A ---------- relies on the hardware manufacturer to assign a unique physical address to each network interface.
AStatic addressing scheme
BConfigurable addressing scheme
CDynamic addressing scheme
DNone of the given
In a static scheme the manufacturer burns a unique address into each interface.
2.
★ Final term Fall 2008Lecture 12
A network uses a ------arranges for computers to be connected in a closed loop.
AStar Topology
BRing Topology
CBus Topology
DNone of the given
In a ring topology the computers form a closed loop.
3.
★ Final term Fall 2008Lecture 14
A system with redundant bridges might have a problem with_______ in the system.
ALoop
BFilters
CSpanning Trees
DAll given choices
Redundant bridges create cycles, so frames can loop endlessly unless a spanning tree is used.
4.
★ Final term Fall 2008Lecture 23
A single networking technology is best for all needs.
ATrue
BFalse
No single network technology suits every need, which is why internetworking is required.
5.
★ Final term Fall 2008Lecture 23
A computer attached to a given network can only communicate with other computers attached to the same network. Is this a problem with multiple networks?
ATrue
BFalse
Isolation of computers on separate networks is the chief problem that internetworking solves.
6.
★ Final term Spring 2010Lecture 23
In the 1970s large organizations began to acquire multiple networks. Each network in the organization formed island. Employees needed to choose a computer appropriate for each task. So they needed multiple screens, keyboards and computers.
AFalse
BTrue
Before internetworking, each network was an island, so users needed separate computers for each network.
7.
★ Final term Spring 2010Lecture 27
Mapping between a protocol address and a hardware address is called Address Resolution.
ATrue
BFalse
Address resolution maps a protocol address to a hardware address.
8.
★ Final term Fall 2008Lecture 29
The general form of an IP datagram is with a header followed by data. The header contains information that controls where and how the datagram is to be sent.
ATrue
BFalse
An IP datagram has a header with delivery information followed by the data.
9.
★ Final term Fall 2008Lecture 29
The process of using a routing table to select a next hop for a given datagram is called_________
AEncapsulation
BReassembling
CRouting or forwarding
DNone of the given
Choosing the next hop from the routing table is called routing or forwarding.
10.
★ Final term Fall 2008Lecture 30
________ field of header indicates whether a datagram is a fragment or a complete datagram.
AFLAGS
BFLAGMENT OFFSET
CIDENTIFICATION
DNone of the given
The FLAGS field bits tell whether the datagram is a fragment and whether more fragments follow.
11.
★ Final term Spring 2010Lecture 30
Reconstruction of original datagram is called reassembly.
ATrue
BFalse
Rebuilding the original datagram from its fragments is called reassembly.
12.
★ Final term Spring 2010Lecture 30
Router detects datagram --------- than network MTU and then it splits into pieces and each piece is ----------than outbound network MTU.
ALarger, smaller
BLarger, larger
CSmaller, larger
DSmaller, smaller
A datagram larger than the MTU is fragmented into pieces that each fit the MTU.
13.
★ Final term Spring 2010Lecture 30
Router detects datagram---------than network MTU
ALarger
BSmaller
CNone of given
DEqual
Fragmentation happens when a datagram is larger than the network MTU.
14.
★ Final term Fall 2008Lecture 34
UDP offers application programs a Message-Oriented Interface, applications can depend on protocol to preserve data boundaries.
ATrue
BFalse
UDP delivers each message as sent, so message boundaries are preserved.
15.
★ Final term Fall 2008Lecture 34
Connectionless service, Message-Oriented protocol, best effort delivery service, arbitrary interaction and operating system independent are the characteristics of _____________
ATCP
BUDP
CIP
DNone of the given
These are the listed characteristics of UDP.
16.
★ Final term Fall 2008
In Direct point to point communication adding the Nth computer requires------------ new connections.
ANone of the given
BN2
CN-1
D(N2 –N)/2
With direct point-to-point links the new computer must be wired to each of the N-1 existing computers.
17.
★ Final term Fall 2008
Hardware that calculates a CRC uses two simple components.
AAND unit and XOR unit
BShift register and XOR unit
CShift register and AND unit
DNone of the given
CRC hardware is built from a shift register and exclusive-OR units.
18.
★ Final term Fall 2008
The Gigabit Ethernet hardware operates at a rate of --------
A10 Mbps
B100 Mbps
C1000 Mbps
DNone of the given
Gigabit Ethernet runs at 1 Gbps, i.e. 1000 Mbps.
19.
★ Final term Fall 2008
_______ protocols of TCP/IP layering model specify how to ensure reliable transfer.
APhysical Layer
BNetwork Interface Layer
CInternet Layer
DTransport Layer
Layer 4, the transport layer, specifies how to ensure reliable transfer.
20.
★ Final term Fall 2008
An Internet Address (IP address) is a unique __________ binary number assigned to a host and used for all communication with host
A48-bit
B32-bit
C24-bit
DNone of the given
An IPv4 address is a 32-bit number.
21.
★ Final term Fall 2008
The address __________ identifies the physical network to which the computer is attached, while the _______ identifies an individual computer on that network.
Aprefix , suffix
Bsuffix , prefix
Csuffix , suffix
DNone of the given
The prefix identifies the network and the suffix identifies the host on it.
22.
★ Final term Fall 2008
___________ places the boundary between the first and second octets
AClass A
BClass B
CClass C
DClass D
Class A uses an 8-bit prefix, so the boundary is after the first octet.
23.
★ Final term Fall 2008
___________ places the boundary between the third and fourth octets.
AClass A
BClass B
CClass C
DClass D
Class C uses a 24-bit prefix, so the boundary is after the third octet.
24.
★ Final term Fall 2008
UDP and TCP are both_________ layer protocols
APhysical
BData link
CNetwork
DTransport
UDP and TCP are both transport layer protocols.
25.
★ Final term Fall 2008
Connection-oriented service, Point-to-point, Complete reliability, Full-duplex communication, Stream interface, Reliable connection startup and Graceful connection shutdown are the services provided by__________
AIP
BNone of the given
CTCP
DUDP
These are the services TCP provides.
26.
★ Final term Fall 2008
________ identifies the application program that sent the data.
ADestination Port
BSource port
CLogical address
DNone of the given
The source port field names the sending application.
27.
★ Final term Fall 2008
The Border Gateway Protocol (BGP) uses_________ for all communication
AUDP
BTCP
CBoth UDP and TCP
DNone of the given
BGP runs over TCP for reliable transport.
28.
★ Final term Fall 2008
Which of the following protocols allows the sender and receiver to enforce polices.
ARIP
BOSPF
CBGP
DRIP and OSPF
BGP provides policies that control which routes are advertised and accepted.
29.
★ Final term Fall 2008
____________ uses distance vector approach to define routing
ABGP
BOSPF
CRIP
DNone of the given
RIP uses the distance-vector algorithm.
30.
★ Final term Fall 2008
Protocol addresses are abstractions provided by ___________.
Ahardware
Bsoftware
Coperating system
Dinternet
Protocol (IP) addresses are created and used by protocol software, not hardware.
31.
★ Final term Fall 2008
These packets serve same purpose on ________ as frames on ______
AIntranet, LAN
BInternet, WAN
CIntranet, WAN
DInternet, LAN
Virtual packets (IP datagrams) play the role in the Internet that frames play on a LAN.
32.
★ Final term Fall 2008
Address mask defines how many bits of address are in suffix?
ATrue
BFalse
The address mask marks the prefix bits, not the suffix.
33.
★ Final term Fall 2008
The term self-identifying is used for Classful IP addresses because the class of the address can be computed from the address__________.
Aitself
Bprefix
Csuffix
Dmask
The leading bits of a classful address reveal its class, so the address itself identifies it.
34.
★ Final term Fall 2008
Find the class of the address:
10100111 11011011 10001011 01101111
AA
BB
CE
DC
The first two bits are 10, which marks a class B address.
35.
★ Final term Fall 2008
Find the class of the address:
11110011 10011011 11111011 00001111
AA
BC
CE
DB
The first four bits are 1111, which marks a class E address.
36.
★ Final term Fall 2008
In which method of Address Resolution Protocol the protocol address is determined by hardware address?
Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange?
AT
BD
CC
DT, C
In closed-form computation the address mapping is a fixed function, so the protocol address is determined by the hardware address.
37.
★ Final term Fall 2008
Which method of Address Resolution Protocol requires hardware broadcast?
Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange?
AD
BT
CC
DT, D
Dynamic message exchange broadcasts a request to find the owner of an address.
38.
★ Final term Fall 2008
Which method of Address Resolution Protocol resolution with minimum delay?
Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange?
AT, D
Bc
CT
DT, C
Table lookup and closed-form computation resolve locally without network messages, so delay is minimal.
39.
★ Final term Fall 2008
In which method of Address Resolution Protocol the implimentation is more difficult?
Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange?
AT, C
BT
CC
DD
Dynamic message exchange needs a request/response protocol, so it is hardest to implement.
40.
★ Final term Fall 2008
On of the design goals for unicast route propagation is __________.
AConsistency
Binconsistency
Cstability
Ddynamic addressing
Unicast routing aims for stable routes because frequent changes cause jitter and out-of-order delivery.
41.
★ Final term Fall 2008
Propagation multicast routing information differs dramatically from unicast route propagation?
ATrue
BFalse
Dynamic group membership and autonomous senders make multicast route propagation very different.
42.
★ Final term Fall 2008
The IP multicast abstraction allows an application running on an arbitrary computer to leave a multicast group at any time. While _____________ application on a computer remain a member of a group.
AOne or more
Bonly one
Cno
Dmany
A host stays a member of a group as long as one or more of its applications remain in it.
43.
★ Final term Fall 2008
To save traffic, an EGP does not summerize routing information from the autonomous system before passing it to another autonomous system.
ATrue
BFalse
An EGP does summarize routing information before passing it to another autonomous system.
44.
★ Final term Fall 2008
In IPv6 the type of address used for collection of computers with same prefix. Are known as_________.
AAnycast
BUnicast
CMulticast
DNon of the given
The handout's cluster address (delivered to one computer sharing a prefix) is the IPv6 anycast address.
45.
★ Final term Fall 2008
Special types of addresses in IPv6 used for multiple destinations; possibly not at same site. Are known as_________.
AUnicast
BAnycast
CMulticast
DNon of the given
A multicast address delivers to a set of destinations that may be at different sites.
46.
★ Final term Fall 2008
Reliability is the responsibility of the _________ layer
ANetwork
BDatalink
CTransport
DApplication
End-to-end reliability is provided by the transport layer.
47.
★ Final term Fall 2008
TCP uses _______ mechanism to control the flow of data.
Adoor
Bwindow
Cacknowledgment
Dretransmission
TCP flow control uses a sliding window advertised by the receiver.
48.
★ Final term Fall 2008
The time for acknowledgement to arrival of packet depends on.
ADistance to destination and Current traffic conditions
BCurrent traffic conditions
CDistance to destination
Dnon of these
Acknowledgement delay depends on both the distance to the destination and current traffic.
49.
★ Final term Spring 2010
________ is used for typical data applications (where the data rate may be unknown and bursty) and allows use of whatever bandwidth is available at a given time.
AConstant Bit Rate (CBR) service
BVariable Bit Rate (VBR) service
CAvailable Bit Rate (ABR) service
DNone of the given
ATM ABR service lets bursty data use whatever bandwidth is currently available.
50.
★ Final term Spring 2010
________ of TCP/IP layering model, corresponds to basic network hardware.
APhysical Layer
BNetwork Interface Layer
CInternet Layer
DTransport Layer
Layer 1, the physical layer, corresponds to basic network hardware.
51.
★ Final term Spring 2010
__________ places the boundary between the second and third octets
AClass A
BClass B
CClass C
DClass D
Class B uses a 16-bit prefix, so the boundary is after the second octet.
52.
★ Final term Spring 2010
Which of the following are interior routing protocols?
ARIP
BOSPF
CBGP
DRIP and OSPF
RIP and OSPF are interior gateway protocols, while BGP is exterior.
53.
★ Final term Spring 2010
__________measures distance in network hops, where each network between the source and destination counts as single hop.
ABGP
BOSPF
CRIP
DNone of the given
RIP uses a hop-count metric.
54.
★ Final term Spring 2010
OSPF is based on--------
ADistance vector routing
BLink state routing
CPath vector routing
DDistance vector routing and Link state routing
OSPF is a link-state routing protocol.
55.
★ Final term Spring 2010
_________ performs local multicast and uses IP-in-IP encapsulation to send multicast datagrams from one site on the Internet to another.
DVMRP does local multicast and tunnels multicast datagrams between sites with IP-in-IP encapsulation.
56.
★ Final term Spring 2010
NEXT HEADER field in the base header defines type of header and it appears at end of fixed-size base header.
ATrue
BFalse
The NEXT HEADER field, at the end of the fixed IPv6 base header, gives the type of the following header.
57.
★ Final term Spring 2010
Although message exchange can be used to bind addresses, sending a request for each binding is hopelessly inefficient.
ATrue
BFalse
Sending an ARP request for every datagram would be wasteful, so responses are cached.
58.
★ Final term Spring 2010
In which method of Address Resolution Protocol the protocol address independent of hardware address?
Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange?
AT, C
BD
CC
DT, D
Only closed-form computation ties the protocol address to the hardware address; table lookup and message exchange do not.
59.
★ Final term Spring 2010
A computer needs a complete stack of protocols to run either a client or a server.
ATrue
BFalse
Both clients and servers rely on the full protocol stack to communicate over the internet.
60.
★ Final term Fall 2008
________is called an end-to-end protocol because it provide a connection directly from an application on one computer to an application on a remote computer.
AIP
BUDP
CTCP
DNone of the given
TCP is called end-to-end because it gives a connection directly between applications on two computers.
61.
★ Final term Fall 2008
_______ is ideal in a situation where the group is small and all members are attached to contiguous Local Area Networks.
AFlood-and -Prune
BConfiguration-and -Tunneling
CCore-Based Discovery
DNone of the given
Flood-and-prune suits small groups on contiguous LANs.
62.
★ Final term Fall 2008
Router that decrements TTL to __ sends ICMP time exceeded message, with router’s address as source address
A3
B2
C1
D0
When a router decrements TTL to zero it discards the datagram and sends ICMP time exceeded.
63.
★ Final term Fall 2008
ARP is almost always used to bind a ___-bit IP address to a ___-bit Ethernet address.
A32, 48
B24, 32
C32, 64
D32, 128
ARP maps 32-bit IPv4 addresses to 48-bit Ethernet addresses.
64.
★ Final term Fall 2008
The number of connections needed for N computer in direct point to point communication is equal to:
A(N2 –N)/2
BN(N- 1)
CN2
DNone of the given
Each of N computers links to every other, giving N(N-1)/2 = (N2-N)/2 links.
65.
★ Final term Fall 2008
A multicast routing scheme in which the protocol software builds a delivery tree from a central point is called__________
Core Based Trees builds the delivery tree from a central core router.
66.
★ Final term Fall 2008
Whenever it handles a packet, IP software needs to separate the destination address into a _________ and ___________.
Apostfix, Infix
Bnon of these
CInfix, prefix
Dprefix, suffix
IP splits a destination address into its network prefix and host suffix.
67.
★ Final term Fall 2008
End-to-end delivery service is connection oriented.
ATrue
BFalse
IP's end-to-end delivery service is connectionless.
68.
★ Final term Fall 2008
Which method of Address Resolution Protocol is useful with any hardware?
Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange?
AT
BC
CD
DC, D
Table lookup works with any hardware because bindings are simply stored.
69.
★ Final term Fall 2008
We use the term _______ to refer to a measure of the path that routing software use when choosing a route.
Arouting path
Brouting metric
Crouting
Dswitching
A routing metric is the measure used to compare paths.
70.
★ Final term Fall 2008
In which method of Address Resolution Protocol the protocol address is determined by hardware address? Were "T" stands for Table lookup, "C" for Closed-form Computation and "D" for Data Exchange? Features / Types of Resolution: Use full with any hardware - T; Address change affects all hosts - T; Protocol address is determined by hardware address - C
AT
BD
CC
DT, C
In closed-form computation the protocol address is derived from the hardware address.
71.
★ Final term Fall 2010
The number of connections needed for N computer in direct point to point communication is equal to:
A(N^2 – N)/2
BN(N- 1)
CN^2
DNone of the given
Each of the N computers links to the other N-1 and each link is shared by two, giving (N^2-N)/2.
72.
★ Final term Spring 2010
----------- provide Application to application communication it also called end to end communication
AIP
BTP
CRIP
DNone of the given
Transport protocols (TP) give application-to-application, end-to-end communication.
73.
★ Final term Spring 2010
A routing table contains_______
AThe destination network ID
BThe hop count to reach the network
CThe router ID of the next hop
DAll of the given
A routing table entry holds the destination network, the next hop and a metric such as hop count.
74.
★ Final term Spring 2010
_______ includes a 32-bits address mask with each address, which allows the address to be classful, classless, or subnetted.
ARIP
BOSPF
CBGP
DNone of the given
OSPF carries a 32-bit mask with each address.
75.
★ Final term Spring 2010
Like most application programs, a client and server use a transport protocol to communicate.
ATrue
BFalse
Clients and servers communicate through a transport protocol such as TCP or UDP.
76.
★ Final term Spring 2010
Information can flow in either or both direction between
AClients
BClients and servers
CServers
DNone of given
Data can flow either or both ways between a client and a server.
77.
★ Final term Spring 2010
IPV6 address consists of ________
A32 Bits
B64 Bits
C128 Bits
Dnone of the given
IPv6 addresses are 128 bits long.
78.
★ Final term Spring 2010
In case TCP, retransmission, acknowledgment from a computer on LAN are expected to arrive within
ASeconds
BMicro seconds
CMilliseconds
DNanoseconds
On a LAN, round-trip times and so acknowledgements are in the millisecond range.
79.
★ Final term Spring 2010
Twice NAT is another variant of NAT. it is used with site that runs server. In this process NAT box is connected to Domain Name.
ATrue
BFalse
Twice NAT lets a site run servers by working with the domain name system.
No MCQs match this filter.Is filter mein koi MCQ nahi.
1.
★ Final term Fall 20113 marks
How ICMP used to test different tools?
Show answerJawab dekhein
Ping sends ICMP echo request messages and waits for echo replies to test whether a destination is reachable and to measure round-trip time. Traceroute sends datagrams with TTL 1, 2, 3, ...; each router where the TTL expires returns an ICMP time exceeded message, revealing the routers on the path, and the destination finally answers (e.g. with an ICMP destination/port unreachable or echo reply). Path MTU discovery sends datagrams with the do-not-fragment bit set and uses the ICMP 'fragmentation needed' replies to find the smallest MTU on the path.
2.
★ Final term Fall 20113 marks
How does host join and leave a group?
Show answerJawab dekhein
A host uses IGMP (Internet Group Management Protocol). To join, an application asks the host's IP software to join the group and the host sends an IGMP membership report to the local multicast router, which then starts receiving and forwarding that group's datagrams onto the network using multicast routing. The router periodically sends IGMP queries; the host remains a member while at least one of its applications is in the group. To leave, the host stops answering queries (or sends a leave message), and when no member remains on that network the router stops forwarding the group's traffic there.
3.
★ Final term Fall 20113 marks
When packet lost what is the procedure TCP adopt?
Show answerJawab dekhein
TCP detects loss with acknowledgements and timers. The sender keeps a copy of each segment and starts a retransmission timer; the receiver acknowledges only data received in order, so a missing segment is not acknowledged. When the timer expires without an acknowledgement the sender retransmits the segment. The timeout is adaptive (based on measured round-trip time), and TCP also treats loss as a sign of congestion and reduces its sending window before increasing it again.
4.
★ Final term Fall 2011Lecture 263 marks
In this subnet block 192.168.1.0/26 what is the range of assignable host address?
Show answerJawab dekhein
A /26 leaves 6 host bits, so the block has 64 addresses: 192.168.1.0 to 192.168.1.63. 192.168.1.0 is the network address and 192.168.1.63 is the broadcast address, so the assignable host range is 192.168.1.1 to 192.168.1.62 (62 hosts).
5.
★ Final term Fall 2011Lecture 102 marks
Write the difference between explicit and implicit frame type.
Show answerJawab dekhein
Explicit frame type: the frame header contains a type field whose value identifies the kind of data carried (e.g. the Ethernet type field). Implicit frame type: the frame has no type field; sender and receiver must agree on the contents or the type is identified inside the data area (e.g. by an LLC/SNAP header).
6.
★ Final term Fall 20112 marks
Give the concept of zero compression regarding IPv6.
Show answerJawab dekhein
In colon hexadecimal notation a run of consecutive 16-bit groups that are all zero can be replaced by two colons (::), which may appear only once in an address. Example: FF0C:0:0:0:0:0:0:B1 is written FF0C::B1.
7.
★ Final term Fall 20112 marks
Can multiple IP addresses assigned or not on different interfaces of a router?
Show answerJawab dekhein
Yes. A router connects to several networks and is assigned one IP address for each interface; each address has the prefix of the network that interface connects to, because an IP address identifies a connection to a network, not the device.
8.
★ Final term Fall 20112 marks
In which process backward compatibility of 100Base-T is done?
