Comp 247 Data Communications Tutorial Answers Week Ten

Question 1 Compare and contrast circuit switched services, dedicated circuit services, and packet switched services. (Core Topic) ANSWER For both circuit switched and dedicated circuit networks, a circuit is established between the two communicating computers. This circuit provides a guaranteed data transmission capability that was available for use by only those two computers and is assigned solely to that transmission. No other transmission is possible until the circuit is closed. In contrast, packet switched services enable multiple connections to exist simultaneously between computers over the same physical circuit or even over different physical circuits. In the POTS and ISDN circuit switched networks each connection goes through the regular telephone network on a different circuit: These circuits may vary in quality, meaning that while one connection will be fairly clear, the next call may be noisy. The data transmission rates on these circuits are usually low. Generally speaking, transmission rates range from 28.8 Kbps to 56 Kbps for dialed POTS circuits to 128 Kbps to 1.5 Mbps for ISDN circuits. You usually pay per use for circuit switched services. With a dedicated circuit network, you lease circuits from common carriers for their exclusive use twenty-four hours per day, seven days per week. All connections are point to point, from one building in one city to another building in the same or a different city. The carrier installs the circuit connections at the two end points of the circuit and makes the connection between them. The circuits still run through the common carrier's cloud, but the network behaves as if you have your own physical circuits running from one point to another: Dedicated circuits are billed at a flat fee per month and the user has unlimited use of the circuit. Once you sign a contract, making changes can be expensive because it means rewiring the buildings and signing a new contract with the carrier. Therefore, dedicated circuits require more care in network design than switched circuits both in terms of locations and the amount of capacity you purchase. With packet switched services, the user again buys a connection into the common carrier cloud). The user pays a fixed fee for the connection into the network (depending upon the type and capacity of the service) and is charged for the number of packets transmitted. Question 2 In Frame relay, do we need to assign DLCI numbers only between the customer and the provider (DTE-DCE interface) or also for virtual circuits between two DCEs (switches)

inside the network and do they need to be unique in any way? What is the procedure for assigning these DLCIs in a switched virtual circuit scenario? (Core Topic) ANSWER DLCIs are assigned not only to define the virtual circuit between a DTE and DCE, but also to define the virtual circuit between two DCEs (Switches) inside the network. A switch assigns a DLCI to each virtual connection in an interface. This means that two different connections belonging to two different interfaces may have the same DLCIs. In other words, DLCIs are unique only for a particular interface. In a switched virtual circuit scenario, each time a DTE wants to make a connection with another DTE, a new virtual circuit connection should be established. How can this be done? In this case, frame Relay cannot do the job by itself, but needs the services of another protocol that has a network layer and network layer addresses (such as ISDN). The signaling mechanism normally used is Q.931 (ISDN signaling protocol). This mechanism makes a connection request using the network layer addresses of the two DTEs. The local DTE sends a SETUP message to the remote DTE, which responds with a CONNECT message. After the connection phase, the virtual circuit is established so that the two DTE can exchange data. Either DTE can issue a RELEASE message to terminate the connection. Question 3 What do you mean by bursty traffic? Explain the differences between CIR and MAR. ANSWER Bursty traffic refers to an uneven pattern of data transmission: sometime very high data transmission rate while other time it might be very low. Bursty traffic requires what is called bandwidth on demand. The user needs different bandwidth allocations at different times. The committed information rate (CIR) is the data rate the circuit must guarantee to transmit. If the network accepts the connection, it guarantees to provide that level of service. Most connections also specify a maximum allowable rate (MAR), which is the maximum rate that the frame relay network will attempt to provide, over and above the CIR. The circuit will attempt to transmit all packets up to the MAR, but all packets that exceed the CIR are marked as discard eligible (DE). If the network becomes overloaded, DE packets are discarded. So while can transmit faster than the CIR, they do so at a risk of lost packets. Question 4 Compare and contrast X.25, Frame Relay, and ATM networks. ANSWER Frame relay is a newer packet switching technology that transmits data faster than X.25. In many ways, frame relay networks work the same as X.25 networks. Users connect to them by FRADs (frame relay access device) that perform the same function as the PADs. Both datagram and virtual circuit services are available.

