Review Questions

1.	What is one advantage of using switches instead of hubs at a customer location? Each port in a switch creates its own broadcast domain. Each port in a switch creates its own collision domain. Each port in a switch creates its own broadcast and collision domains. It depends on the type of servers.
2.	Your customer asks you why they should upgrade to Fast Ethernet. After you tell the customer it is 10 times faster, they ask what are the disadvantages of upgrading. What do you tell them? Installation difficulty. Expense. The physical distance. There are no disadvantages.
3.	You have excessive broadcasts on your customer’s LANs. What solutions would benefit the customer? (Choose all that apply.) Adding more hubs Adding Layer 3 switching Creating VLANs Using more routers
4.	Which of the following is a characteristic of a switch, but not of an active hub? Switches forward packets based on the IPX or IP address in the frame. Switches forward packets based only on the IP address in the packet. Switches forward packets based on the IP address in the frame. Switches forward packets based on the MAC address in the frame.
5.	How does the cut-through switching technique work? The LAN switch copies the entire frame into its on-board buffers and then looks up the destination address in its forwarding (switching) table and determines the outgoing interface. The switch waits only for the header to be received before it checks the destination address and starts forwarding the packets. Cut-through switching uses broadcast addresses as source addresses. Cut-through switching uses a Class II repeater in a collision domain.
6.	How do switches use the store-and-forward switching technique? The switch waits only for the header to be received before it checks the destination address and starts forwarding the packets. The LAN switch copies the entire frame into its on-board buffers and then looks up the destination address in its forwarding (switching) table and determines the outgoing interface. The switch uses a Class II repeater in a collision domain. The switch uses broadcast addresses as source addresses.
7.	Which of the following are needed to support full-duplex Ethernet? (Choose all that apply.) Multiple paths between multiple stations on a link Full-duplex NICs Loopback and collision detection disabled Automatic detection of full-duplex operation by all connected stations
8.	Which of the following are not used by full-duplex Ethernet? (Choose all that apply.) Collision detection MAC addressing 10-megabit transmission Loopback
9.	You are creating VLANs at your customer location. The manager comes by to ask why and reminds you of your statement that spending money on switches would solve all their problems. What do you tell her? (Choose all that apply.) You need to break up the collision domains. You need to break up the broadcast domains. VLANs help to extend the time-to-live in the IP header. Security between departments can be accomplished with VLANs since no traffic leaves a VLAN unless it is first routed.
10.	Which of the following are among the advantages of creating VLANs? (Choose all that apply.) Datagram filtering Breaking up broadcast domains Creating networks by function rather than by location Long time-to-live packet control when using IP
11.	Which two of the following statements describe frame tagging? Frame tagging examines particular info about each frame. An ID is placed in the header of each frame as it traverses the switch fabric. A user-assigned ID is added to each frame. Filter tables are built as part of frame tagging.
12.	Which of the following statements describes a full-duplex transmission? Full-duplex transmission uses a single cable. Full-duplex transmission uses a point-to-point connection from the transmitter of the transmitting station to the receiver of the receiving station. Full-duplex transmission involves data transmission in both directions, but only one direction at a time. Full-duplex transmission involves data transmission in only one direction.
13.	If a frame is received at a switch and only the destination hardware address is read before the frame is forwarded, what type of switching method are you using? Cut-through Store-and-forward Store-and-cut FragmentFree
14.	Which two Layer 2 features are used on Cisco switches to optimize multicast? OSPF Multicast CGMP Multicast IP forwarding IGMP snooping
15.	Which of the following statements is true for the store-and-forward switching method? It is the default switching method for all Cisco switches. It only reads the destination hardware address before forwarding the frame. Latency varies depending on frame length. Latency is constant.
16.	What does the spanning-tree algorithm (STA) do? STA is implemented by STP to prevent loops. STA forwards packets through a switch. STA restores lost frames. STA prevents API duplication in bridged networks.
17.	What are three advantages of a Layer 3 backbone over a Layer 2 backbone? Reducing the number of connections between L3 switches (peering) Multiple spanning-tree instances Scalability to a very large size Multicast and broadcast control on the backbone
18.	What is the IEEE specification for the Spanning-Tree Protocol (STP)? 802.2u 802.3q 802.1d 802.6
19.	A Layer 3 switch transmits data outside the source network by using which of the following? Bridging tables EIGRP tables Route processors RIP
20.	Of the three switching types, which one has the highest latency? Cut-through FragmentFree Store-and-forward Store-and-cut

Answers

1.	B Individual collision domains per switch port provide for a more efficient network design. Hubs create a flat network design and can be very inefficient since workstations must share the bandwidth.
2.	C The timing in Fast Ethernet is shorter (10 times faster than 10Mbps Ethernet). The maximum frame size (time slot) is 1,518 bytes. The physical distance is reduced because both Fast Ethernet and regular Ethernet specifications state that the round-trip time must not exceed 512 bit times. Since Fast Ethernet transmits faster, a signal of 512 bits covers a shorter distance.
3.	B, C By using a combination of Layer 3 switches and VLANs, you can break up broadcast domains and create more bandwidth where needed.
4.	D A switch forwards packets based on the MAC address of the frame; an active hub only repeats signals.
5.	B After the first 64 bytes are received and checked for fragments, the frame is forwarded to the destination hardware address.
6.	B Store-and-forward switching has the highest latency but provides for the most error-free network design.
7.	B, C Most NICs support full-duplex configuration, which disables loopback and collision detection.
8.	A, D When an NIC is configured for full-duplex Ethernet, the collision detection and loopback are disabled.
9.	B, D VLANs are a configuration option available in switches, and they allow further control of network traffic.
10.	B, C When VLANs are used to control broadcast traffic, the need for routers is reduced. VLANs also serve the purpose of controlling traffic regardless of the physical port connection in the switch fabric.
11.	B, C Two methods of identifying a frame as it moves throughout the switch fabric are adding a VLAN ID to each frame and adding a user- assigned ID to each frame.
12.	B Using multiple parallel lines, full-duplex transmission allows for simultaneous transmit and receive paths.
13.	A Cut-through switching is the fastest of the switching types but is more likely to forward fragmented or damaged frames.
14.	B, D One problem is that the default for an L2 switch is to forward all multicast traffic to every port that belongs to the same VLAN on the switch. The commonly used solutions are Cisco Group Management Protocol (CGMP) and IGMP snooping.
15.	C Although store-and-forward has the highest latency of the switching methods, it checks the entire contents of each frame, thus providing for error-free delivery.
16.	A The spanning-tree algorithm calculates the best path from each bridge or switch port to a root bridge or switch and provides for loop- free networks.
17.	A, C, D By using an L3 switching backbone over an L2 backbone you can reduce the number of connections between backbone switches; this process is called peering. Since it is Layer 3, STP is not used. In addition, you can scale to a very large size and provide broadcast control, which is not possible in an L2 backbone without creating multiple VLANs.
18.	C 802.1d is the IEEE specification for STP.
19.	C A Layer 3 switch uses a route processor to determine how a packet is forwarded outside the source network. A Layer 2 switch uses a MAC address table to create forward/filter decisions.
20.	C Store-and-forward is the slowest of the switching methods, but it provides for error detection.