Show answerJawab dekhein
Through auto-negotiation (auto-sensing): dual-speed 10/100 interfaces and hubs/switches detect the speed the other end supports and automatically operate at 10 Mbps or 100 Mbps, so 100Base-T works with older 10Base-T equipment.
9.
★ Final term Fall 20112 marks
Does OSPF only share information with an area or does it allow communication between different areas?
Show answerJawab dekhein
OSPF allows an autonomous system to be divided into areas. Full link-state information is flooded only inside an area; area border routers summarise routes and pass the summaries to other areas through the backbone area. So OSPF does allow communication between areas, but in summarised form, which reduces broadcast traffic, memory and computation and lets OSPF scale to many routers.
10.
★ Final term Fall 20082 marks
How can we prove that we have 2,147,483,648 addresses in class A?
Show answerJawab dekhein
In a class A address the first bit is fixed at 0, leaving 31 free bits. So the class A space has 2^31 = 2,147,483,648 addresses (2^7 = 128 network prefixes, each with 2^24 = 16,777,216 suffixes; 128 x 16,777,216 = 2,147,483,648).
11.
★ Final term Fall 20082 marks
Why is internet multicast routing difficult?
Show answerJawab dekhein
Group membership is dynamic (hosts join and leave at any time), members can be anywhere in the Internet, and any host may send to a group without being a member (autonomous senders). Routers must therefore keep changing delivery trees as membership changes and must avoid flooding datagrams to networks with no members.
12.
★ Final term Fall 20082 marks
Define what is the Extension Headers in IPv6.
Show answerJawab dekhein
IPv6 extension headers are optional headers placed between the fixed base header and the data. Each header's NEXT HEADER field identifies the type of the following header. They carry optional information, such as fragmentation, source routing, hop-by-hop options and authentication, only when it is needed.
13.
★ Final term Fall 2008Lecture 303 marks
How does IP software reassemble fragments that arrive out of order?
Show answerJawab dekhein
Every datagram carries an IDENTIFICATION number that is copied into each of its fragments. The destination groups arriving fragments by source address and identification, uses the FRAGMENT OFFSET field to place each fragment's data at its correct position, and uses the MORE FRAGMENTS flag to recognise the last fragment. When all pieces with no gaps are present it rebuilds the original datagram (reassembly); a reassembly timer discards the pieces if some fragment never arrives.
14.
★ Final term Fall 2008Lecture 263 marks
What is the first address in the block if one of the addresses is 167.199.170.82/27?
Show answerJawab dekhein
With /27, the first 27 bits are kept and the remaining 5 bits are set to 0. The last octet 82 = 01010010; keeping its first 3 bits gives 01000000 = 64. So the first address is 167.199.170.64/27 (the block runs to 167.199.170.95).
15.
★ Final term Fall 20083 marks
In internet routing how does a host join or leave a group?
Show answerJawab dekhein
A host uses IGMP (Internet Group Management Protocol). To join, an application asks the host's IP software to join the group and the host sends an IGMP membership report to the local multicast router, which then starts receiving and forwarding that group's datagrams onto the network using multicast routing. The router periodically sends IGMP queries; the host remains a member while at least one of its applications is in the group. To leave, the host stops answering queries (or sends a leave message), and when no member remains on that network the router stops forwarding the group's traffic there.
16.
★ Final term July 20122 marks
How long TCP should wait before retransmitting?
Show answerJawab dekhein
There is no fixed time. TCP uses adaptive retransmission: it measures the round-trip time of each segment, keeps a weighted average of the round-trip time and of its variation, and sets the timeout slightly above the average. The wait therefore grows when Internet delays grow and shrinks when they fall.
17.
★ Final term July 20122 marks
Is TCP/IP suite including ARP? What kind of messages are in ARP.
Show answerJawab dekhein
Yes, ARP is part of the TCP/IP protocol suite; it maps IP addresses to hardware addresses on a physical network. It has two kinds of messages: an ARP request (broadcast, asking for the hardware address of a given IP address) and an ARP reply (sent back by the target with its hardware address).
18.
★ Final term July 20122 marks
What are some of the metrics used by routing protocols?
Show answerJawab dekhein
Common routing metrics are hop count, delay, bandwidth (throughput), load, reliability and an administrative cost assigned by the manager. For example, RIP uses hop count, while OSPF lets the manager assign link costs.
19.
★ Final term July 20123 marks
How TCP provides reliability?
Show answerJawab dekhein
TCP achieves reliability with positive acknowledgement and retransmission: every byte has a sequence number; the receiver sends acknowledgements for data received; the sender keeps a copy of unacknowledged data and retransmits it when an adaptive retransmission timer expires. A checksum detects corrupted segments, sequence numbers let the receiver reorder data and discard duplicates, a sliding window provides flow control, and a three-way handshake gives reliable connection startup and graceful shutdown.
20.
★ Final term July 2012Lecture 173 marks
Is Link state is better than Distance vector is? if yes then explain reason
Show answerJawab dekhein
Generally yes, especially for large networks. In link-state routing each router receives the status of every link and computes shortest paths itself (Dijkstra), so routes converge quickly after a change and problems like count-to-infinity and slow convergence of distance-vector routing are avoided; it also scales using areas. The cost is greater complexity, memory and computation than simple distance-vector routing.
21.
★ Final term July 20123 marks
What is the difference between an interior gateway protocol and an exterior gateway protocol? Name an example of each.
Show answerJawab dekhein
An Interior Gateway Protocol (IGP) is used among routers inside one autonomous system (one administrative authority) to exchange routes to destinations within that system, e.g. RIP or OSPF. An Exterior Gateway Protocol (EGP) is used between autonomous systems to exchange reachability information about destinations throughout the Internet and to apply routing policies, e.g. BGP.
22.
★ Final term Spring 20102 marks
To solve the problem the routers and networks in the Internet are divided into groups. How large do you think are these groups?
Show answerJawab dekhein
The groups are autonomous systems: sets of networks and routers under one administrative authority. There is no fixed size; an autonomous system can be as small as one organization's few networks or as large as an ISP's or a corporation's many networks and routers, as long as it is managed by one authority.
23.
★ Final term Spring 20102 marks
Traceroute continues to increment the Time To Live until the value is large enough for the datagram to reach its final destination. What happens when the TTL is sufficiently large for the datagram to reach its destination?
Show answerJawab dekhein
The datagram reaches the destination host, so no router sends an ICMP time exceeded message. Traceroute sends its probes to an unused UDP port (or as echo requests), so the destination answers with an ICMP destination (port) unreachable message (or an echo reply). That reply tells traceroute that the destination has been reached and it stops.
24.
★ Final term Spring 20102 marks
How can TCP monitor internet delays?
Show answerJawab dekhein
TCP records the time each segment is sent and the time its acknowledgement arrives, giving a round-trip time sample. It keeps a weighted average of these samples and of their variance and uses them to set an adaptive retransmission timeout, so it follows changes in Internet delay.
25.
★ Final term Spring 20102 marks
Which wireless standard is used in WIFI technology?
Show answerJawab dekhein
Wi-Fi is based on the IEEE 802.11 wireless LAN standards (e.g. 802.11b/g/n).
26.
★ Final term Spring 20103 marks
What kind of information is in a routing table?
Show answerJawab dekhein
Each entry gives a destination (network address with its address mask) and the next hop: the IP address of the next router (or an indication of direct delivery) and the interface to use. Tables usually also hold a default route and may store a metric (cost) for each route.
27.
★ Final term Spring 2010Lecture 103 marks
Define frames without type field?
Show answerJawab dekhein
Frames without a type field use implicit frame typing. The frame header does not say what the data is, so either the sender and receiver agree in advance on the type of data, or the first bytes of the data area carry the type information (for example an IEEE 802.2 LLC/SNAP header that identifies the protocol).
28.
★ Final term Spring 20103 marks
What is the meaning of Facilities For Transit Routing as a characteristic of the Border Gateway Protocol?
Show answerJawab dekhein
BGP can classify each autonomous system as a transit system, which agrees to carry traffic between other autonomous systems, or a stub system, which carries only its own traffic. BGP therefore supports transit routing: an AS can advertise routes and pass traffic on for other ASes, under the control of its policies.
29.
★ Final term Fall 20082 marks
Is there a comparison between TCP/IP reference model and OSI reference model.
Show answerJawab dekhein
Both are layered reference models used to show how protocols fit together. The OSI model has seven layers (physical, data link, network, transport, session, presentation, application) and was designed as a standard before protocols existed. The TCP/IP model used in the handouts has five layers (physical, network interface, internet, transport, application); it was built around the protocols actually used in the Internet and has no separate session and presentation layers, whose functions go into the application layer. Layers correspond roughly but not one-to-one.
30.
★ Final term Fall 2008Lecture 302 marks
How can a datagram be transmitted across a physical network that does not understand the datagram format?
Show answerJawab dekhein
By encapsulation. The sender places the entire datagram in the data area of a hardware frame and sets the frame type field to show that the frame carries an IP datagram. The hardware treats it like any other frame, and the receiver uses the frame type to pass the data to IP.
31.
★ Final term Fall 20082 marks
What is the basic concept of Twice NAT (Network Address Translation)?
Show answerJawab dekhein
Twice NAT translates both the source and the destination addresses of datagrams. It is used when a site runs a server that outside hosts must reach: the NAT box works with the site's Domain Name System, so when an outside host looks up the server's name, NAT creates a mapping and returns an address; it then rewrites both addresses in packets passing in each direction.
32.
★ Final term Fall 2008Lecture 293 marks
What format is used for an internet packet?
Show answerJawab dekhein
An internet packet is an IP datagram: a header followed by a data (payload) area. The IPv4 header contains version, header length, service type, total length, identification, flags, fragment offset, time to live, protocol type, header checksum, source IP address, destination IP address and optional options with padding.
33.
★ Final term Fall 20083 marks
"To achieve a hierarchy, OSPF allows an autonomous system to be partitioned for routing purposes". Does this feature make OSPF more complex or powerful?
Show answerJawab dekhein
It makes OSPF more powerful. Dividing the autonomous system into areas means full link-state information is flooded only inside each area and only summaries cross area boundaries, so traffic, memory and computation are reduced and OSPF scales to large networks. It adds some configuration complexity, but the gain in scalability outweighs it.
34.
★ Final term Fall 20083 marks
Why does IPv6 use separate Extension Headers?
Show answerJawab dekhein
So that the base header stays small and fixed, which makes processing fast; optional information (fragmentation, source routing, options, security) is added only when needed, each header naming the next through the NEXT HEADER field; and new features can be added later by defining new extension headers, which makes IPv6 extensible. For example, fragmentation information is kept in a separate fragment header.
35.
★ Final term Fall 20082 marks
Is there a comparison between TCP/IP reference model and ISO reference model.
Show answerJawab dekhein
Both are layered reference models used to show how protocols fit together. The OSI model has seven layers (physical, data link, network, transport, session, presentation, application) and was designed as a standard before protocols existed. The TCP/IP model used in the handouts has five layers (physical, network interface, internet, transport, application); it was built around the protocols actually used in the Internet and has no separate session and presentation layers, whose functions go into the application layer. Layers correspond roughly but not one-to-one.
36.
★ Final term Fall 20082 marks
Does OSPF only share information within an area or does it allow communication between areas?
Show answerJawab dekhein
OSPF allows an autonomous system to be divided into areas. Full link-state information is flooded only inside an area; area border routers summarise routes and pass the summaries to other areas through the backbone area. So OSPF does allow communication between areas, but in summarised form, which reduces broadcast traffic, memory and computation and lets OSPF scale to many routers.
37.
★ Final term Fall 20082 marks
What are the implementations of Network Address Translation?
Show answerJawab dekhein
NAT can be implemented in software on a computer that connects a site to the Internet (for example Windows Internet Connection Sharing or Linux IP masquerading), or in stand-alone devices such as wireless routers, DSL and cable modems and ISP equipment. Common variants are basic NAT, NAPT (address and port translation) and Twice NAT.
38.
★ Final term Fall 20083 marks
Describe the difference between static and dynamic routing?
Show answerJawab dekhein
In static routing the routing table is set when the system boots (usually by the manager) and does not change afterwards. In dynamic routing the table is initialised at boot and then routing software running a routing protocol learns routes and keeps updating the table, so it adapts to failures and changes.
39.
★ Final term Fall 2008Lecture 263 marks
What is the first address in the block if one of the addresses is 140.120.84.24/20?
Show answerJawab dekhein
With /20, keep the first 20 bits and set the other 12 bits to 0. The first two octets (16 bits) stay 140.120; the third octet 84 = 01010100 keeps its first 4 bits, giving 01010000 = 80; the last octet becomes 0. The first address is 140.120.80.0/20.
40.
★ Final term Fall 20083 marks
Write three new features of IPv6.
Show answerJawab dekhein
IPv6 uses 128-bit addresses; it has a completely new header format with a fixed base header and optional extension headers placed before the data; it provides flow labels and quality-of-service support for audio and video; and it is extensible, so new features can be added easily.
41.
★ Final term past paperLecture 42 marks
CRC detect what kind of error?
Show answerJawab dekhein
CRC detects all single-bit errors, all errors that change an odd number of bits, and all burst errors shorter than the CRC length, and most longer bursts (vertical/burst errors that simple parity misses).
42.
★ Final term past paper2 marks
Define IPv6 NEXT HEADER.
Show answerJawab dekhein
NEXT HEADER is a field at the end of the IPv6 base header (and in each extension header) that identifies the type of the header that follows: either an extension header or the payload protocol such as TCP or UDP.
43.
★ Final term past paper3 marks
Describe three features of closed-form computation.
Show answerJawab dekhein
In closed-form computation the hardware address is computed from the protocol address by a simple function (for example the host part of the IP address is used as the hardware address), so the protocol address is determined by the hardware address. It needs no table and no network messages, so resolution is very fast with minimum delay. It requires configurable hardware addresses that the manager can set to match the protocol addresses.
44.
★ Final term past paper3 marks
TCP timer for retransmission.
Show answerJawab dekhein
When TCP sends a segment it starts a retransmission timer; if the acknowledgement does not arrive before the timer expires, the segment is sent again. Because Internet delays vary, the timeout is adaptive: TCP measures the round-trip time of each acknowledged segment, keeps a weighted average and the variance, and sets the timeout from them; the timeout is increased (backed off) after repeated retransmissions.
45.
★ Final term past paper3 marks
Difference between forwarding and routing.
Show answerJawab dekhein
Forwarding is the per-datagram action of looking up the destination in the routing table, choosing the next hop and sending the datagram out of the right interface. Routing is the process of building and maintaining the routing tables, either manually (static) or by routing protocols that exchange route information (dynamic).
46.
★ Final term past paper3 marks
Define ATM technology.
Show answerJawab dekhein
ATM (Asynchronous Transfer Mode) is a connection-oriented, high-speed switching technology designed to carry voice, video and data. It sends all data in fixed-size 53-byte cells (5-byte header, 48-byte payload) over virtual circuits identified by VPI/VCI, supports quality-of-service guarantees, and usually uses fiber links between ATM switches.
47.
★ Final term 20112 marks
What are the functions that the IP multicast abstraction allows an application running on an arbitrary computer to do?
Show answerJawab dekhein
It allows an application on any computer to join a multicast group at any time, to leave the group at any time, and to send a message to a group even if it is not a member. The host stays a member as long as one or more of its applications belong to the group.
48.
★ Final term 2011Lecture 262 marks
Give the address 130.4.102.1/24 , find the number of subnet bits?
Show answerJawab dekhein
130 lies in 128-191, so the address is Class B with a default 16-bit prefix. The /24 mask uses 24 prefix bits, so 24 - 16 = 8 bits are used for the subnet.
49.
★ Final term 2011Lecture 272 marks
What is the functionality of address resolution software in layering?
Show answerJawab dekhein
Address resolution software sits between the IP layer and the network interface. It hides the hardware address details: upper layers use only IP (protocol) addresses, and the resolution layer maps the next-hop IP address to the hardware address needed to send the frame. This keeps upper-layer software general and independent of the hardware.
50.
★ Final term 20112 marks
Why we need the variants of NAT? Explain it with the proper reasons?
Show answerJawab dekhein
Basic NAT only changes IP addresses, so each internal host needs its own public address while it communicates and it does not support connections started from outside. NAPT also translates protocol port numbers, so many internal hosts can share one public address (the most common form). Twice NAT works with the site DNS so that outside hosts can reach servers inside the site.
51.
★ Final term 20113 marks
How ICMP can be used to test different tools?
Show answerJawab dekhein
ICMP messages are used by test tools. Ping tests whether host A is reachable from host B: it sends an ICMP echo request and the internet layer of the destination returns an ICMP echo reply, which also gives the round-trip time. Traceroute sends datagrams with TTL 1, 2, 3 and so on; each router that drops a datagram because TTL reaches zero returns an ICMP time exceeded message, revealing the path to the destination.
52.
★ Final term 20113 marks
ABC industry is using different network technologies in its branches. Can all branches communicate with each other? If No, then give reason?
Show answerJawab dekhein
Not directly. Each network technology has its own frame format, addressing and signalling, so a computer on one technology cannot talk to a computer on an incompatible one. The branches can communicate only if they are joined into an internet using routers and a common protocol such as TCP/IP, which gives universal service over the different technologies.
53.
★ Final term 20113 marks
Which type of NAT fails if an application uses the IP addresses instead of domain name? And why?
Show answerJawab dekhein
Twice NAT. It lets outside hosts reach servers inside the site by working with the DNS: when an outside host looks up the server name, the DNS/NAT returns a public address and creates the translation entry. If an application uses an IP address directly, no DNS lookup happens, no mapping is created, and the NAT box cannot translate the incoming datagram, so the communication fails.
54.
★ Final term 2011Lecture 302 marks
How can a datagram are transmitted across a physical network that does not understand the datagram format?
Show answerJawab dekhein
By encapsulation. The whole IP datagram is placed in the data area of a frame of that network; the frame type field marks the contents as an IP datagram and the next-hop IP address is mapped to a hardware address for the frame header. The hardware treats the datagram as ordinary data. At the next router the frame is discarded and the datagram is encapsulated again in a frame of the next network.
55.
★ Final term 2011
Describe the process of routing packets
Show answerJawab dekhein
Routing moves a datagram from source to destination across an internet. Each router receives the datagram, extracts the destination IP address and compares it with its routing table (using each entry's mask and destination). The matching entry gives the next hop (or a default route if nothing else matches). The router decrements TTL, encapsulates the datagram in a frame for the next network and sends it on, until the router attached to the destination network delivers it directly. If no route exists the datagram is dropped and an ICMP error is returned.
56.
★ Final term 2011
How ICMP can be used to trace a route?
Show answerJawab dekhein
Traceroute sends a datagram with TTL 1; the first router decrements TTL to zero, discards it and returns an ICMP time exceeded message containing its address. It then sends TTL 2, 3 and so on, learning each router in turn. To detect the destination it either sends an ICMP echo request (answered by an echo reply) or sends to an unused UDP port so the destination returns an ICMP destination (port) unreachable message.
57.
★ Final term 2011
What is the scale or level of requirement in of IPv6?
Show answerJawab dekhein
IP had to change because the scale of networking changed dramatically: the Internet grew from a few tens of computers to tens of millions, link speeds rose from 56 Kbps to 1 Gbps and more, and frame sizes in hardware increased. IPv6 was designed for this scale, mainly with a much larger 128-bit address space.
58.
★ Final term 2011
What are the three approaches for datagram forwarding?
Show answerJawab dekhein
For multicast datagram forwarding three approaches are used: flood-and-prune (send everywhere, then routers with no members prune the tree; good for small groups on contiguous LANs), configuration-and-tunneling (multicast routers are configured and datagrams are tunnelled through IP-in-IP between sites), and core-based discovery (a core router forms the tree and members join toward it).
59.
★ Final term 20123 marks
How TCP and IP interact with each other?
Show answerJawab dekhein
TCP uses IP to carry its messages. A TCP segment is encapsulated in an IP datagram, IP routes the datagram through the internet to the destination host, and there IP passes the contents to TCP. IP only delivers datagrams between computers (best effort); TCP adds application-to-application connections and reliability on top of it.
60.
★ Final term 20122 marks
Describe four factors for network classificatio n ?
Show answerJawab dekhein
Networks are classified by: size (LAN, MAN, WAN); connectivity or topology (point-to-point links versus broadcast/shared media); medium (wired such as copper or fibre, or wireless); and mobility (fixed or mobile computers).
61.
★ Final term 20122 marks
differentiate between IP and transport protocol with the help of example .
Show answerJawab dekhein
IP gives computer-to-computer (machine-to-machine) delivery: its addresses identify computers, and it gives best-effort datagram delivery. A transport protocol gives application-to-application (end-to-end) communication: it adds ports to identify applications and may add reliability. Example: IP carries a datagram from host A to host B, while TCP or UDP delivers the data inside it to a web server program on port 80 of host B.
62.
★ Final term 20122 marks
Give the main advantage and disadvantage of RIP .
Show answerJawab dekhein
Advantages: RIP is simple to understand, configure and deploy; it is widely supported; it can propagate a default route; and hosts can listen passively to update their tables. Disadvantages: its only metric is hop count, so it ignores delay and bandwidth; the maximum is 15 hops (16 means unreachable), so it does not scale to large networks; it converges slowly after changes and can suffer count-to-infinity; and its periodic broadcast of whole tables adds traffic.
63.
★ Final term 2012Lecture 262 marks
Tel the first assignable IP address from a 128.140.80.24/20 .
Show answerJawab dekhein
With /20 the third octet 80 = 01010000 keeps its first 4 bits (0101), giving the network 128.140.80.0 (block 128.140.80.0 to 128.140.95.255). The network address itself is not assigned to a host, so the first assignable address is 128.140.80.1.
64.
★ Final term 20122 marks
how was the NAT implemented .
Show answerJawab dekhein
NAT can be implemented as a stand-alone hardware box between the site and its ISP, inside a router (often by the ISP), or as software on a computer that connects to the Internet and shares the connection (for example at home with DSL or cable). In each case the NAT device keeps a translation table: an entry is created when an outgoing datagram is seen, the private source address (and in NAPT the port) is replaced with the public one, and replies are translated back using the table.
65.
★ Final term 20122 marks
IS ATM include LAN and WAN network. If yes what kind of connection is established.
Show answerJawab dekhein
Yes. ATM was designed to be used both as a LAN and as a WAN technology. It is connection-oriented: a virtual circuit (permanent or switched) is set up between the endpoints before data is sent as fixed-size cells.
66.
★ Final term 20123 marks
is IP multicasting beneficial? Defend your answer with proper reason.
Show answerJawab dekhein
Yes. With multicast the sender transmits one copy of a datagram and routers duplicate it only where paths to group members split, instead of the sender sending a separate unicast copy to every receiver. This saves bandwidth and sender processing, suits audio/video conferencing and other one-to-many applications, and allows members to join or leave at any time without the sender knowing each receiver.
67.
★ Final term 2012Lecture 133 marks
Can the length of the segment be increased 500 meter by adding three repeater one with each segment. It can be done or not.
Show answerJawab dekhein
A single Ethernet segment is still limited to 500 m; a repeater does not lengthen a segment, it joins segments and regenerates the signal. One repeater doubles and two repeaters triple the total length, so three repeaters can join four 500 m segments. This cannot be repeated without limit: the Ethernet standard allows only a small number of repeaters (at most four) between any two stations because CSMA/CD needs the signal delay to stay small enough to detect collisions.
68.
★ Final term 20123 marks
How an administrator can handle static and dynamic routing.
Show answerJawab dekhein
With static routing the administrator builds the routing table by hand (or in a startup file); it does not adjust to network changes, so every change must be made manually. With dynamic routing the administrator enables a routing protocol (such as RIP or OSPF) and the routers build and update the tables automatically, reacting to failures and changes; the administrator mainly configures the protocol and policies.
69.
★ Final term 2012Lecture 283 marks
IS TCP/IP suit include ARP. What kind of messages are in ARP.
Show answerJawab dekhein
Yes, the TCP/IP suite includes the Address Resolution Protocol (ARP). ARP defines two message types: a request, which is broadcast and contains the IP address to be resolved, and a reply (response), sent back directly by the computer that owns that IP address and containing its hardware address. Both use the same message format.
70.
★ Final term Fall 20122 marks
IS ATM including LAN and WAN network. If yes what kind of connection is established?
Show answerJawab dekhein
Yes. ATM was designed to be used both as a LAN and as a WAN technology. It is connection-oriented: a virtual circuit (permanent or switched) is set up between the endpoints before data is sent as fixed-size cells.
71.
★ Final term Fall 20123 marks
Transit routing.
Show answerJawab dekhein
In BGP, transit routing means carrying traffic that neither starts nor ends in the autonomous system, i.e. passing traffic of other ASes through it. BGP classifies each AS as a transit AS, if it agrees to pass such traffic, or a stub AS, if it does not. A multi-homed corporation can declare itself a stub so that it refuses transit traffic.
72.
★ Final term Fall 2012Lecture 23
Are TCP/IP protocols organized into conceptual layers?
Show answerJawab dekhein
Yes. TCP/IP protocols are organized into five conceptual layers: physical, network interface, internet, transport and application; each layer handles one part of the communication problem.
73.
★ Final term Fall 2012Lecture 293 marks
What is the size of the datagram header?
Show answerJawab dekhein
The IP datagram header is 20 octets when it has no options; options can make it up to 60 octets (the header length field counts 32-bit words). The whole datagram, header plus data, can be at most 65,535 octets.
74.
★ Final term Fall 2012Lecture 133 marks
Can the length of an Ethernet be increased by adding a repeater?
Show answerJawab dekhein
Yes. A repeater joins two segments and regenerates the signal, so one repeater doubles and two repeaters triple the maximum cable length. It cannot be repeated without limit, because the standard allows only a few repeaters between any two stations to keep delay low enough for collision detection.
75.
★ Final term Fall 20122 marks
What is meant by client and server?
Show answerJawab dekhein
In the client-server model used by network applications, the server is the program that waits passively for requests and provides a service, and the client is the program that actively contacts the server to request the service.
76.
★ Final term 2012Lecture 142 marks
Is bridge is intelligent?
Show answerJawab dekhein
Yes. A bridge learns which computers are on each segment from the source addresses of frames, forwards a frame only when its destination is on another segment, filters local frames, and does not pass collisions or noise, unlike a simple repeater.
77.
★ Final term 20122 marks
What is meant by Zero Compression in IPv6?
Show answerJawab dekhein
In colon hex notation a run of consecutive zero groups can be replaced by two colons, written once in an address. For example FF0C:0:0:0:0:0:0:B1 is written FF0C::B1. This makes addresses with many zeros much shorter.
78.
★ Final term 20123 marks
Why three-way handshake technique is used by TCP?
Show answerJawab dekhein
TCP uses a three-way handshake (SYN, SYN+ACK, ACK) so that connection start-up is reliable: both sides agree that they want the connection and are ready, and each side tells the other its random 32-bit initial sequence number and has it acknowledged. Because new random sequence numbers are chosen, delayed or duplicate segments from an old connection (for example after a reboot) are not confused with the new one. A similar exchange closes connections gracefully.
79.
★ Final term 20123 marks
Why we need server?
Show answerJawab dekhein
A server provides a service (for example files, web pages or mail) to many clients. It is a special program that usually starts when the system boots and runs forever, waits passively for contact from clients, accepts requests from arbitrary clients and can handle several remote clients at the same time. It usually needs a powerful computer and operating system.
80.
★ Final term Fall 20112 marks
define jitter
Show answerJawab dekhein
Jitter is the variation (variance) in transmission delay between packets. It matters for real-time voice and video; in a LAN it occurs when some packets are delayed because the network is busy.
81.
★ Final term Fall 20112 marks
define TCP
Show answerJawab dekhein
TCP (Transmission Control Protocol) is the TCP/IP transport protocol that gives applications a reliable, connection-oriented, full-duplex byte stream between two computers, using IP to carry its segments.
82.
★ Final term Fall 20112 marks
what is meant by the client server paradigm ?
Show answerJawab dekhein
In the client-server model used by network applications, the server is the program that waits passively for requests and provides a service, and the client is the program that actively contacts the server to request the service.
83.
★ Final term Fall 2011Lecture 233 marks
why organization does not use single router
Show answerJawab dekhein
An organization seldom connects all its networks with one router for two reasons: the router must forward every packet, so a single router's processor cannot handle all the traffic; and using several routers gives redundancy, so the internet keeps working if one router fails.
84.
★ Final term Fall 2011Lecture 172 marks
Distance Vector Routing
Show answerJawab dekhein
In distance-vector routing each switch or router keeps a table of destinations with distance and next hop, and periodically sends its list of (destination, distance) pairs to its neighbours. A neighbour that learns a shorter path through the sender updates its table. It is simple and is used by RIP.
85.
★ Final term Fall 2011Lecture 302 marks
what stand for MTU, define
Show answerJawab dekhein
MTU stands for Maximum Transmission Unit: the largest amount of data a frame of a given hardware technology can carry in its data area. A datagram larger than the MTU must be fragmented.
86.
★ Final term Fall 20112 marks
why the Internet Multicast Routing is difficult
Show answerJawab dekhein
Because membership is dynamic and unrestricted: any computer may join or leave a group at any time, members can be anywhere in the Internet, and senders need not be members. Routers must therefore keep changing the delivery trees as members come and go, which is much harder than unicast routing.
87.
★ Final term Fall 20112 marks
What the basic function of Twice NAT?
Show answerJawab dekhein
Twice NAT is a variant of NAT used when a site runs servers that must be reached from the Internet. The NAT box works with the site's domain name server: when an outside host looks up a server name, the reply carries a public address and NAT creates a mapping, and NAT then translates both the source and the destination addresses of the datagrams.
88.
★ Final term Fall 20113 marks
Difference between Static and Dynamic Routing
Show answerJawab dekhein
In static routing the routing table is set up (by the administrator or at boot time) and does not change unless someone edits it manually, so it cannot react to failures. In dynamic routing the table is initialized at boot and then routing protocol software (such as RIP or OSPF) continually learns routes from other routers and updates the table automatically when the network changes.
89.
★ Final term Fall 2011Lecture 103 marks
How a receiver know that incoming data is datagram or other
Show answerJawab dekhein
Through the frame type field. Sender and receiver agree on the value used in the frame header type field for IP (for Ethernet 0x0800), so when a frame arrives with that type the receiver passes its data area to IP as a datagram.
90.
★ Final term Fall 2011Lecture 53 marks
what is data stuffing
Show answerJawab dekhein
Data (byte or bit) stuffing lets frame delimiters and control characters be distinguished from data. Before sending, the sender changes the data slightly: whenever a reserved character (such as the start or end of frame mark, or the escape character) appears in the data, it is replaced by an escape sequence. The receiver reverses the change, so the frame boundaries stay unambiguous and the original data is recovered.
91.
★ Final term Fall 20112 marks
What is means by "It provides facilities for Transit Routing."?
Show answerJawab dekhein
It is a BGP characteristic: BGP classifies each autonomous system as a transit system, if it agrees to pass traffic of other ASes through it, or as a stub system, if it does not. BGP lets an organization declare itself a stub even if it is multi-homed, so it refuses transit traffic.
92.
★ Final term Fall 20112 marks
Does OSPF support for multi access network?
Show answerJawab dekhein
Yes. OSPF supports multi-access (broadcast) networks such as Ethernet; it elects a designated router on such a network so that routers do not all exchange link-state information with each other.
93.
★ Final term Fall 2011Lecture 113 marks
What is difference in NIC and CPU Processing?
Show answerJawab dekhein
A network interface card contains enough hardware to handle frames independently of the system CPU: it sends and receives frames, checks addresses and errors and buffers data, and some NICs have their own microprocessor and use DMA to move data to memory. The CPU is involved only when a complete frame is ready (by interrupt), so network processing does not use CPU time for every bit.
94.
★ Final term Fall 20113 marks
What is Extension Header in Ipv6?
Show answerJawab dekhein
In IPv6, optional information is not kept in the base header; it is placed in optional extension headers that follow the base header and come before the data. Each extension header has a NEXT HEADER field naming the next one. Examples are the fragmentation and routing headers. Routers can process datagrams faster and new features can be added as new header types.
95.
★ Final term Fall 20113 marks
What is congestion control, How TCP Segment format is done?
Show answerJawab dekhein
Congestion control: its goal is to avoid adding retransmissions to an already congested network. TCP assumes that lost segments are due to congestion and reduces its window quickly when loss occurs, then increases it gradually. TCP segment format: TCP uses one format, the segment, for all messages, both data and acknowledgements. The header holds source and destination ports, sequence number, acknowledgement number, header length, code bits, window, checksum, urgent pointer and options, followed by the data.
96.
★ Final term Fall 2011Lecture 242 marks
Find the class in 00000001.001011.1001.111
Show answerJawab dekhein
The first bit is 0, which marks a Class A address.
97.
★ Final term Fall 20112 marks
What is the difference between unicast and multicast?
Show answerJawab dekhein
A unicast address and transmission is for a single destination computer; a multicast address and transmission is for a group of destinations, which may be at different sites, and one transmission reaches all members of the group.
98.
★ Final term Fall 2011Lecture 112 marks
What is the role of DMA in NIC?
Show answerJawab dekhein
With Direct Memory Access the NIC copies frame data directly from main memory when sending and directly into main memory when receiving, without the CPU copying each byte, so the CPU is free for other work.
99.
★ Final term Fall 2011Lecture 253 marks
Change the following into equivalent binary 154.31.161.13 202.32.15.7 192.168.1.5
TCP provides: connection-oriented service; point-to-point connections; complete reliability; full-duplex communication; a stream interface; reliable connection start-up; and graceful connection shutdown.
101.
★ Final term Fall 2011Lecture 133 marks
What are the distance limitations in Fiber Optic?
Show answerJawab dekhein
Optical fibre has very low signal loss and delay and very high bandwidth, so a fibre link can extend across several kilometres, far beyond the few hundred metres of copper LAN cable. It is used, with fibre modems, to connect sites or buildings far apart.
102.
★ Final term Fall 2011Lecture 42 marks
Limitations of parity checking?
Show answerJawab dekhein
Parity can detect only errors that change an odd number of bits. If an even number of bits are changed the parity still looks correct, e.g. data 10010001 with even parity bit 1 corrupted to 10110011 with parity bit 1 (two bits changed) is not detected. So parity is useful mainly for single-bit errors and cannot correct errors.
103.
★ Final term Fall 20112 marks
where should an ICMP message be sent?
Show answerJawab dekhein
An ICMP error message is sent back to the original source of the datagram that caused the problem (the source address in that datagram). ICMP messages are not sent about problems with ICMP error messages themselves.
104.
★ Final term Fall 20112 marks
what is the basic concept of twice NAT?
Show answerJawab dekhein
Twice NAT is a variant of NAT used when a site runs servers that must be reached from the Internet. The NAT box works with the site's domain name server: when an outside host looks up a server name, the reply carries a public address and NAT creates a mapping, and NAT then translates both the source and the destination addresses of the datagrams.
105.
★ Final term Fall 20113 marks
Could IP be redesigned to use hardware addresses instead of the 32-bit addresses it currently uses. Why or why not?
Show answerJawab dekhein
No. Hardware addresses differ in size and format between technologies, so they cannot give one uniform address for all hosts. They are flat (chosen by the manufacturer with no network part), so routers could not route by network prefix and would need an entry per host. They also change when an interface card is replaced. IP needs hierarchical, uniform, software-assigned addresses for routing across an internet.
106.
★ Final term Fall 2011Lecture 273 marks
Three features of dynamic message method in ARP.
Show answerJawab dekhein
With message exchange: the protocol address is independent of the hardware address, so either can change without reconfiguration; the hardware must support broadcast, because requests are broadcast; and the implementation is more difficult (requests, replies, timeouts and caching), and resolution takes longer than table lookup or computation.
107.
★ Final term Fall 20113 marks
why TCP is called end to end protocol?
Show answerJawab dekhein
Because it provides communication directly between an application on one computer and an application on another. TCP connections are virtual connections created by TCP software on the two end computers only; routers in the Internet just forward IP datagrams and give no support for the connection.
108.
★ Final term Spring 20102 marks
Define what is extension head in IPv6
Show answerJawab dekhein
In IPv6, optional information is not kept in the base header; it is placed in optional extension headers that follow the base header and come before the data. Each extension header has a NEXT HEADER field naming the next one. Examples are the fragmentation and routing headers. Routers can process datagrams faster and new features can be added as new header types.
109.
★ Final term Spring 20102 marks
What is implementation of NAT?
Show answerJawab dekhein
NAT can be implemented as a stand-alone hardware box between the site and its ISP, inside a router (often by the ISP), or as software on a computer that connects to the Internet and shares the connection (for example at home with DSL or cable). In each case the NAT device keeps a translation table: an entry is created when an outgoing datagram is seen, the private source address (and in NAPT the port) is replaced with the public one, and replies are translated back using the table.
110.
★ Final term Spring 20103 marks
Write a note on "limited connectivity" of Wireless LAN.
Show answerJawab dekhein
Unlike a wired LAN, in a wireless LAN not every station can reach every other one: signals are weak and fade with distance, and walls and other obstacles block them. So a station may not hear another station's transmission and cannot rely on collision detection; wireless LANs use collision avoidance (CSMA/CA) instead.
111.
★ Final term Spring 20103 marks
In internet routing how can a host join or leave group?
Show answerJawab dekhein
A host uses the Internet Group Management Protocol (IGMP) to tell a nearby multicast router about group membership. To join, the host sends an IGMP membership report for the group when its first application joins; the router then arranges for the group's datagrams to be forwarded onto that network. The router periodically queries the hosts to see whether members remain. When the last application on the host leaves the group, the host informs the router (or stops reporting), and when no members remain on the network the router stops forwarding that group's traffic there.
112.
★ Final term Spring 2010Lecture 343 marks
Provide three characteristics of UDP?
Show answerJawab dekhein
UDP is an end-to-end transport protocol that gives application-to-application communication. It provides connectionless service, is message-oriented (it keeps message boundaries), gives best-effort delivery with no guarantee against loss, duplication or reordering, allows arbitrary interaction (one-to-one, one-to-many, many-to-one), and is operating-system independent.
1.
★ Final term Fall 20115 marks
What is ICMP and what type of errors internet layer can detect?
Show answerJawab dekhein
ICMP (Internet Control Message Protocol) is the error-reporting and control companion of IP; its messages are carried inside IP datagrams and are sent back to the source of a problem datagram. Errors it reports include: source quench (router congested, datagram discarded), time exceeded (TTL reached zero, or reassembly timer expired), destination unreachable (network, host or port cannot be reached, or fragmentation needed but not allowed), redirect (host should use a different first-hop router) and parameter problem (bad header field). It also carries informational messages such as echo request/reply and address mask request/reply.
2.
★ Final term Fall 20115 marks
In which situation RIP support for default routers?
Show answerJawab dekhein
RIP can advertise a default route. When the routers inside an organization do not know routes to every external destination, the router that connects the organization to its ISP advertises a default route through RIP, and RIP installs it in each router. Any datagram whose destination is not in a router's table is then sent toward that border router.
3.
★ Final term Fall 20115 marks
Give pros and cons of static and dynamic routing.
Show answerJawab dekhein
Static routing: routes are entered when the system boots (manually) and do not change. Pros: simple, no routing protocol traffic or CPU overhead, predictable. Cons: does not adapt to link or router failures, must be changed by hand, does not scale to large networks. Dynamic routing: routing software (a routing protocol such as RIP or OSPF) learns routes and keeps updating the table. Pros: adapts automatically to failures and changes, scales to large internets. Cons: uses bandwidth, memory and CPU, is more complex, and routes may be unstable while the protocol converges.
4.
★ Final term Fall 20085 marks
What is the role of area in open shortest path first (OSPF)?
Show answerJawab dekhein
OSPF lets a manager divide an autonomous system into areas. Routers inside an area exchange full link-state information and each computes shortest paths for that area. Area border routers summarise the area's routes and send the summaries to other areas through the backbone area. This limits flooding of link-state messages, reduces memory and computation, hides internal changes from other areas and lets OSPF scale to large autonomous systems.
5.
★ Final term Fall 20085 marks
Write a note on Address Resolution.
Show answerJawab dekhein
To send a datagram over a physical network, the sender must map the next hop's IP (protocol) address to its hardware address; this is address resolution, and it is done only for computers on the same physical network. Three techniques exist: table lookup (bindings stored in a table), closed-form computation (hardware address computed from the protocol address with a function, used with configurable hardware addresses), and dynamic message exchange. On Ethernet, ARP uses message exchange: the sender broadcasts an ARP request containing the target IP address, the target replies with its hardware address, and the sender caches the binding to avoid repeated requests. ARP messages are carried directly in frames with their own frame type.
6.
★ Final term Fall 200810 marks
What is the concept of classes for IP Addresses? Briefly describe each class.
Show answerJawab dekhein
Classful addressing divides the IPv4 address space into classes, and the first bits of an address tell its class and where the prefix/suffix boundary lies (addresses are self-identifying). Class A: first bit 0, 8-bit prefix and 24-bit suffix, first octet 0-127, few networks with very many hosts. Class B: first bits 10, 16-bit prefix and 16-bit suffix, first octet 128-191. Class C: first bits 110, 24-bit prefix and 8-bit suffix, first octet 192-223, many networks with up to 254 hosts. Class D: first bits 1110, first octet 224-239, used for multicast. Class E: first bits 1111, first octet 240-255, reserved.
7.
★ Final term July 20125 marks
Is there a technique for achieving reliability through TCP?
Show answerJawab dekhein
TCP achieves reliability with positive acknowledgement and retransmission: every byte has a sequence number; the receiver sends acknowledgements for data received; the sender keeps a copy of unacknowledged data and retransmits it when an adaptive retransmission timer expires. A checksum detects corrupted segments, sequence numbers let the receiver reorder data and discard duplicates, a sliding window provides flow control, and a three-way handshake gives reliable connection startup and graceful shutdown.
8.
★ Final term July 20125 marks
Give reasons for which IPv4 need to be changed?
