The -------------term refers to the general concept of a small block of data.
- APackets
- BData
- CFrame
- DNone of given
Packet is the generic term for a small block of data sent across a network.
Questions from past midterm papers.Pichhle midterm papers ke sawal.
132 MCQs · 61 short · 26 long questions · 132 MCQs from past papers132 MCQs · 61 short · 26 long sawal · 132 MCQs past papers se
The -------------term refers to the general concept of a small block of data.
Packet is the generic term for a small block of data sent across a network.
CRC can detect more errors than a simple checksum.
A cyclic redundancy check detects more error patterns, including burst errors, than a simple checksum.
___________scheme, which is designed to help detect transmissions errors, send one extra bit of information with each character
A parity scheme adds one extra bit per character.
Computer networks are often called -------------- because they use packet technology.
Networks that use packet technology are called packet (switching) networks.
________________ has advantages arisen from the size and ease of computation.
Checksums are small and easy to compute, which is their chief advantage.
The term ___________ is used to denote the definition of a packet used with a specific type of network.
Frame denotes the packet format defined for a specific network technology.
-------------- have advantages arisen from the size and ease of computation.
Checksums are small and easy to compute.
Computers attached to an ether use ------ in which a computer waits for the ether to be idle before transmitting a frame.
Ethernet stations use CSMA/CD: they sense the carrier and wait for an idle ether before sending.
Computers attached to an ethernet use ------ in which a computer waits for the ether to be idle before transmitting a frame.
Ethernet stations use CSMA/CD and wait for an idle ether before sending.
A ____________ provide a mechanism that a customer can use to set a physical address.
A configurable scheme lets the customer set the interface physical address.
Ethernet uses a ------ bit static addressing scheme in which each device is assigned a unique address by the manufacturer.
Ethernet hardware addresses are 48 bits assigned by the manufacturer.
Ethernet uses a ____________ static addressing scheme in which each device is assigned a unique address by the manufacturer.
Ethernet addresses are 48 bits, fixed by the manufacturer.
The third field of the header consists of ------ bit Ethernet frame type.
The Ethernet frame type field is 16 bits.
An ------------method, the network hardware designers specify how type information is included in the frame and the value use to identify various frame types.
In explicit frame typing the frame carries a type field whose values the hardware designers define.
Formally named --------------- informally known as thin wire ethernet or thin net.
10Base2 is the formal name of thin coaxial (thinnet) Ethernet.
Formally named __________ informally known as the thick wire Ethernet or Thick net.
10Base5 is the formal name of thick wire Ethernet.
Formally named __________ informally known as the twisted pair Ethernet or TP Ethernet.
10Base-T is the formal name of twisted pair Ethernet.
One repeater ---------------, two repeaters --------------- the maximum cable length limitation.
Each repeater adds one more maximum-length segment, so one doubles and two triple the length.
One repeater _______, two repeaters ________ the maximum cable length limitation.
Each repeater adds one more maximum-length segment.
_____________ sense a signal on one cable and then transmits an amplified copy on the other cable.
A repeater amplifies and retransmits signals between cables.
A Bridge can ---------------.
A bridge extends a LAN by connecting segments and filters or forwards each frame.
A Bridge forwards or filters a frame by comparing the information in its address table to the frame's__________
The forwarding decision is made by looking up the frame destination MAC (layer 2) address.
A bridge uses the ___________ to determine which computers connect to which segment, and uses the __________ to determine whether to forward a copy of frame.
A bridge learns from source addresses and forwards based on destination addresses.
A system with redundant bridges might have a problem with_______ in the system.
Redundant bridges can create forwarding loops.
Most WAN systems include a mechanism that can be used to eliminate the common case of duplication routing is called___________
A default route replaces many routing-table entries that share the same next hop with a single entry.
The next hop to which a packet is sent depends only on
Next-hop forwarding looks only at the destination address.
_______ computes shortest paths in a graph by using weights on edges as a measure of distance.
Dijkstra's algorithm finds least-weight paths using edge weights as distances.
Basic LAN technologies such as Ethernet, Token Ring, and FDDI use a _____.
Basic LANs send each frame independently without setting up a connection.
--------------- has a jitter zero
An isochronous network is designed to deliver data with no variation in delay, i.e. zero jitter.
The network with Throughput T and Delay D has a total --------------- bits in transit at a time.
The delay-throughput product D x T gives the number of bits in transit.
The product of delay and throughput measures the _____ of data that can be present on the network.
Delay times throughput gives the volume of data in transit.
A network with throughput T and delay D has a total of ________ bit in transit at any time.
The number of bits in transit is the delay-throughput product T x D.
TTL stands for __________
TTL is Time to Live.
In -------------, network occupies the smaller area like a room a floor or a building
A local area network covers a small area such as a room, floor or building.
------------- was especially concerned about the lack of high powered computers.
ARPA research projects needed access to scarce high-powered computers, which motivated its networking research.
The network that uses a ------------- , usually consist of a single long cable to which computer attach.
In a bus topology all computers attach to one long shared cable.
Local Talk is a LAN technology that employs -------------
Apple's LocalTalk is the handout's example of a bus topology LAN.
LAN that use ATM technology have a -------------
ATM LANs connect each computer to a central switch, giving a star topology.
When an application--------------- data, it makes a copy of the data available to all other computers on the network.
Broadcasting delivers a copy of the data to every computer on the network.
__________is used for compressed audio and video where the data rate depends on the level of compression that can be achieved.
Variable bit rate service suits compressed media whose data rate varies with compression.
ATM assigns each VC a _____________ identifier that is divided two parts to produce a hierarchy.
An ATM virtual circuit is identified by 24 bits split into an 8-bit VPI and a 16-bit VCI.
A typical port on an ATM switch operates at __________ or higher.
ATM switch ports typically run at OC-3, which is 155 Mbps.
Which of the following is a connecting device?
Bridges, repeaters and hubs are all devices used to connect network segments.
A bridges function in the _________ layers(s).
A bridge works on frames, operating in the physical and data link (MAC) layers.
The Fast Ethernet hardware operates at a rate of _______________
Fast Ethernet runs at 100 Mbps.
_____________Program sends a message to a remote computer and reports whether the computer responds.
Ping sends an echo request and reports whether a reply comes back.
In ______________ network occupies larger areas like cities & countries.
A wide area network spans cities and countries.
In Point-to-Point topology there are two topologies.
Point-to-point topologies in the handout are star and tree.
Unlike Frame Relay and ATM, SMDS (Switched multi-megabit Data service) offers_______.
SMDS is a connectionless service, while Frame Relay and ATM are connection-oriented.
An interface for twisted pair Ethernet must have a ____________ connector, and must generate signals according to the_____________ specification.
Twisted pair Ethernet uses RJ-45 connectors and 10Base-T signalling.
Most LANs that employ ring topology use an access mechanism known as-------
Ring LANs control access by passing a token around the ring.
The maximum size of an Ethernet segment is___________
An original Ethernet cable segment is limited to 500 meters.
A network uses a ------ if all computers attach to a central point
In a star topology every computer connects to a central point (hub).
---------- was especially concerned about the lack of high powered computers.
ARPA needed access to scarce high-powered computers for its research projects.
An interface for twisted pair Ethernet must have an ____________ connector , and must generate signals according to the_____________ specification.
Twisted pair Ethernet uses RJ-45 connectors and 10Base-T signalling.
___________ is used for audio and video, since these have predefined maximum data rates
Constant bit rate service suits uncompressed audio and video with fixed maximum rates.
ATM is designed to work on___________.
ATM was designed for optical fiber, though it can use twisted pair.
FDDI can transmits data at a rate of ----------
FDDI transmits at 100 Mbps over fiber.
In broadcast topology there are two categories____.
The handout lists satellite/radio and ring as the two broadcast topologies.
The --------- bit preamble in the Ethernet frame format that precedes the frame contains alternating 1s and 0s that allow the receiver’s hardware to synchronize with the incoming signal.
The Ethernet preamble is 64 bits.
An interface for twisted pair Ethernet must have a ____________ connector, and must generate signals according to the_____________ specification.
Twisted pair Ethernet uses RJ-45 connectors and 10Base-T signalling.
In 3-way handshake TCP requires to generate a random ________ sequence number.
TCP sequence numbers are 32 bits.
TCP stands for _______
TCP is the Transmission Control Protocol.
Twice NAT allows a site to run servers.
Twice NAT works with DNS so that a site behind NAT can run servers.
Preliminary version of IP was called _________.
The preliminary version of IPv6 was called IP Next Generation.
_______ uses window mechanism to control the flow of data.
TCP uses a sliding window for flow control.
Which protocol is used to test different tools.
Tools such as ping and traceroute are built on ICMP.
The Source can configure outgoing datagram’s to avoid _________
Using the path MTU, the source sizes datagrams to avoid fragmentation.
The time for acknowledgement to arrive depends on __________
Acknowledgement delay depends on both distance and current traffic.
which is not the type of error messages defined by ICMP.
Source quench, time exceeded and destination unreachable are all ICMP error messages.
Due to revolutionalization of IP-V6 the speed has increased from _____
The course notes that network speeds grew from 56 Kbps to 1 Gbps.
IGPs stand for _________
IGP means Interior Gateway Protocol.
NAT is not useful at a residence with Cable Modem or DSL connectivity.
NAT is useful at a residence to share one cable or DSL address among several computers.
______ layer Provides reliable delivery of datagram.
The transport layer (TCP) provides reliable delivery.
The process of learning the path MTU is known as path MTU discovery.
Learning the smallest MTU along a path is called path MTU discovery.
End to End delivery Service of IP datagram is _____
IP delivers each datagram independently with a connectionless service.
EGP stands for _________
EGP means Exterior Gateway Protocol.
Reliability is not the responsibility of the Transport layer.
Reliability is a transport layer responsibility (TCP).
IPV6 address with ____ leading zeros is interpreted to hold an IPV4 address.
An IPv6 address with 96 leading zero bits holds an IPv4 address in its last 32 bits.
NAT software does not allow a PC to connect with the Internet and act as a NAT device at the same time.
NAT software lets a PC connect to the Internet and act as a NAT device at the same time.
TCP uses the term segment to refer to a ________.
TCP calls each of its messages a segment.
IP datagram can contains ______ octets.
The 16-bit length field limits an IP datagram to at most 65,535 octets.
The amount of buffer space available at any time is called the window.
TCP's advertised window is the receiver's currently available buffer space.
_____ Source is responsible for fragmentation.
In IPv6 only the source fragments; routers drop datagrams that are too large.
Routers use _____ to forward datagrams along prearranged path.
Routers use the IPv6 flow label to forward datagrams along a prearranged path.
_______ protocol uses three way handshake to begin a connection.
TCP opens connections with a three-way handshake.
Typical internet routing uses uses a combination of two metrics__________________.
Internet routing combines administrative cost with hop count.
Network Address and Port Translation (NAPT) is by far the most popular form of
NAPT is the most popular form of NAT.
The routers within an autonomous system use a _____________ to exchange routing information.
Routers inside one autonomous system exchange routes with an IGP.
Interior Gateway Protocols (IGPs) and Exterior Gateway Protocols (EGPs) two broad classes of Internet Routing Protocol.
Internet routing protocols fall into two classes, IGPs and EGPs.
The Network Layer Protocol ICMP stands for____________
ICMP is the Internet Control Message Protocol.
