is connected by all machines, if the cable breaks down, the entire network will be down too.
Because of this, even with low cost of implementing the network, the cost of managing the network is quite high to prevent it from breaking down.
There are 2 types of bus topology.
Linear Bus Topology
Distributed Bus Topology
Linear bus topology is a type of network where all the nodes of the network are connected to a common transmission medium that has exactly 2 endpoints. All data that is transmitted between the nodes in network is transmitted over common transmission medium and is capable to be received by all nodes in the network simultaneously.
The 2 endpoints of the common transmission medium have a device called terminator that prevent the signal being reflected back to the transmission medium in the opposite direction which will interfere other signals.
Distributed bus topology is a network where has more than 2 endpoints that are created to add more branches to the main section. It works exactly like linear bus.
Below is the example of a linear bus topology (Figure 1.1) and distributed bus topology
Star topology is connected to a central device called hub or switch. Star topology is also commonly used in networking office. In contrast to the bus topology, all the nodes in star topology are connected to the hub with a point to point connection. Physically it looks like a star but logically it works almost the same like a bus topology. For any machine to communicate with other machines in star topology, firstly the source must send the information to the hub, and then the hub must transmit that information to the destination machine. For example, if Computer B wants to send information to Computer F, it must send the information to the hub first, and then the hub will pass the information to Computer F. Only 1 signal can be transmit at one time.
Star topology is considered as the easiest topology to design and implement. The advantages are it's easy to set up and expand which means a new node can be added anytime. Another advantage is if one link to the hub breaks, the entire network will not be affected except for the only machine that is using the link. It also has its own benefits from centralization because as the central hub is the bottleneck, it increases the capacity easily. Not only that, it's also easier to detect faults and remove parts.
The disadvantages are when the hub breaks down; the entire network will be down too. Besides that, the performance is entirely dependent on the capabilities of the hub. Network size is limited by the numbers of hub available and wiring up the system might be complex and expensive.
Below is the example of star topology
Tree topology also known as hierarchical topology or star of stars topology is a combination of star and bus topology that uses an active hub to connect all the machines together. The hub is the most important component in the network because it links all machines in the network together.
The function of the hub is to accept information from one machine and repeat the information
to other machines and hub. It connects multiple star networks to other star networks. Tree topology allows the network to have many servers and can branch out in many ways. This topology can be usually seen at colleges, universities and schools because it can identify the relevant systems in their own network and yet to be able to connect into a bigger network in other way.
The structure of tree topology is widely spread and divided into many branches. Like a real tree.
It follows the hierarchical pattern where each level is connected to a higher level.
There will be at least 3 levels of hierarchy in tree topology. The higher levels in the hierarchy are expected to perform more functions than the lower levels in the network. The tree topology can be extended easily to function and there is no limitation to how big it can be extended.
Like other topologies, it has its own advantages and disadvantages too.
The advantages of tree topology are it's supported by various networks vendors and hardware vendors. All the machines or computers will have access to a larger immediate networks and it's the best topology for branch out networks. Beside these, a point to point connection is possible too.
The disadvantages for tree topology are that the length of network depends on the length of cable that is being used. It is also entirely dependent on the trunk which is the main backbone of the whole network, if it fails, the entire network will fail too. Another disadvantage is since the network is too big, it is difficult to configure after a certain point.
Below is the example of tree topology.
In ring topology, all the machines are connected in a closed loop or ring pattern. The first machine is cabled to the second and so on until it reaches back to the first machine. The repeaters are joined by a point to point links in closed loop. Each machine has their own unique address that is used for identification purposes. Firstly the signal passes through each machine that is connected to the ring in one direction. The machine transfer information to the next machine on the ring, which checks the address of the information. If the address of the information matches with the machine address, the machine copies the information and passes it to the next machine and so on until it reaches the source machine. The source then removes the information from the ring.
The advantages of ring topology are as follows:
Easy to install and expand
Can use fiber optic cable
The disadvantages are:
If a link or node breaks down, the entire network will not work anymore
Complex hardware is required
Adding a new client disrupts the entire network
Below is the example of ring topology.
Mesh topology also known as fully connected topology connects directly to every machine in the network via multiple hops and they are generally not mobile. Mesh network are self healing.
The node can still operate when one node breaks down or connection goes bad. As a result, the network may typically look very reliable, as there is often more path between a source and destination in the network. Although mostly are used in wireless set-up, this concept can be used in wired network and software interaction. The number of connection are determined by
n(n-1)/2 where n is the number of machines in the network. This type of topology is not cost effective.
The advantages of mesh topology are there are machines that have a dedicated connection to other machines. This makes it reliable and the security is higher. If one of the link breaks down, the network will not affected and will continue to function.
The major disadvantage of mesh topology is that it uses too many connections and therefore it requires a lot of wiring. For example, if the network has 50 machines, machine number 1 would require 49 network connections to connect it to machine 2 through 49.
Below is the example of mesh topology.
From this question, I've learn that network topologies are very important to every organization that is intended to set up a network. This is because it helps them to decide which is better or more suitable for their plan and which is more cost effective, reliable and also easier to maintain.
Without network topologies, we will have no idea what kind of network we are running on and we will not figure out how to set up a better network to save more cost and have a better performance network. Based on the question, to set up a small networking office, the most suitable network topology to be used is star topology. This is because it only needs one hub to connect all the machines together. It is cost effective, easier to set-up and even expands if needed to. Other than that, if one link breaks down, it will not affect the entire network. Capacity can be increased easily and easier to remove parts and detect faults too.
