✔ Scroll down and test yourself — answers are hidden under the “View Answer” button.
Attempt all questions first.
✔️ Click SUBMIT at the end to unlock VIEW ANSWER buttons.
A typical core Internet router must have:
- Multiple interfaces with IP addresses
- A switching fabric
- Buffers for packet queuing
- Scheduling algorithms for packet transmission
Why all options are correct?
(a) Multiple network interfaces, each with its own IP address
A core router connects to many different networks or links (often high-speed fiber links). Each network interface (port) operates as a separate network endpoint and therefore must have its own IP address. This allows the router to correctly send, receive, and route packets on each connected link.
Hence, this statement is true.
(b) A switching fabric between its input ports and output ports
Inside a router, packets arriving on input ports must be forwarded to the correct output ports. This internal data path is called the switching fabric.
It may be implemented using memory, a shared bus, or a crossbar switch. The switching fabric enables high-speed packet transfer inside the router.
This is an essential internal component, so the statement is true.
(c) Buffers to hold a queue of packets at input or output ports
Packets can arrive faster than they can be transmitted, especially during periods of congestion. To handle this, routers maintain buffers (queues) at:
- Input ports (to wait for access to the switching fabric)
- Output ports (to wait for link transmission)
Without buffers, packets would be dropped immediately during traffic bursts.
Therefore, this statement is true.
(d) A scheduling algorithm to determine which queued packet goes next
When multiple packets are waiting in a queue, the router needs a rule to decide which packet is sent next. This is handled using scheduling algorithms.
Common scheduling algorithms include:
- FIFO (First-In-First-Out)
- Priority Scheduling
- Round Robin
- Weighted Fair Queuing (WFQ)
These algorithms help manage fairness, delay, and Quality of Service (QoS).
Hence, this statement is true.
The maximum number of connections that can be ongoing in the network at any one time is 48
Given Network
The circuit-switched network consists of four switches A, B, C, and D connected in a square topology.
The number of circuits on each link is: A ↔ B has 14 circuits, B ↔ C has 10 circuits, C ↔ D has 13 circuits, and D ↔ A has 11 circuits. Each circuit can support only one connection at a time.
Key Idea
In a circuit-switched network, each connection occupies one circuit on every link it uses. To maximize the total number of simultaneous connections, the network should favor single-link connections instead of multi-hop paths. This ensures that circuits are not wasted on multiple links for the same connection.
Maximum Number of Ongoing Connections
Since all links can be used independently, the number of ongoing connections equals the total number of available circuits:
A–B supports 14 connections, B–C supports 10 connections, C–D supports 13 connections, and D–A supports 11 connections.
Maximum connections = 14 + 10 + 13 + 11 = 48
Explanation:
Packet jitter refers to the variation in packet delay. The main cause is variable queuing delay at routers and switches, which occurs when network congestion changes over time. Different packets experience different waiting times, leading to jitter.
Other causes of packet jitter can be due to cross traffic, bursty behavior of the end-user or end-host, etc.
Layering is used in computer networks because it simplifies design, promotes interoperability, and enables reuse of protocol implementations.
Explanation: Why layering is used?
Layering divides the complex task of network communication into separate, well-defined layers, where each layer has a specific responsibility (e.g., physical transmission, routing, transport reliability). This design brings major software engineering benefits such as:
- Each layer can be developed independently
- Implementations can be reused across different systems
- Changes in one layer do not break other layers
- Standard interfaces promote interoperability
Ethernet switches learn MAC addresses by observing the source address of incoming frames, while hubs do no learning at all.
Explanation:
In Ethernet networks, switches (also called bridges) maintain a MAC address table (forwarding table).
This table maps: MAC address to switch port.
The switch learns this mapping by observing the source MAC address of incoming frames and recording which port they arrived on.
Why the Other Options Are Incorrect
(a) Ethernet switches do not learn addresses from the destination MAC address. The destination address is used only to decide where a frame should be forwarded. Learning requires knowing where a device is located, which can be determined reliably only from the source MAC address.
(b) Ethernet hubs and repeaters operate at the physical layer and do not inspect frame headers. They do not maintain MAC address tables and simply forward incoming signals to all ports. Therefore, they are incapable of learning addresses.
(c) A correctly functioning Ethernet switch may send a frame to multiple ports when the destination address is unknown or when the frame is a broadcast or multicast. This behavior is normal and does not indicate incorrect operation.
No comments:
Post a Comment