The Current version of IP-Version 4 is ___ old
The course states IPv4 was about 20 years old.
TCP is a connectionless and reliable transport protocol.
TCP is reliable but connection-oriented.
ATM header is about the ---------- of the cell.
The 5-octet header is about 10% of the 53-octet cell.
Jitter is significance for ------------------
Variance in delay matters for voice, video and data traffic.
How many methods exist for building routing table?
Routing tables are built either statically or dynamically.
Asynchronous Transfer Mode uses the concept of ______________
ATM is connection-oriented.
Which of the following statement is true regarding ATM?
All three statements describe ATM.
A ____ can forward or block messages based on the information in the message itself
A proxy (application-level) firewall inspects message content; a packet filter looks only at headers.
Transmission media are usually categorized as _______.
Media are classed as guided (cables) or unguided (wireless).
Which of the following is not a guided medium?
The atmosphere carries unguided (wireless) signals.
Which multiplexing technique transmits digital signals?
Time division multiplexing interleaves digital data in time slots, while FDM and WDM are analog/optical frequency techniques.
All destinations on same switch have same _________
Destinations attached to the same switch share the same next hop.
An exterior switch is one with which ………………….
An exterior switch has computers attached; an interior switch has none.
Frame relay is………………….
Frame relay is a connection-based Telco service for delivering blocks of data.
An interior switch is one with which ………………….
An interior switch connects only to other switches, with no computers attached.
Which of the following is not a benefit of source independence?
All three listed items are benefits of source independence, so none of them is a non-benefit.
Which of the following statement regarding ATM is true?
ATM connections are virtual channels identified by a 24-bit VPI/VCI value.
How many methods exist for computing routing table?
There are two methods: static routing and dynamic routing.
Which one of the following is a property of static routing?
Static routing is set at boot time, is inflexible and adds little network overhead.
Which multiplexing technique shifts each signal to a different carrier frequency?
FDM places each signal on its own carrier frequency.
LAN networks can not be extended to very large networks because of the limitations of...........
LAN technologies have distance limits set by signal strength and access timing.
Which of the following statement is wrong regarding ATM?
All three statements about ATM are true, so none is wrong.
Which one of the following is a property of dynamic routing?
Dynamic routing updates tables automatically and can route around failures.
Which of the following statement is true regarding ATM?
ATM serves both LAN and WAN goals and is connection-oriented.
Source independence is a phenomena in which next hop to destination does not depend on the …………………
With source independence the next hop is chosen from the destination only, not from the packet's source.
If a computer on the network shares resources for others to use, it is called ____
A computer that shares resources with others is a server.
Terminators are used in ______ topology.
Bus cables need terminators at both ends.
In _____ topology, if a computer’s network cable is broken, whole network goes down.
A break in the shared bus cable stops the whole network.
For large networks, _______ topology is used.
Star (hub/switch based) wiring scales best for large networks.
ISO stands for
ISO is the international standards organization.
ISO OSI model is used in
The OSI model describes communication in a network environment.
Network cable lies on _____ layer
Cables belong to the physical layer.
____ Layer decides which physical pathway the data should take.
The network layer handles routing.
ISDN is an example of ______ network
ISDN sets up dedicated circuits.
X.25 is an example of ______ network
X.25 is a packet-switched network.
_____________ allows LAN users to share computer programs and data.
A file server stores shared programs and data.
Print server uses ________ which is a buffer that holds data before it is send to the printer.
A spool buffers print data until the printer takes it.
A standalone program that has been modified to work on a LAN by including concurrency controls such as file and record locking is an example of____
A standalone program adapted with locking for LAN use is LAN-aware software.
The ______ portion of LAN management software restricts access, records user activities and audit data etc.
Access restriction and auditing are security management functions.
Which connector STP uses?
Twisted pair Ethernet cabling uses RJ-45 connectors.
What is the central device in star topology?
A star topology centers on a hub or switch.
No MCQs match this filter.Is filter mein koi MCQ nahi.
What is the disadvantages of repeater?
A repeater does not understand frames; it only amplifies and retransmits electrical signals. Therefore it also copies collisions, noise and corrupted signals from one segment to the other, it cannot filter traffic (all segments still share one bandwidth and form one collision domain), and only a limited number of repeaters can be placed between two stations because of timing limits.
Why the Connection Oriented Services use the connection identifier instead of full address?
A connection identifier (such as an ATM VPI/VCI) is much shorter than a full destination address, so each cell or packet carries less header overhead, and switches can forward it quickly with a simple table lookup indexed by the identifier. The full address is needed only once, when the connection is set up.
What is the logical and physical topology of 10 Base T and token ring network?
10Base-T: physically a star (each computer is wired to a central hub), but logically a bus, because the hub electronics make all stations share one medium like a single cable. Token ring (IBM style): physically a star (stations are wired to a central hub/MAU), but logically a ring, because the token and frames pass from station to station in a closed loop. Logical topology is set by the network technology; physical topology by the wiring scheme.
Why 10-base T wiring scheme is called star shaped bus wiring scheme.
In 10Base-T every computer is connected by its own twisted pair cable to a central hub, so the physical wiring looks like a star. Inside the hub, electronics connect all the cables so that they behave like one shared Ethernet cable: every station hears every transmission and CSMA/CD is used. So logically it works as a bus. Hence it is a star-shaped bus.
What is the difference between LAN and WAN?
A LAN (local area network) covers a small area such as a room, floor, building or campus, is usually owned by one organization, and offers high data rates with low delay. A WAN (wide area network) spans large distances such as cities, countries or continents, usually uses leased carrier facilities and packet switches, and generally has lower data rates and higher delay than a LAN; it can grow to connect many sites.
Define the term Jitter.