Data link control is to provide functions like flow control, error detection and error control.
These functions are used to make the communication better in-between a network.
Data link control is a layer above the physical interfacing, with the transmission medium
referred as data link.
Flow control is one of the functions provided in data link control. Its main purpose is to enable a
receiver to regulate data flow from sender to avoid buffer overflow; sender may not send data faster than receiver can absorb or download.
Error detection is performed by receiver to check for error in receive code and to recover actual bits.
Error control is performed by sender to retransmit damaged frames that is not acknowledged by receiver.
Flow control is a technique for assuring the sending entity does not overwhelm the receiving entity which means is to prevent buffer overflow. The receiving entity typically allocates a data buffer of some maximum length for a transfer. When data are received, the receiver must do a certain amount of processing before passing the data to higher level software. If flow control is absence, the receiver's buffer will fill up and over flow while it is processing its old data.
The figure above shows a flow control in the absence of error.
Each arrow represents a single frame transiting a data link between two stations. The data are sent in sequence of frames, with each frame containing a portion of the data and some control information. The time it takes to emit all bits of a frame onto the medium is the transmission time. This is proportional to the length of the frame. The propagation time is the time it takes for a bit to traverse the link between source and destination. In this section, all frames are successfully received and no frames are lost.
Stop and wait flow control
This is the simplest form of flow control. The source transmit frame and then the destination receive frame and replies with acknowledgement. The source must wait until it receives the acknowledgement before sending the next frame. The destination can thus stop the flow of data simply by withholding acknowledgement. It works well for a few large frames however it is often that a source breaks up a large block of data into smaller ones and transmits it in many frames. This is done because of limited buffer size, error detected sooner when whole frame are received. On error, retransmission of smaller frames is needed and to prevent one station occupying medium for too long periods.
Stop-and-wait is not very good for multiple frames for a single message.
That is because only 1 frame at a time can be transit
Sliding windows flow control
Allows multiple frames in transit and receiver has buffer W long.
The transmitter can send up to W frames without acknowledgement and each frame is numbered.
Acknowledgement includes number of next frame expected and the sequence number are bounded by size of field (k).
Figure 2.3 and 2.4 is example of sliding window.
Return not ready (RNR) allows receiver to acknowledge frames without indication that it is not yet ready to accept more frames. At some point, receiver must send a Return ready (RR) acknowledgement to allow transmission of more frames.
Errors occur due to change in one or more bits during transmission.
Error detection is added by transmitter for error detection code.
The value of parity bit is such that character has even or odd parity numbers of ones.
Even number of bit errors goes undetected.
Cyclic Redundancy Check
For a block of k bits transmitter generates n bit sequence.
Transmit k+n bits which is exactly divisible by some number
Receive divides frame by that number where if no remainder = no error.
Error control is a mechanism to detect error and correction of errors.
Two types of errors during transmission is
Lost frames means that frames did not arrive at receiver end and possibly address damaged due to noise.
Damaged frames means that some bits in the received frame are altered during transmission.
Automatic repeat request (ARR)
This is a technique error control.
Error detection is one of them including positive acknowledgement, retransmission after timeout and negative acknowledgement and retransmission.
Stop and wait ARQ
This is a very simple process. The source will transmit single frame and then wait for ACK.
If received frame are damaged, it will be discarded. Transmitter will have timeout, if no ACK within timeout, retransmit will happen.
If ACK is damaged, transmitter will not recognize it.
Go-back N ARQ
This is based on sliding window. If no error, ACK as usual with next frame expected.
It used window to control number of outstanding frame. If there's error, it will reply with rejection. Receive discards that frame and future frame till frame is correctly received.
Received frame is erroneous as determined by receiver
Receiver discards the frame, leading to two possibilities
(a) Sender sends next frame
Receiver receives next frame out of order and sends an REJ for previous (damaged frame)
Sender retransmits damaged frame and all subsequent frames
(b) Sender does not send additional frames right away
Receiver receives nothing and does not return either RR or REJ
Upon timeout, sender transmits an rr frame with a P bit set to 1
Receiver will interprets the RR frame with a P bit of 1 as a command to be acknowledged by
sending an RR, with the next frame expected
When sender receives the RR, it retransmits the lost frame
Damaged RR has two subcases
(a) Receiver receives frame and sends an rr which is lost in transit
It is possible that a subsequent acknowledgement arrives before the timeout at sender, causing no problem
(b) Sender's timeout expires
Sender transmits an rr frame with P bit as 1 A P-bit timer is set
If receiver fails to respond, the P-bit timer expires
Sender issues a new rr and resets the P-bit timer
The above procedure is repeated a number of times, and ¬nally, the sender initiates a reset procedure
3. Damaged rej
Handled as case 1b
• Selective-Reject arq
Only those frames are retransmitted that receive a negative acknowledgement, or that time out
SREJ is sent when a frame is received out of order
Receiver continues to accept incoming frames and bu¬€ers them till the lost frame is received
Receiver must maintain a large enough bu¬€er to save post-SREJ frames
Transmitter also requires more complex logic, limiting the use of this technique
In a nutshell, data link control plays a big part in our daily network communication and data transfer to ensure that it is always error free and whenever there's error, they will be able to fix it.