Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

CMPG 315 > Chapter 5: Network layer > Flashcards

Chapter 5: Network layer Flashcards

1
Q

The network is highest…

A

highest layer that runs on network specific hardware, everything afterwards is software.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What the previous networks dont have but really need…

A

A way to establish communications between hosts that arent directly connected to one another, but connected by virtue of intermediaries. Speed should be as optimal as possible

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Store-and-Foreward packet switching

A

In order to transfer packets from one host to another via intermediaries, a technique known as store-and-forward packet switching is used.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How store-and-forward works

A

each of the routers between the source and the destination will fully buffer the packets they receive before sending them to the next node in the path.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

One of the most important functions of the network layer

A

Identification of the optimal route. Routing algorithms are used.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Connectionless network

A

Sends data in a fire-and-forget method. No attempt is made to check if the information is received correctly or at all.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Connection orientated

A

Establish a communication channel before any actual information is transmitted. The goal is usually to ensure that the information is received and received correctly.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Routing algorithms with connectionless network

A

Send all the packets individually using individual paths. But the route will change with every packet, making it more difficult to implement congestion control or quality of life (qos) features.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Datagram network

A

Connectionless network, where packets are called datagrams.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Virtual circuit (VC)

A

connection oriented, and gets its name from the fact that a “circuit” connecting the communicating hosts is created before the data is sent.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Network differences:
Circuit setup

A

Datagram: not needed

VC: Required

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Addressing differences

A

Datagram: each packet contains the full source and destination address

VC: each packet contains a short VC number

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

State information differences

A

Datagram: routers do not hold state information about connections

VC: each VC requires router table space per connection

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Routing differences

A

Datagram: each packet is routed independently

VC: Route chosen when VC is set up; all packets follow it.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Effect of router failure (differences)

A

Datagram: None, except that packets are lost
VC: All VC’s that pass through the failed router are terminated

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Quality of Service (differences)

A

Datagram: difficult
VC: Easy if enough resources can be allocated in advance to each VC

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Congestion control differences

A

Datagram: difficult

VC: Easy if enough resources can be allocated in advance for each VC

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

How routers store information needed to calculate routes

A

Each server has a routing table, which contains information about other routers in the network.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Important example for studying

A

So topological routing tables explored on efundi page!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Optimality principle

A

If the path from router A to router K is optimal, and router F is on that path, then the optimal path from router F to K will follow that path.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Sink tree

A

When optimality principle is applied to a network graph, you can create a sink tree that only contains optimal paths.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Sink tree example

A

SEE AND STUDY THE EFUNDI EXAMPLE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Dijkstra’s Alogirthm (shortest path)

A

By describing each connection as a path with a known length, you can identify the optimal path from any source to any destination by simply minimising the length of the path that a packet would need to travel.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

How Dijkstra’s algorithm works

A

Uses labelling to identify the shortest route. As longer routes are used, labels keep track of the shortest path.