Show answerJawab dekhein
IPv4's 32-bit address space is running out as the Internet grows. New applications such as real-time audio and video need quality-of-service support that IPv4 lacks; the variable IPv4 header with options is slow to process; and IPv4 is hard to extend with new features. IPv6 addresses these with 128-bit addresses, a fixed base header with optional extension headers, flow labels for QoS, and easier extensibility.
9.
★ Final term July 20125 marks
What type of protocol designed for use within an organization?
Show answerJawab dekhein
Interior Gateway Protocols (IGPs) are designed for routing within one organization's autonomous system. Examples are RIP, a distance-vector protocol that uses hop count, runs over UDP and can advertise a default route, and OSPF, a link-state protocol that uses Dijkstra's algorithm, supports areas, authentication and manager-assigned costs. Exterior protocols such as BGP are used between organizations.
10.
★ Final term Spring 20105 marks
What are the characteristics of the Border Gateway Protocol?
Show answerJawab dekhein
BGP is the main exterior gateway protocol of the Internet. It provides routing among autonomous systems; it advertises reachability information including the path of autonomous systems to each destination; it provides policies so a manager can control which routes are advertised and accepted; it offers facilities for transit routing (transit vs stub ASes); it uses TCP for reliable transport; and it sends incremental updates after the initial exchange.
11.
★ Final term Spring 20105 marks
Describe how NAT can be useful for residence with cable modem or DSL?
Show answerJawab dekhein
A home with a cable modem or DSL line usually receives a single globally valid IP address from the ISP. A NAT device (often built into the home router or modem) gives every home computer a private address and translates addresses (and port numbers, NAPT) of outgoing datagrams to the single global address, and translates replies back to the right internal computer. All home computers can therefore share one address and one subscription, no extra addresses need to be bought, and internal computers are hidden from unsolicited outside connections.
12.
★ Final term Fall 2008Lecture 265 marks
Consider the IP address 178.200.127.5 and the corresponding subnet mask 255.255.255.0, then find out the following:
a. The number of bits used for subnetting
b. Total number of hosts in the subnet
c. The network address of the subnet
d. The subnet address of the IP address
Show answerJawab dekhein
178.200.127.5 is a class B address (default mask 255.255.0.0). a) The mask 255.255.255.0 adds 8 bits, so 8 bits are used for subnetting. b) 8 host bits remain: 2^8 - 2 = 254 hosts per subnet. c) ANDing the address with the mask gives the subnet's network address 178.200.127.0. d) The address therefore belongs to subnet 178.200.127.0 (subnet number 127 of class B network 178.200.0.0), and its host part is 5.
13.
★ Final term Fall 2008Lecture 175 marks
Write down the comparison of Distance-vector and Link-state algorithm?
Show answerJawab dekhein
Distance-vector: each router periodically sends its whole table of distances to its neighbours and updates its own table from what it receives; it is simple to implement but converges slowly and can suffer count-to-infinity; used in RIP. Link-state: each router tests its links and floods link-status messages to all routers; each router builds the complete topology and computes shortest paths independently with Dijkstra's algorithm; it is more complex and needs more memory and computation but converges faster and scales better; used in OSPF.
14.
★ Final term Fall 200810 marks
Describe in detail what is the purpose of the following table? What sort of information can be extracted?
First four bits of address | Table index in decimal | Class of address
0000 | 0 | A
0001 | 1 | A
0010 | 2 | A
0011 | 3 | A
0100 | 4 | A
0101 | 5 | A
0110 | 6 | A
0111 | 7 | A
1000 | 8 | B
1001 | 9 | B
1010 | 10 | B
1011 | 11 | B
1100 | 12 | C
1101 | 13 | C
1110 | 14 | D
1111 | 15 | E
Show answerJawab dekhein
The table lets IP software compute the class of a classful IPv4 address quickly: the first four bits of the address are used as an index (0-15) into the table and the entry gives the class. Indexes 0-7 (first bit 0) are class A, 8-11 (first bits 10) class B, 12-13 (110) class C, 14 (1110) class D and 15 (1111) class E. From the class one can extract where the prefix/suffix boundary lies (A: 8-bit prefix, B: 16-bit, C: 24-bit), and whether the address is unicast (A, B, C), multicast (D) or reserved (E). This works because classful addresses are self-identifying.
15.
★ Final term Fall 200810 marks
List down and describe at least five characteristics of Routing Information Protocol.
Show answerJawab dekhein
RIP is an interior gateway protocol used inside an autonomous system. It uses a hop-count metric, each network between source and destination counting as one hop. It uses a distance-vector algorithm, routers periodically sending their tables to neighbours. It sends all messages over UDP. It runs over LANs: version 1 uses hardware broadcast and version 2 allows multicast. It can advertise a default route so an organization can install a default route in every router. Hosts can listen passively to RIP messages to update their own tables.
16.
★ Final term Fall 20085 marks
As the Internet grew, the original Classful addressing scheme became a limitation, what is the designed solution.
Show answerJawab dekhein
The fixed class boundaries wasted addresses (class C too small, class B too large) and the address space ran short. The designed solutions were subnet addressing and classless addressing: an address mask (or CIDR /n notation) states where the prefix ends, so the boundary can be anywhere. Subnetting divides one network's suffix among several physical networks, and classless addressing (CIDR) assigns blocks of any size and lets routers aggregate routes. NAT and IPv6 were later added to further relieve the address shortage.
17.
★ Final term Fall 20085 marks
What are IPv6 ADDRESS NOTATION?
Show answerJawab dekhein
A 128-bit IPv6 address is too long for dotted decimal (it would need 16 numbers, e.g. 105.220.136.100.255.255.255.255.0.0.18.128.140.10.255.255). IPv6 instead uses colon hexadecimal: eight 16-bit groups written in hex and separated by colons, e.g. 69DC:8864:FFFF:FFFF:0:1280:8C0A:FFFF. Zero compression replaces one run of zero groups with two colons, e.g. FF0C:0:0:0:0:0:0:B1 becomes FF0C::B1. An IPv6 address with 96 leading zero bits holds an IPv4 address.
18.
★ Final term Fall 200810 marks
List some characteristics of a client.
Show answerJawab dekhein
A client is an arbitrary application program that becomes a client temporarily when it needs remote access, and it can also perform other computation. It is invoked directly by a user and runs locally on the user's computer. It actively initiates contact with a server. It can access multiple services as needed but usually contacts one server at a time. It does not need special hardware or a sophisticated operating system.
19.
★ Final term past paperLecture 3410 marks
Define the characteristics of UDP.
Show answerJawab dekhein
UDP is an end-to-end transport protocol giving application-to-application communication through protocol port numbers. It provides connectionless service with no connection setup; it is message-oriented, so each message is sent as one datagram and data boundaries are preserved; it uses best-effort delivery, so messages may be lost, duplicated or delivered out of order with no acknowledgement, retransmission or flow control; it allows arbitrary interaction (one-to-one, one-to-many, many-to-one, many-to-many); it is independent of the operating system; and it has a small 8-byte header (source port, destination port, length, optional checksum) carried in IP.
20.
★ Final term Fall 20105 marks
As the Internet grew, the original Classful addressing scheme became a limitation, what is was the designed solution.
Show answerJawab dekhein
Classful addressing forced every network to take a Class A, B or C block, so many addresses were wasted and the address space was being exhausted. Two mechanisms were designed: subnet addressing, which lets a site divide one classful network into several subnets using an address mask, and classless addressing (CIDR), which drops the fixed classes and lets the prefix/suffix boundary fall on any bit, so an ISP can give a customer a block of the size it needs (for example a /28 for up to 14 hosts).
21.
★ Final term 20115 marks
What are the main advantages and disadvantages of routing information protocol (RIP)?
Show answerJawab dekhein
Advantages: RIP is simple to understand, configure and deploy; it is widely supported; it can propagate a default route; and hosts can listen passively to update their tables. Disadvantages: its only metric is hop count, so it ignores delay and bandwidth; the maximum is 15 hops (16 means unreachable), so it does not scale to large networks; it converges slowly after changes and can suffer count-to-infinity; and its periodic broadcast of whole tables adds traffic.
22.
★ Final term 2011Lecture 275 marks
Network engineer has three address resolution methods. How many methods does TCP/IP support in a real environment? Write names of methods and support your answer with solid reason?
Show answerJawab dekhein
TCP/IP can use all three methods: table lookup (bindings stored in a table in memory), closed-form computation (the hardware address is computed from the IP address, possible when hardware addresses are configurable) and dynamic message exchange (a request is broadcast and the owner replies, as ARP does on Ethernet). Which one is used depends on the addressing scheme of the underlying hardware, so TCP/IP supports all of them rather than one.
23.
★ Final term 20115 marks
Have any technique used for achieving reliability in TCP?
Show answerJawab dekhein
Yes. TCP achieves reliability over the unreliable IP service by: sequence numbers on every byte so data can be ordered and duplicates removed; positive acknowledgements from the receiver; retransmission of data whose acknowledgement does not arrive before a timer expires; adaptive retransmission, where the timeout is computed from measured round-trip times; a checksum to detect corrupted segments; a sliding window for flow control; congestion control; and a three-way handshake for reliable connection start-up and graceful shutdown.
24.
★ Final term 2011
Why EGP not use routing metric?
Show answerJawab dekhein
An exterior gateway protocol connects autonomous systems that are run by different organizations, each of which may measure cost in its own way. Metric values from different autonomous systems are not comparable, so an EGP advertises only reachability (and, in BGP, the path of autonomous systems) and leaves route choice to policies instead of a shared metric.
25.
★ Final term 20125 marks
Is there any technique for achieving reliability through TCP?
Show answerJawab dekhein
Yes. TCP achieves reliability over the unreliable IP service by: sequence numbers on every byte so data can be ordered and duplicates removed; positive acknowledgements from the receiver; retransmission of data whose acknowledgement does not arrive before a timer expires; adaptive retransmission, where the timeout is computed from measured round-trip times; a checksum to detect corrupted segments; a sliding window for flow control; congestion control; and a three-way handshake for reliable connection start-up and graceful shutdown.
26.
★ Final term 20125 marks
In a star organization there are 120 systems connected in a network. Give your comments about delay, delay should be smaller or larger. Give reasons?
Show answerJawab dekhein
Delay should be kept as small as possible. In a star network every frame passes through the central hub or switch, so with 120 systems the queuing delay at the centre grows as more stations send at once, and propagation plus switching delay add to it. High delay slows interactive applications and acknowledgements and lowers effective throughput, so the engineer should keep delay low, for example with a fast switch, enough capacity at the centre and by splitting the network into smaller segments.
27.
★ Final term 20125 marks
describe characteristics of BGP .
Show answerJawab dekhein
BGP is the main exterior gateway protocol of the Internet. It provides routing among autonomous systems; it allows policies that control which routes are advertised and accepted; it sends the path of autonomous systems for each destination (path vector); it uses TCP for reliable transport; it is the EGP of choice today, the current version being BGP-4; and it provides facilities for transit routing, classifying each AS as transit or stub.