Jitter is the variation in delay: some packets of the same flow take longer to arrive than others. It matters for real-time audio and video; a network with zero jitter (an isochronous network) delivers every packet with the same delay.
Give a comparison of wiring Schemes.
Thick Ethernet (10Base5): thick coaxial cable forms a bus; each computer needs an external transceiver and AUI cable; segment up to 500 m; costly and hard to install. Thin Ethernet (10Base2): thin, flexible coaxial cable connected directly to each computer with BNC T-connectors; transceiver on the card; cheaper, shorter segments (about 185 m); a break in the cable disrupts the whole segment. Twisted pair Ethernet (10Base-T): each computer has its own twisted pair cable with RJ-45 connectors to a central hub (star wiring, logical bus); cheapest and easiest to install and manage, and one failed cable affects only one computer.
What is the difference between the physical and logical topologies?
Physical topology is the actual layout of the wiring, i.e. how cables connect the computers (for example a star of cables to a hub). Logical topology is how the network technology actually passes data among the computers (for example a bus or a ring), and is defined by the network technology. The two can differ: 10Base-T is a physical star but a logical bus.
Define Vector-Distance Algorithm.
Distance-vector routing is a distributed routing algorithm in which each packet switch keeps a table of (destination, distance, next hop) and periodically sends its list of destinations and distances to its neighbours. When a switch receives a neighbour's list, it updates its own entry for a destination if the route through that neighbour is shorter (neighbour's distance plus the cost of the link). Repeating this lets all switches compute shortest routes.
What is the concept of store and forward technology?
In store-and-forward switching, a packet switch receives a complete packet and stores it in memory (a queue) before processing it. When the outgoing link to the next hop is free, the switch forwards the packet. This lets the switch check the packet, choose the next hop, and handle bursts of traffic by queueing packets, at the cost of some delay; if packets arrive faster than they can be sent, the queue grows and packets may be dropped.
Define static and dynamic routing.
Static routing: routes are entered into the routing tables manually (or by a program at startup) and do not change by themselves; it is simple and has no routing traffic, but cannot adapt if a link or switch fails. Dynamic routing: routing software on the switches/routers exchanges information with neighbours and continuously updates the tables as the network changes, so traffic is rerouted around failures automatically.
What is meant by Bridges STARTUP and STEADY State?
Startup state: when a bridge first starts its address table is empty, so it does not know where any computer is; it forwards every frame to all other segments while learning locations from the source addresses of frames it sees. Steady state: after computers have sent frames, the bridge knows which segment each computer is on, so it forwards a frame only when the destination is on a different segment and filters frames whose destination is on the same segment. This reduces traffic and lets each segment carry local traffic simultaneously.
Six devices are connected in network A,B,C,D,E and F. All devices are on same packet switch. Computer A wants to send packet to E. How will packet switch forwards the packet?
Since A and E are attached to the same packet switch, the switch receives the packet from A on A's port and stores it in memory. It examines the destination address; the address shows that E is a computer attached locally to this switch, so no other switch is involved. The switch looks up the port for E and forwards the packet out of that port directly to E (store and forward).
Network administrator wants to build a tunnel between sites offices. How will he build using private virtual network?
Each site is connected to the public Internet and a VPN device (router or software) is placed at each site. When a packet must go to the other site, the VPN device encrypts the whole packet and encapsulates it inside a new packet addressed to the VPN device at the other site; this forms a tunnel through the Internet. The remote VPN device removes the outer header, decrypts the packet and delivers it on its local network. Outsiders see only encrypted traffic between the two VPN devices, so the sites behave like one private network at low cost.
Who examine performance of network?
A network analyzer, also called a network monitor or network sniffer, examines network performance. It runs in promiscuous mode, captures frames on the LAN, and reports statistics such as utilization, frame sizes and collision rate, and helps debug problems.
How can we compute shortest path in WAN?
Represent the WAN as a weighted graph: packet switches are nodes, links are edges, and each edge weight is the link cost (distance, delay, etc.). Run Dijkstra's shortest-path algorithm from each switch (or use a distributed distance-vector algorithm) to find the least-cost path to every destination, and store the first hop of each path in the switch's routing table.
100mbps Ethernet NIC is connected to control Hub which supports 10mbps. What will be Data Rate in this scenario ?
The data rate will be 10 Mbps. A 10/100 NIC negotiates the speed with the hub and both use the highest rate supported by both ends, which is 10 Mbps.
NIC 100mbps Ethernet NIC and other speed 10mbps. Are connected with each other What will be NIC Data Rate in this scenario?
10 Mbps: the two interfaces autonegotiate and run at the highest rate both support, which is 10 Mbps.
two far away site offices want to connect which is better LAN and WAN?
WAN. A LAN technology is limited to a small area such as a building or campus, while a WAN is designed to span large distances (cities, countries) using leased links and packet switches, so far-away offices must be connected by a WAN.
What is 10base T?
10Base-T is twisted pair Ethernet: 10 Mbps baseband Ethernet over unshielded twisted pair cable. Each computer connects with an RJ-45 connector to its own cable that runs to a central hub, so the wiring is a physical star while the hub makes it behave as a logical bus using CSMA/CD.
What is Multicasting?
Multicasting is sending one copy of a frame or packet that is delivered to a selected group of computers (those that have joined the multicast group) rather than to a single computer (unicast) or to all computers (broadcast). The interface hardware accepts frames sent to the multicast addresses it has been told to recognize.
What is FDDI?
FDDI (Fiber Distributed Data Interconnect) is a ring LAN technology that uses optical fiber between stations and token passing for access, at 100 Mbps. It uses a pair of counter-rotating fiber rings; if a station or fiber fails, the neighbouring stations loop the data back onto the second ring (self-healing), so the network keeps working.
Define 802.11 Wireless lans and CSMA?
802.11 is the IEEE standard for wireless LANs, in which computers communicate by radio, usually through an access point. Because a wireless station cannot reliably detect collisions (and not all stations can hear each other), 802.11 uses CSMA/CA (collision avoidance): a station senses the medium and, before sending data, exchanges short control messages so other stations stay quiet. CSMA (carrier sense multiple access) is the access method in which a station listens for a carrier (signal) on the shared medium and transmits only when the medium is idle; Ethernet adds collision detection (CSMA/CD).
Dijkstra's Algorithm
Dijkstra's algorithm computes the shortest paths from one node to all other nodes in a weighted graph. It keeps a set of nodes whose shortest distance is known; repeatedly it picks the unvisited node with the smallest tentative distance, adds it to the set, and updates the distances of its neighbours through it. For routing, the first hop of each shortest path is put into the switch's routing table.
Disadvantages of Repeaters
A repeater only amplifies and retransmits electrical signals and does not understand frames, so it copies collisions, noise and errors from one segment to the others, cannot filter traffic (all segments share bandwidth and form one collision domain), and only a limited number of repeaters can be used between two stations because of timing limits.
Diffe b/w LAN and WAN
A LAN covers a small area (room, building, campus), is usually owned by one organization, and has high speed and low delay. A WAN spans large distances (cities, countries), uses packet switches and leased long-distance links, and usually has lower speed and higher delay.
Why fiber need for ATM
ATM was designed to carry voice, video and data at very high speeds (OC-3, 155 Mbps and above) with guaranteed quality of service. Optical fiber provides the very high bandwidth needed over long distances, has low error rates, and is immune to electrical interference, so cells are rarely lost or corrupted. Therefore ATM was designed to work on fiber (though it can also run over twisted pair at lower speeds).
define bridge
A bridge is a device that connects two LAN segments and forwards complete frames between them. It does not forward collisions or invalid frames, and it learns where computers are from source addresses so that it forwards a frame only when its destination is on the other segment (filtering otherwise). It is transparent to the computers and makes the segments act as one LAN.
Where are destination and source address located in frame?
They are in the frame header, which comes before the data (payload). In an Ethernet frame, after the preamble, the header starts with the 48-bit destination address, followed by the 48-bit source address and then the 16-bit frame type field.
If 100mbps Ethernet NIC is connected to a hub which supports 10mbps, what will be the data rate in this scenario?
10 Mbps: the NIC and hub autonegotiate and use the highest speed both support.
In connection oriented network, which type of connection identifier is used? Support your answer with example.
A connection-oriented network uses a short connection identifier assigned when the connection is set up, instead of the full destination address in each packet. Example: ATM uses a 24-bit identifier made of an 8-bit VPI (virtual path identifier) and a 16-bit VCI (virtual circuit identifier); every cell carries the VPI/VCI and each switch uses it to forward the cell and rewrites it for the next link.
Imagine a network of 6 devices A, B, C, D, E and F. All devices are on same packet switch. If computer A wants to send packet to computer E, How would packet switch forward this packet?
Since A and E are attached to the same packet switch, the switch receives the packet from A on A's port and stores it in memory. It examines the destination address; the address shows that E is a computer attached locally to this switch, so no other switch is involved. The switch looks up the port for E and forwards the packet out of that port directly to E (store and forward).
Network administrator needs audio and video service for his network. He has to send or receive data at same data rate. Which type of ATM quality service specification would be applied there? Give example?
Constant Bit Rate (CBR) service. CBR guarantees a fixed data rate for the whole connection, which suits uncompressed audio and video that have predefined maximum data rates; for example, a digital voice call or uncompressed video stream sent at a fixed rate over ATM. (VBR would be used for compressed audio/video, ABR for ordinary data.)
We have two satellite locations, at first location one bridge is configured and at second location 2 bridges are configured. Which location performed will be faster? Give reason.
The first location (one bridge) will be faster. A bridge is a store-and-forward device: it must receive each complete frame, check it and look up the destination before forwarding it, which adds delay. A frame that crosses two bridges suffers this delay twice and may also wait in two queues, so end-to-end delay is higher and throughput can be lower than with one bridge. Fewer bridges between communicating stations means less delay.
There are two sites office A & B, both are using VPN. If a user of another network wants to send some packet to A, then what will be the behavior of network A?
Computers inside A use the VPN and usually private addresses; the VPN device at A accepts as VPN traffic only encrypted packets from its partner at B. A packet from a user of another network is not part of the VPN, so it is not admitted into the private network (it is discarded or handled as ordinary Internet traffic under the site's policy).
What type of switching technique is used in WAN?
Packet switching with store and forward. A WAN is built from packet switches connected by long-distance links; each switch receives a complete packet, stores it in memory, chooses the next hop from its routing table and forwards the packet when the outgoing link is free.
Write the names of identifiers used in ATM.
ATM uses the VPI (Virtual Path Identifier, 8 bits) and the VCI (Virtual Circuit Identifier, 16 bits); together they form the 24-bit identifier of a virtual circuit carried in each cell header.
What type of topology used in Ethernet?
Ethernet uses a bus topology: all stations share one medium and use CSMA/CD. Original thick and thin Ethernet are physically a bus; 10Base-T is wired as a physical star to a hub but is still a logical bus.
Defines simplex and full duplex connection?
A simplex connection carries data in one direction only, from one computer to the other. A full duplex connection carries data in both directions at the same time. With a technology that offers only simplex connections, two computers that need two-way communication must set up two connections, one from A to B and one from B to A.
How WAN capable to handle many computers?
A WAN does not use one shared medium; it is built from packet switches joined by point-to-point links. Each switch connects to some local computers and to other switches, and forwards packets by store and forward using routing tables. To serve more computers or sites, more switches and links are simply added, so the WAN can grow to handle many computers while keeping reasonable performance.
Briefly explains the working structure of thick Ethernet.
Thick Ethernet (10Base5) uses a thick coaxial cable as a shared bus with terminators at both ends to stop signal reflection. Each computer has a transceiver attached to the coax, and an AUI (drop) cable connects the NIC to the transceiver; the AUI cable carries digital signals, power and control, and the transceiver sends and senses the analog signal on the coax.
Imagines a network for six devices: A, B, C, D, E and F. all devices are on a same packet switch. Computer A want to send packet to computer E. how would packet switch forward this packet?
Since A and E are attached to the same packet switch, the switch receives the packet from A on A's port and stores it in memory. It examines the destination address; the address shows that E is a computer attached locally to this switch, so no other switch is involved. The switch looks up the port for E and forwards the packet out of that port directly to E (store and forward).
Does ATM eliminate the varying delay times associated with different sized packets? If yes then give reason.
Yes. ATM sends all data in fixed-size cells of 53 octets, so every cell takes the same time to transmit and switches can process them quickly in hardware. A short cell never has to wait behind a long packet, so delay is small and predictable (low jitter), which is why ATM suits voice and video.
If a 100 Mbps Ethernet NIC is connected to a hub supporting 10 Mbps, what will be the data rate?
10 Mbps. A 10/100 NIC detects the hub's speed and operates at the highest rate both support, which is the hub's 10 Mbps.
In connection-oriented networks, which type of connection identifier is used? Give an example.
A connection (virtual circuit) identifier is placed in each cell or packet instead of the full destination address. Example: ATM uses a 24-bit identifier made of an 8-bit VPI and a 16-bit VCI.
Network admin needs audio/video service at same data rate. Which ATM quality service applies? Give example.
Constant Bit Rate (CBR) service, which guarantees a fixed data rate and low delay variation, for example an uncompressed voice call or a live video conference.
Shortest path in WAN?
Dijkstra's algorithm computes the shortest path from a node to every other node in the graph representing the WAN, using edge weights as distances.
Switching techniques in WAN?
WANs use packet switching (circuit switching is the alternative): packet switches receive whole packets, store them and forward them toward the destination (store and forward).
Names of identifiers used in ATM
VPI (Virtual Path Identifier) and VCI (Virtual Channel/Circuit Identifier).
Ethernet topology?
Ethernet uses a bus topology: all stations share one cable (logically a bus even when wired as a star through a hub).
Difference between LAN and WAN
A LAN covers a small area such as a room, floor or building and gives high speed at low cost. A WAN covers large distances such as cities or countries and connects many sites, usually through packet switches and leased lines.
Why fiber is needed for ATM?
ATM was designed for very high speeds (e.g. 155 Mbps and above) and for voice and video with guaranteed service. Fiber gives the high bandwidth, low error rate and long distances these speeds need, so ATM switches are normally connected with fiber.
Simplex and full-duplex connections?
Simplex: data flows in one direction only. Full-duplex: data flows in both directions at the same time.
WAN capability to handle many computers?
A WAN can grow to many computers by adding packet switches and links; switches use store-and-forward and hierarchical addresses (switch, computer) so each switch needs only next-hop information, and the WAN scales as needed.
Thick Ethernet working structure?
Thick Ethernet (10Base5) uses a thick coaxial cable as a bus. Each station's NIC connects through an AUI cable to a transceiver attached directly to the coax; the transceiver sends and receives signals, senses the carrier and detects collisions. Terminators at both ends of the cable stop signal reflections.
ATM eliminates varying delay of packets? Why?
Yes, largely. All ATM cells are the same small size (53 bytes), so each takes the same time to send and switch, and a cell never waits behind a long packet; delay variation (jitter) is therefore small.
FDDI?
FDDI (Fiber Distributed Data Interconnect) is a ring LAN technology using fiber optics at 100 Mbps with token passing. It uses two counter-rotating rings, so if a link or station fails the stations loop back onto the second ring and the network keeps working (self-healing).
Static vs Dynamic routing
Static routing: the table is set (usually manually) when the system boots and does not change. Dynamic routing: routing software running protocols such as RIP or OSPF learns routes and keeps updating the table automatically.
Difference between physical and logical topologies
Physical topology is the actual layout of the wiring and devices. Logical topology is the way data flows between stations, which can differ; for example, 10Base-T is physically a star but logically a bus.
Bridges STARTUP vs STEADY state
At startup a bridge's address table is empty, so it forwards frames to the other segment while it learns source addresses. In steady state it has learned where the computers are, so it forwards a frame only when the destination is on the other segment and filters all local traffic.
How can switched virtual network (SVC) be established?
A switched virtual circuit is set up on demand. The host sends a signalling request to its ATM switch with the destination address and required quality of service; the switches find a path, reserve resources and assign VPI/VCI values along it; the destination accepts, and the connection is then used until either end requests its release.
Write advantages and disadvantages of public and private Networks.
Private network: owned and used by one organization. Advantages: the owner has full control over technology, configuration and security, and can guarantee capacity for its own traffic. Disadvantages: high initial cost and ongoing cost of operation and maintenance, and the organization needs its own expertise; long-distance links usually still must be leased from a carrier. Public network: owned by a service provider (carrier) and shared by many subscribers who pay for service. Advantages: no need to buy or manage equipment, lower cost through sharing, flexible connection to many sites. Disadvantages: less control, data crosses shared facilities so privacy and security are a concern, and performance depends on the provider and on other users.
What are the main factors due to which mechanism of error detections schemes differ?
Error detection schemes differ mainly in: (1) the overhead, i.e. how many extra bits of redundant information must be sent with the data; (2) the amount of computation needed by sender and receiver to produce and check the extra information, which affects speed and hardware cost; and (3) the kinds of errors they can detect, e.g. parity detects only an odd number of changed bits, a checksum misses some errors that cancel out, while a CRC detects more errors including burst errors.