Simply selecting all the nodes with the shortest paths will yield the shortest route.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
SEE DIJKSTRAS EXAMPLE ON EFUNDI
PLS
26
Flaw of shortest path algorithms
It requires full knowledge of the network topology in order to be used effectively. The router must have information above what other routers have. Therefore, often times these algorithms are paired with others.
27
Flooding
The router sends all of its packets to all its neighbours simultaneously. If all the routers repeat this process with the same packet, it should reach its destination.
28
Problems with flooding
Routers can sometimes flood the network with an infinite number of identical packets (can be controlled and managed) Wastes a lot of bandwidth, making it impractical
29
Why use flooding then?
Effective way of sharing a message that each router must receive. And also, it allows for a solution to the critical problem of the shortest path algorithm to form, making it feasible
30
Distance vector routing (Bellman-Ford)
Maintains a table (also known as vector) that contains the shortest distance to every known router in the network, as well as the neighbour that traffic needs to be sent to in order to reach that node.
31
Distance vector routing implications
Each router knows how far away a destination is, and to which node the packet needs to be sent to first.
32
Updates to distance vector routing implications
This table is regularly updated and involves each of the routers in the network sharing their tables with one another. They use the other's table information to update their own information, and account for any deviations.
33
Why it was abandoned in its application on the internet.
It lacks a method of seeing if a router goes offline, which has severe consequences Routers don't see an offline router as offline, but rather faulty. Also, depending on what order the routers received their information, neighbours to offline routers may not see it as offline but rather just far away.
34
Count-to-infinity problem
When routers continually increases their distances between neighbouring offline routers due to misunderstanding of the routers not knowing one is offline.
35
Link-state routing
At regular intervals, each router will: 1. Discover its neighbours and learn their network addresses. 2. Set the distance or cost metric to each of its neighbours 3. Construct a packet telling all it has just learned. 4. Send this packet to and receive packets from all other routers 5. Compute the shortest path to every other router.
36
Link-state routing is used...
Used to route internet traffic today, and requires routers to have greater amount of memory and computational power than what would be needed for distance vector routing, as shortest path will always be computed using full network topology, but because it is constantly updated it wont have a count-to-infinity error.
37
Most used link-state routing protocols
IS-IS (intermediate system- intermediate system) OPSF (open shortest path first)
38
Both IS-IS and OPSF use
Dijkstra's algorithm to determine the shortest path to each router.
39
Both OSPF and IS-IS are classified as
Internal Gateway protocols (IGP), which means that they function within a single autonomous system.
40
Biggest difference between IS-IS and OSPF
IS-IS can carry information about multiple network layer protocols simultaneously, whereas OSPF cant
41
Reason for the difference between IS-IS and OSPF (IS-IS side)
IS-IS is so low in the network layer that it works directly with the data-link layer. Works regardless of protocols used and is address independent. It is also safe from IP attacks, which OSPF is vulnerable to.
42
Reason for the difference between IS-IS and OSPF (OSPF side)
OSPF makes use of the network layer protocols to exchange data. It cannot be used with network protocols that do not support it. It also needs a rigid network structure with central core, and is address dependent.
43
Advantages of OSPF over IS-IS
OSPF, while harder to set up and maintain, it is harder to make mistakes using it. It also offers support for virtual links, and is more adaptable to corporate network environments.
44
Other routing Alogirhtms
Hierarchical Broadcast Multicast Anycast
45
Hierarchical
This functions by dividing the network into hierarchical regions, and then routing traffic between the regions. Routers only need to worry about routers within its own region. All traffic that needs to go to a different region is handled by a designated server.
46
Hierarchical continued
Other regions can be made larger without installing more routers. And routers don't need to know everything about other routers.
47
Downside of hierarchical
Path lengths are increased, because when routers need to send data to other regions, they dont use shortest path but instead just send it to the designated server.
48
Broadcast
Focus on minimising bandwidth usage when a router needs to send a packet to all the hosts on the network.
49
Broadcast method of achieving its goal
Reverse path forwarding. The router only accepts packets from a router if the packet is received via the route that the receiving router usually uses to send packets to the sending router.
50
Broadcast reason for using that method
Because of the optimality principle, if the path from A to B is shortest, then the path from B to A will probably also be the shortest. All packets not received on the reverse route are probably duplicates and are discarded.
51
Multicast
When a packet needs to be sent to multiple (but not all) hosts in the network, then only the correct hosts receive the packet.
52
How multicast can be achieved
designate the subset of routers to a group using some method, then all hosts not in that group destroy the message. Then only the correct hosts receive the packet.
53
Anycast
Concerned with delivering a packet to whichever host is closest. Because it is simply delivering to the nearest host, distance vector or link-state routing can be used to find anycast route.
54
IPv4 protocol and Addressing scheme
Two parts of an IPv4 protocol address: First part identifies the subnet (discrete section of network) 2nd part identifies the specific host within that network.
55
How IPv4 distinguishes between the 2 parts of the name
It makes use of a subnet mask. In binary, it consists of a continuous set of 1s followed by a continuous set of 0s. The mask also uses the amount of digits. The number of 0s indicate the amount of hosts in a network.
56
Maximum number of addresses
2^x, where x is the number of 0s.
57
Length of IPv4 addresses
32 bits in length
58
Most common way to use IPv4 addresses
In dotted decimal form. The 32 bits is divided in the 4 groups of 8 bits, called octets. Each octet is then converted to decimal
59
All IPv4 addresses, in decimal form...
Have four values, each between 0 and 255, and the values are separated using dots.
60
Subnet masks further details
They are crucial in determining the separation between the network and host section of the address.
61
Two approaches to determine these subnet masks
Classful addressing, and classless addressing
62
Classful addressing
The subnet mask is determined by the class of the network. The class also determines the range of values that the first octer i the IP address may have. So only the IP address is needed to determine the subnet mask.
63
Classless network
The network does not belong a type of class, so it is more flexible to the number of addresses required. But it does mean that classless requires that information about the subnet mask is provided
64
Two ways subnet mask information can be provided.
It can be provided directly in binary With an additional value that indicates the number of bits in the prefix. This value is separated by means of a forward slash.

CMPG 315 (6 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions