Connection-orientated Service
Jan 18, 2018
Connection-orientated Service
Connection in data networks
• Another type of network, such as the telephone network, was developed based on the connection idea.– Before communication, a connection must be set
up, which consists of physical circuits.– After the communication is over, tear down the
connection.
Advantage
• The resources in the connection is reserved, so interference-free from other connections. Can guarantee certain Quality of Service requirements.
• The connection (the physical circuits in the routers) has been set up, the processing of data packets is minimal. No need to check the header for full destination address. The packet does not even have to carry the address information.
Disadvantage
• People may reserve more resources than they need. Such resources, even when idle, cannot be used by others. A waste.
• The set up of the connection requires time. For a large network, the delay can be 100ms. Okay for telephone networks, not oaky for some data services which just need to send a single packet.
Virtual Circuit• Implementing circuit in a datagram network?
– Datagram network forwards packets, there is no physical circuits that can be set up for connections.
– Still, if we set up a virtual circuit, the routers will know how to route the packets of this virtual circuit. To reduce the complexity of routers.
– Before transmission, a virtual circuit is set up. Then, a packet will carry the ID of the virtual circuit. The virtual circuit ID is much smaller, and thus much easier to deal with, than the full destination address. It is also much smaller than the full destination address, can save some overhead.
Virtual Circuit• When setting up the
virtual circuit, a VC ID is picked. The router knows where to forward a packet with a certain VC ID.
• A practical problem in a distributed environment – different stations may pick the same VC ID. – Labels can be
swapped without causing confusion.
B
E
F
D
CA
H1
H2
A’s TableIn OutH1, 1 C, 1H2, 1 C, 2
H3
C’s TableIn OutA, 1 E, 1A, 2 D, 1
Exercise
• Find the shortest path from A to all other nodes in the network.
Exercise• Suppose a network runs the distance vector
algorithm. Suppose node A is adjacent to B, C, and D, with link distance of 5, 2, and 9, respectively. A receives advertisements from B, C, and D about node E. From B: 5. From C: 10. From D: 3. So to reach E, A will pick which node as the next hop and the distance isa. B, 12b. C, 10c. D, 12d. None of the above.
Exercise• A router has the following (CIDR) entries in its routing table:
Address/mask: Next hop 160.36.224.0/19: Interface 0 160.36.240.0/20: Interface 1 160.36.232.0/21: Router 1 Default: Router 2 For a packet with destination address of 160.36.243.15, the router will forward it toa. Interface 0b. Interface 1c. Router 1d. Router 2
Exercise
• A router just received the following new IP addresses: 57.6.40.0/21, 57.6.60.0/22, 57.6.36.0/22, all with the same outgoing line. They a. can be aggregated to 57.6.32.0/19.b. can be aggregated to 57.6.48.0/19c. cannot be aggregated.d. None of the above.
Exercise
• Which of the following statements is true?a. BGP is for routing between the ASes, and suffers
from the count-to-infinity problem.b. OSPF is for routing within ASes, and is a distance
vector algorithm.c. ARP is used when a node wants to know the IP
address of itself.d. None of the above,
Exercise
• Suppose an input buffered switch has 4 inputs/outputs. If the buffer states at input 0 to 3 are (0,2,0,0), (1,2,3,0),(0,0,2,2), and (0,0,1,0), respectively, what is the maximum number of packets that can be sent in this time slot?a. 2.b. 3.c. 4.d. None of the above.
Exercise• Line FG goes down and
suppose we want to find path from F to D.
• What content will node F have in its routing table when DV is used? What will it do in the first step?
• What content will node F have in its routing table when Path Vector is used? What will it do in the first step?
AG
F
EI J
H
DC
B
4 1
1
1
2
1
2
6
6
6
6
6
6
6
Exercise• Line FG goes down.• What content will node F have in
its routing table when DV is used? What will it do in the first step?– B: 5– E: 4– I: 5
• What content will node F have in its routing table when Path Vector is used? What will it do in the first step?– B: 5, BCD– E: 4, EFGCD– I: 5, IFGCD
AG
F
EI J
H
DC
B
4 1
1
1
2
1
2
6
6
6
6
6
6
6