How much data can be present on an original Ethernet segment at one time? To find out, compute the delay-throughput product. The original Ethernet operate at 10 mega bits per second, and a segment was limited to 500 meters. Assume the signals propagate down the cable at 66 percent of the speed of light.
Propagation speed = 0.66 x 3 x 10^8 m/s = 1.98 x 10^8 m/s. Delay over 500 m = 500 / 1.98 x 10^8 = about 2.53 x 10^-6 s (2.53 microseconds). Delay-throughput product = 2.53 x 10^-6 s x 10 x 10^6 bit/s = about 25 bits. So only about 25 bits can be in transit on one segment at a time.
Which type of information is obtained from network sniffer and in which mode Network sniffer operates?
A network analyzer (network monitor or sniffer) is used to check the performance of a network or to debug it. It can report statistics such as capacity utilization, distribution of frame sizes, collision rate or token circulation time, count frames of a given type or size, and show frames sent to or from particular computers; from the type field it can also tell whether a frame carries an LLC/SNAP header. It operates in promiscuous mode: the network interface accepts every frame on the LAN, not only frames addressed to it, so software can examine all traffic. This also shows that LAN communication is not private.
Thick Ethernet, Thin Ethernet and Twisted pair Ethernet.
Thick Ethernet (10Base5) is the original Ethernet wiring: a heavy, inflexible coaxial cable forms a bus up to 500 m per segment; each computer connects through an external transceiver attached to the cable and an AUI cable to the interface card. Thin Ethernet (10Base2) uses thinner, cheaper, flexible coaxial cable, also as a bus; the transceiver electronics are on the interface card and the cable connects directly to each computer with BNC T-connectors, with terminators at the ends; segments are shorter (about 185 m). Twisted pair Ethernet (10Base-T) uses unshielded twisted pair cable with RJ-45 connectors running from each computer to a central hub (maximum about 100 m), so it is physically a star but logically a bus; it is cheap, easy to install, and a broken cable affects only one computer. All three run at 10 Mbps and use CSMA/CD.
Describe permanent virtual circuits (PVC).
A permanent virtual circuit is a connection that is set up manually by the network administrator rather than on demand by a computer. The administrator configures each switch along the path with the forwarding entries and assigns the connection identifiers (VPI/VCI). Once configured, the PVC stays in place and survives reboots and power failures, so the two endpoints can communicate at any time without a setup delay. PVCs are used for long-term connections, such as between two sites of a company; their drawback is that changes need manual configuration and the circuit uses resources even when idle.
How can a bridge know whether to forward frames?
A bridge learns automatically. It runs in promiscuous mode and examines every frame on each attached segment. From the source address of each frame it learns which segment that computer is on and records it in its address table. For each frame it then looks up the destination address: if the destination is on the same segment the frame came from, the bridge filters (does not forward) it; if the destination is on another segment, it forwards the frame only to that segment; if the destination is unknown, or is a broadcast/multicast, it forwards the frame to all other segments. Entries are updated as new frames arrive.
Compare connection oriented and connectionless Service.
Connection-oriented service works like a telephone call: a connection is set up before data is sent, data then flows along that connection (often identified by a short connection identifier), and the connection is released afterwards. It can guarantee quality of service and lets the network know about resource needs in advance, but has setup delay and keeps state in switches; ATM and Frame Relay are examples. Connectionless service works like the postal system: each packet carries the full destination address and is sent independently with no setup, so it is simple, flexible and has no setup delay, but it cannot guarantee resources and packets may follow different paths; Ethernet and other basic LANs, and SMDS, are examples.
What is the concept of packet switches?
A WAN is built from packet switches. A packet switch is a small dedicated computer with a processor, memory and I/O interfaces: some interfaces connect to computers at the local site and others connect by high-speed digital links to other packet switches. It uses store-and-forward: each arriving packet is placed in memory, the switch examines its destination address, uses its routing table to choose the next hop, and forwards the packet when that link is free. Because packets from many users share the same links, the network can grow by adding more switches and links without changing the computers.
Write a note on Weight Graph.
A WAN can be modelled as a graph: each packet switch is a node and each direct link between switches is an edge. In a weighted graph each edge is given a weight (label) that represents the cost of using that link, for example distance, delay, or monetary cost, or a value set by the administrator to prefer some links. Routing algorithms such as Dijkstra's shortest-path algorithm use these weights as the measure of distance and choose the path with the smallest total weight, which may not be the path with the fewest hops. The result is used to fill each switch's next-hop routing table.
We have two satellite locations. At first location one bridge is configured and at 2nd location two bridges are configured. Which location's performance will be faster. Give reasons to support your answer.
The first location (one bridge) will be faster. A bridge is a store-and-forward device: it must receive each complete frame, check it and look up the destination before forwarding it, which adds delay. A frame that crosses two bridges suffers this delay twice and may also wait in two queues, so end-to-end delay is higher and throughput can be lower than with one bridge. Fewer bridges between communicating stations means less delay.
Star Textile has huge network of systems and switches. A task is given to IT Officer that if some link goes the entire network works smoothly. Which routing method will be use.
Dynamic routing should be used, not static routing. With dynamic routing (for example a distance-vector or link-state algorithm) the switches continuously exchange routing information; when a link fails, the neighbouring switches detect it, propagate the change, and all switches recompute their routing tables so that traffic is sent along another available path. With static routing, routes are fixed by hand and the destinations behind a failed link would become unreachable until the administrator changed the tables. In a large network dynamic routing also saves manual configuration work.