28.
★ Final term 20125 marks
describe IPV6 addressing notation .
Show answerJawab dekhein
IPv6 addresses are 128 bits. Written in dotted decimal they would need 16 numbers, e.g. 105.220.136.100.255.255.255.255.0.0.18.128.140.10.255.255, which is unwieldy. So IPv6 uses colon hexadecimal: eight groups of 16 bits in hex separated by colons, e.g. 69DC:8864:FFFF:FFFF:0:1280:8C0A:FFFF. Zero compression replaces one run of zero groups with two colons, so FF0C:0:0:0:0:0:0:B1 becomes FF0C::B1. An IPv6 address with 96 leading zero bits is interpreted as holding an IPv4 address.
29.
★ Final term 20125 marks
have there is a technique for achieving reliability through TCP .
Show answerJawab dekhein
Yes. TCP achieves reliability over the unreliable IP service by: sequence numbers on every byte so data can be ordered and duplicates removed; positive acknowledgements from the receiver; retransmission of data whose acknowledgement does not arrive before a timer expires; adaptive retransmission, where the timeout is computed from measured round-trip times; a checksum to detect corrupted segments; a sliding window for flow control; congestion control; and a three-way handshake for reliable connection start-up and graceful shutdown.
30.
★ Final term Fall 20125 marks
Compare IPv6 with IPv4.
Show answerJawab dekhein
Address size: IPv4 uses 32-bit addresses, IPv6 128-bit addresses. Address types: IPv4 has unicast, multicast and broadcast; IPv6 has unicast, multicast and anycast (cluster) with no broadcast. Header: IPv4 header is variable (20-60 bytes) with options inside it; IPv6 has a fixed 40-byte base header with optional extension headers. Configuration: IPv4 hosts usually need manual or DHCP setup; IPv6 supports automatic configuration. IPv6 adds flow labels for quality of service, fragmentation only by the source, and is easier to extend. Notation: IPv4 uses dotted decimal, IPv6 colon hexadecimal with zero compression.
31.
★ Final term 20125 marks
Difference b/w PIM-SM and PIM-DM
Show answerJawab dekhein
PIM-SM (sparse mode) uses the same approach as CBT: it builds a shared multicast tree from a central point (rendezvous router) and routers join it explicitly, so traffic only goes where members ask for it; it suits groups whose members are few and widely spread. PIM-DM (dense mode) is meant for use within an organization: routers flood multicast packets to all locations, and routers that have no members send prune messages to stop the flow; it needs no setup before sending and suits short-lived sessions with many members. Both are protocol independent: they use whatever unicast routing exists instead of depending on one routing protocol.
32.
★ Final term 20125 marks
Describe NAT using at Home.
Show answerJawab dekhein
At a home with DSL or cable modem service the ISP gives one public IP address, but there may be several computers. NAT lets all of them share that address: the home computers use private addresses, and a NAT device (a dedicated box/home router, or NAT software on one PC that is also connected to the Internet) replaces the private source address (and port, in NAPT) with the public one on outgoing datagrams, records the mapping in a table, and translates replies back to the right computer. The household therefore needs no extra IP addresses.
33.
★ Final term Fall 20115 marks
how congestion control by tcp?
Show answerJawab dekhein
TCP assumes that lost segments are caused by congestion. When loss is detected it shrinks its congestion window quickly (to one segment) so it does not add retransmissions to an already congested network. It then uses slow start, growing the window with each acknowledgement, and after reaching a threshold grows it slowly (congestion avoidance). The sender transmits no more than the smaller of the congestion window and the receiver's advertised window.
34.
★ Final term Fall 20115 marks
IPv6 addressing
Show answerJawab dekhein
IPv6 uses 128-bit addresses, each made of a network prefix and a host suffix; there are no address classes, so the prefix/suffix boundary can fall anywhere. Special address types: unicast (one destination computer), multicast (a set of destinations, possibly at different sites) and cluster/anycast (a set of computers sharing a prefix; the datagram is delivered to one of them). Addresses are written in colon hexadecimal with zero compression, and an address with 96 leading zeros holds an IPv4 address.
35.
★ Final term Fall 20115 marks
Write a note on IPV6 Addressing.
Show answerJawab dekhein
IPv6 uses 128-bit addresses, each made of a network prefix and a host suffix; there are no address classes, so the prefix/suffix boundary can fall anywhere. Special address types: unicast (one destination computer), multicast (a set of destinations, possibly at different sites) and cluster/anycast (a set of computers sharing a prefix; the datagram is delivered to one of them). Addresses are written in colon hexadecimal with zero compression, and an address with 96 leading zeros holds an IPv4 address.
36.
★ Final term Fall 2011Lecture 345 marks
What are the characteristics of UDP
Show answerJawab dekhein
UDP is an end-to-end transport protocol that gives application-to-application communication. It provides connectionless service, is message-oriented (it keeps message boundaries), gives best-effort delivery with no guarantee against loss, duplication or reordering, allows arbitrary interaction (one-to-one, one-to-many, many-to-one), and is operating-system independent.
37.
★ Final term Fall 20115 marks
What are multicast routing protocols? Give names of 5 of them.
Show answerJawab dekhein
Multicast routing protocols let routers build and maintain delivery trees so multicast datagrams reach all members of a group. Five examples: Distance Vector Multicast Routing Protocol (DVMRP), Core Based Trees (CBT), Protocol Independent Multicast - Sparse Mode (PIM-SM), Protocol Independent Multicast - Dense Mode (PIM-DM), and Multicast extensions to OSPF (MOSPF).
38.
★ Final term Fall 2011Lecture 345 marks
Message Oriented in UDP
Show answerJawab dekhein
UDP offers applications a message-oriented interface: each message the application sends is sent as one UDP datagram, and UDP neither divides a message into packets nor combines messages. Advantage: applications can rely on UDP to preserve message boundaries. Disadvantages: each UDP message must fit into a single IP datagram, and very small or very large messages can use the underlying network inefficiently.
39.
★ Final term Fall 20115 marks
There are so many multicast protocol, Name only five one of those?
Show answerJawab dekhein
Multicast routing protocols let routers build and maintain delivery trees so multicast datagrams reach all members of a group. Five examples: Distance Vector Multicast Routing Protocol (DVMRP), Core Based Trees (CBT), Protocol Independent Multicast - Sparse Mode (PIM-SM), Protocol Independent Multicast - Dense Mode (PIM-DM), and Multicast extensions to OSPF (MOSPF).
40.
★ Final term Fall 20115 marks
What is Base Header in IPv6?
Show answerJawab dekhein
Every IPv6 datagram starts with a fixed-size 40-octet base header containing version, traffic class, flow label, payload length, next header, hop limit, and the 128-bit source and destination addresses. The NEXT HEADER field at the end of the base header gives the type of the header that follows (an extension header or the data, e.g. TCP). Extension headers, some of variable size, also carry a NEXT HEADER field, forming a chain.
41.
★ Final term Fall 20115 marks
If IPV4 works so well. Why change it?
Show answerJawab dekhein
Address space: 32-bit addresses are being exhausted; most networks are Class C and too small for many organizations, and Class B addresses are almost used up. Type of service: applications such as real-time audio and video need different delivery guarantees, but IPv4 type of service is rarely implemented. Multicast and other features are not well supported. The scale of the Internet, link speeds and frame sizes have grown greatly. These factors led to IPv6.
42.
★ Final term Fall 20115 marks
Main advantages and disadvantages of Routing Information Protocol.
Show answerJawab dekhein
Advantages: RIP is simple to understand, configure and deploy; it is widely supported; it can propagate a default route; and hosts can listen passively to update their tables. Disadvantages: its only metric is hop count, so it ignores delay and bandwidth; the maximum is 15 hops (16 means unreachable), so it does not scale to large networks; it converges slowly after changes and can suffer count-to-infinity; and its periodic broadcast of whole tables adds traffic.
43.
★ Final term Spring 2010Lecture 345 marks
What is meant by massage oriented interface in UDP also give the advantages and disadvantages of interface
Show answerJawab dekhein
UDP offers applications a message-oriented interface: each message the application sends is sent as one UDP datagram, and UDP neither divides a message into packets nor combines messages. Advantage: applications can rely on UDP to preserve message boundaries. Disadvantages: each UDP message must fit into a single IP datagram, and very small or very large messages can use the underlying network inefficiently.
44.
★ Final term Spring 20105 marks
What is the role of OSPF?
Show answerJawab dekhein
OSPF (Open Shortest Path First) is an interior gateway protocol that routes within an autonomous system using the link-state algorithm. Each router floods the status of its links, builds the topology and computes shortest paths with Dijkstra. It carries a 32-bit mask with each address (so classless and subnetted addresses work), supports multi-access networks, authenticates messages, and lets the AS be divided into areas so that it scales to large networks.
45.
★ Final term Fall 20085 marks
What is IPv6 ADDRESS NOTATION?
Show answerJawab dekhein
IPv6 addresses are 128 bits. Written in dotted decimal they would need 16 numbers, e.g. 105.220.136.100.255.255.255.255.0.0.18.128.140.10.255.255, which is unwieldy. So IPv6 uses colon hexadecimal: eight groups of 16 bits in hex separated by colons, e.g. 69DC:8864:FFFF:FFFF:0:1280:8C0A:FFFF. Zero compression replaces one run of zero groups with two colons, so FF0C:0:0:0:0:0:0:B1 becomes FF0C::B1. An IPv6 address with 96 leading zero bits is interpreted as holding an IPv4 address.
46.
★ Final term Fall 2008Lecture 2410 marks
Describe in detail what is the purpose of the following table? What sort of information can be extracted? (Table: First Four Bits of address / Table index in decimal / Class of Address: 0000 0 A; 0001 1 A; 0010 2 A; 0011 3 A; 0100 4 A; 0101 5 A; 0110 6 A; 0111 7 A; 1000 8 B; 1001 9 B; 1010 10 B; 1011 11 B; 1100 12 C; 1101 13 C; 1110 14 D; 1111 15 E)
Show answerJawab dekhein
Whenever it handles a packet, IP software must split the destination address into prefix and suffix, so it must first know the address class. Classful addresses are self-identifying: the class is computed from the address itself. The table lets software find the class quickly by using the first four bits of the address as an index (0-15) into a 16-entry table instead of testing bits one by one. Information extracted: indexes 0-7 (first bit 0) are Class A; 8-11 (first bits 10) Class B; 12-13 (110) Class C; 14 (1110) Class D (multicast); 15 (1111) Class E (reserved). From the class the software then knows the prefix length (8, 16 or 24 bits for A, B, C) and so the network and host parts.