Frame relay differs from X.25 and traditional networks in three important ways. First, frame relay operates only at the data link layer. Frame relay, like other packet switched networks takes the incoming packets from the user network and converts them to its own packet structure for internal transmission. However, unlike X.25, frame relay does not replace the user's network layer or data link layer packets with its own. Instead, it encapsulates the entire incoming packet with its own data link layer packet, leaving the user's network and data link layer packets intact. Frame relay uses variable length packets which adapt to the size of the incoming packet (up to 8K), unlike X.25 that uses fixed length packets. Second, frame relay networks do not perform error control. X.25 networks (and virtually all other types of networks) perform error checking at each computer in the network. Any errors in transmission are corrected immediately, so that the network layer and application software can assume error-free transmission. However, this error control is one of the most time consuming processes in a network. Most networks today are relatively error-free, so frame relay networks do not ensure error-free delivery of the packets (they do perform error checking, but simply discard packets with errors; they do not generate NAKs and ask for retransmission). It is up to the software at the source and destination to perform error correction and to control for lost messages. Since the user's data link packet remains intact, it is simple for the devices at the edge of the frame relay network to check the error control information in the user's data link layer packet to ensure that no errors have occurred and to request transmission of damaged or lost packets. A third major difference is that frame relay defines two connection data rates that are negotiated per connection and for each virtual circuit as it is established. The committed information rate (CIR) is the data rate the circuit must guarantee to transmit. If the network accepts the connection, it guarantees to provide that level of service. Most connections also specify a maximum allowable rate (MAR), which is the maximum rate that the frame relay network will attempt to provide, over and above the CIR. The circuit will attempt to transmit all packets up to the MAR, but all packets that exceed the CIR are marked as discard eligible (DE). If the network becomes overloaded, DE packets are discarded. So while can transmit faster than the CIR, they do so at a risk of lost packets. ATM is similar to frame relay. All data are packet-switched, and there is no error control at the intermediate computers within the network; error control is the responsibility of the source and destination (ATM is considered an unreliable packet service). CIR and MAR (which ATM calls available burst rate (ABR)) can be negotiated when circuits are established. Unlike Frame Relay, ATM uses fixed length packets, offers QoS based services and is scalable (it is easy to multiplex basic ATM circuits into much faster ATM circuits). Question 5 Refer to the lecture slide number 88. At the sending side, look at the encapsulation process at the employees router VPN device. Answer the following questions: (Core Topic) a) Is this not a flaw since the encapsulating protocol needs to be either at the same layer or a layer below the encapsulated protocol and not in the layer above it (as is in this case)? b) Why is there a need to encapsulate original IP packet using a secure ESP header within an UDP packet and then within another IP packet? Why cannot we simply encapsulate the original IP packet within another IP? c) Alternatively, since original IP packet is secured using ESP, cant we simply encapsulate it within a DL frame (in this case: PPP) and push it into the network?

ANSWER a) We do this to handle encryption. The whole IP packet (IP+ TCP + HTTP) is encrypted and is treated like an application layer packet. This packet can only be decrypted by the VPN gateway. This packet gets passed to the transport layer (UDP) and then to the IP layer again for further encapsulation. This packets destination address is the VPN gateway address and the source address of the employees computers normal Internet interface. b) The reason being that the VPN software treats the encrypted packet as an application layer packet. c) You would do this in a normal communication scenario where VPNs are not involved. Since the encrypted packet is treated like data from the application layer it undergoes standard transport, network and then the data link layer encapsulations in sequence. Question 6 How does circuit switching technique differs from packet switching based virtual circuit technique? (Core Topic) ANSWER Although it seems that a circuit switched connection and a virtual circuit connection are the same, there are differences: Path versus Route: A circuit switched connection creates a path between two points. The physical path is created by setting the switches for the duration of the (dial up line) or the duration of the lease (leased line). A virtual circuit connection creates a route between two points. This means each switch creates an entry in its routing table for the duration of the session. (SVBC) or duration of the lease (PVC). Whenever, the switch receives a packet belonging to a virtual connection, it checks the table for the corresponding entry and routes the packet out of one of its interfaces. So in a nutshell, in circuit switching all switches close in such a way to create a path between two points A and B whereas in virtual circuit approach, all switches create a route entry in a such way to create a route for this connection. Dedicated versus Sharing: In a circuit switched connection, the links that make a path are dedicated, they cannot be used by other connections. In a VC connection, the links that make a route can be shared (using multiplexing) by other connections. So in a nutshell, in circuit switching, no multiplexing is done at the switches. However, in the virtual circuit approach we can have multiplexing at the switches.

Question 7 Explain the concept of a virtual circuit (Virtual channel, Virtual path) in an ATM network. ANSWER

ATM defines a virtual circuit between each sender and receiver, and all packets use the virtual circuit identifier as the address. Each virtual circuit identifier has two parts, a path number and a circuit number within that path. Each ATM switch contains a circuit table that lists all virtual circuits known to that switch (analogous to a routing table in TCP/IP). Because there are potentially thousands of virtual circuits and because each switch knows only those virtual circuits in its circuit table, a given virtual circuit identifier is used only between one switch and the next. A VP provides a connection or a set of connections between two switches. A VC is a part of VP such that all cells belonging to the single message follow the same VC and remain in their original order until they reach their destination. AVP identifier in the cell makes the switching process a bit more efficient.