Write note on Bridges
A bridge is a device that connects two (or more) LAN segments and passes frames between them, operating at the physical and data link layers. Unlike a repeater it handles complete frames: it does not forward collisions or damaged frames, so each segment is a separate collision domain. A bridge is adaptive (learning): it examines the source address of each frame to learn which segment each computer is on, then filters a frame whose destination is on the same segment and forwards it only when the destination is on another segment (unknown and broadcast frames go everywhere). This lets each segment carry local traffic at the same time and extends the LAN. Bridges are transparent to computers. When bridges form a loop, the distributed spanning tree algorithm is used to disable some links so that frames do not circulate forever.
Different b/w Cells and Packets
A cell is a small fixed-size unit of data; in ATM every cell is 53 octets (5-octet header and 48-octet payload). A packet (frame) is variable in size, with a larger header carrying full addresses. Because cells are all the same size, switches can process them in hardware quickly, the time to send a cell is constant, and delay and jitter are low and predictable, which suits voice and video. But each cell has a high header overhead compared to its payload, and large data must be divided into many cells. Variable-size packets carry data more efficiently but have variable transmission time, making delay less predictable.
define default route how it is diff from routing
Routing is the process of choosing the next hop for each destination; a routing table normally has one entry per destination (or per site) giving the next hop. A default route is a single special entry used for any destination that does not match an explicit entry in the table. Many tables contain long lists of destinations that all use the same next hop; a default route replaces them with one entry, making the table smaller and lookups faster. The switch first looks for a specific entry and uses the default route only if none is found.
There are two sites office A & B, both are using VPN. If a user of another network wants to send some packet to A, then what will be the behavior of network A? How data is protected in VPN environment?
In a VPN, sites A and B are joined through the Internet by VPN devices (servers/routers) at each site, and the VPN behaves like a private network. Computers inside A normally use private addresses that are not reachable from the Internet. A packet from a user of some other network is not part of the VPN: the VPN device at A accepts as VPN traffic only encrypted packets coming from its partner at B, so the outside packet is not let into the private network (it is discarded or treated as ordinary Internet traffic under the site's policy). Data is protected by tunnelling with encryption: when a computer in A sends to B, the VPN device encrypts the whole original packet and encapsulates it in a new packet addressed to the VPN device at B; only B's VPN device can decrypt it, then it removes the outer header and delivers the original packet inside B. Anyone on the Internet sees only encrypted data between the two VPN devices.
How data is protected in VPN environment?
By encryption and tunnelling: the VPN device at the sending site encrypts each packet going to the other site and encapsulates it in a new packet addressed to the VPN device at the other site. Only that device can decrypt it; it removes the outer header and delivers the original packet locally. Outsiders on the Internet see only encrypted data between the two VPN devices.
Why ATM design chose cells over packet?
ATM chose small fixed-size cells because: (1) memory management is simpler, since every buffer is the same size and variable-length packets cause memory fragmentation; (2) hardware does not have to accommodate the largest possible packet or detect where a packet ends, it can simply count bits as a cell arrives, so switching can be done fast in hardware; (3) the time to send a cell is constant, so no complicated interrupt scheme is needed to detect completion, whereas sending a variable-length packet takes a variable time; (4) because delays are fixed and small, quality of service (such as low delay and jitter for voice and video) can be guaranteed much more easily than with variable-length packets.
Star Textile network: Ensure network works smoothly if a link fails. Which routing method?
Dynamic (distributed) routing. Each router runs a routing protocol, exchanges information with its neighbours and recomputes its own table, so when a link fails the routes are changed automatically to avoid it, unlike static routing.
Ambiguity may cause network failure. How parity bit helps?
A parity bit is an extra bit added to each data unit so that the total number of 1s is even (even parity) or odd (odd parity). The receiver counts the 1s; if the count does not match the agreed parity it knows the data was corrupted and can discard it or ask for retransmission. Parity detects any single-bit (odd number of bits) error but cannot detect an even number of changed bits and cannot correct errors.
Define default route. How is it different from routing?
A default route is a single routing-table entry used for every destination that has no explicit entry; it replaces many entries that share the same next hop and keeps tables small. Routing is the overall process of building routing tables and choosing a next hop for each packet; the default route is just one entry used within that process.
ATM design chose cells over packets. Why?
Fixed-size cells make switch hardware simple and fast, since every cell takes the same time and buffers are uniform, which avoids memory fragmentation. Small cells keep delay and jitter low for voice and video, because no cell waits behind a long packet, which makes quality-of-service guarantees easier. Cell switching also allows high-speed parallel hardware.
Reception is a CPU process. Explain.
Frame reception is actually handled mainly by the network interface hardware, not the CPU. The NIC watches the medium, recognises frames addressed to the computer (or broadcast/multicast), checks the CRC and stores the frame in memory (often with DMA) without CPU help. Only then does it interrupt the CPU, whose protocol software processes the headers and passes the data to the right application.
Concept of packet switches
A packet switch is the basic building block of a WAN: a small dedicated computer with a processor, memory and I/O interfaces, some connecting to computers and some to other switches (often over leased lines). It receives a complete packet, stores it in memory, looks up the next hop for the destination and forwards it (store and forward), so many computers and sites can share the links.
Note on weighted graph
A WAN can be modelled as a graph whose nodes are packet switches and whose edges are links. In a weighted graph each edge has a weight such as distance, delay or cost. The shortest path is the path with the lowest total weight, which is not necessarily the one with the fewest hops; Dijkstra's algorithm computes such paths from each node and the results fill the routing tables.
Difference between cells and packets
Cells are small and fixed in size (an ATM cell is 53 bytes), so they take a constant time to send, are easy to buffer and switch in hardware, and give low, predictable delay. Packets are variable in length and usually larger, so transmission time varies, memory management is more complex, small packets can wait behind long ones and delay is less predictable, although header overhead is lower for large data.