CS6250 Computer Networks Exam 1
● Describe the relationships among ISPs, IXPs, and CDNs. - ANS-The Internet is a complex
environment. It is built with a community of networks. These networks encompass (ISPs)
Internet Service Providers, (IXPs) Internet Exchange Points, (CDNs) Content Delivery
Networks. In 2019, there were apx. 500 IXPs round the arena.
Huge scale Tier-1 ISPs that function at a worldwide scale, and basically they shape the
"spine" community over which smaller networks can join
IXPs are interconnection infrastructures, which offer the physical infrastructure, wherein a
couple of networks (eg ISPs and CDNs) can interconnect and alternate site visitors
regionally.
CDNs are networks which can be created with the aid of content providers with the goal of
getting extra manage of how the content is delivered to the give up-users, and also to lessen
connectivity expenses. Some example CDNs encompass Google and Netflix. These
networks have more than one facts centers - and each certainly one of them can be housing
masses of servers - which are disbursed internationally.
● Explain Additive Increase/Multiplicative Decrease (AIMD) inside the context of TCP. -
ANS-TCP decreases the window when the level of congestion goes by halving the window
length, and it will increase the window whilst the extent of congestion goes down by adding
to the window size. This causes convergence to the gold standard bandwidth by means of
quickly cutting use in times of congestion at the same time as slowly increasing usage while
the congestion clears.
The concept in the back of additive boom is to increase the window via one packet each
RTT (Round Trip Time).
Once TCP Reno detects congestion, it reduces the price at which the sender transmits. So,
while the TCP sender detects that a timeout occurred, then it sets the CongestionWindow
(cwnd) to half of of its previous price.
● Explain how TCP CUBIC works. - ANS-CUBIC makes use of a cubed polynomial as its
increase characteristic. To maintain the TCP equity it uses a multiplicative lower and
decreases the window to half of.
TCP CUBIC is fair regardless of RTT due to the fact the calculation that is used relies upon
on the elapsed time from two congestion events, rather than being dependent on the RTT of
a connection.
● Explain TCP throughput calculation. - ANS-The calculation is:
P = Probability of packet loss
MSS = Maximum Segment Size
RTT = Round-Trip Times
BW = information consistent with cycle / time consistent with cycle
BW = MSS/RTT * C / sqrt(P)
, ● How does a bunch infer congestion? - ANS-The host infer congestion from the network
behavior particularly via 2 alerts:
First is the packet postpone. As the community receives congested, the queues in the router
buffers increase. This ends in extended packet delays. Thus, an increase inside the
spherical-trip time, which can be expected primarily based on ACKs, may be a trademark of
congestion within the network. However, it turns out that packet put off in a network tends to
be variable, making delay-based totally congestion inference quite complex.
Another signal for congestion is packet loss. As the network gets congested, routers begin
dropping packets. Note that packets can also be misplaced due to different reasons which
include routing mistakes, hardware failure, TTL expiry, mistakes inside the hyperlinks, or go
with the flow control issues, although it is uncommon.
● How does a course server work? - ANS-A path server:
● Collects and shares routing statistics from its peers or members that connects with (i.E.
IXP individuals that connect with the RS).
● Executes its personal BGP decision system and also re-put it up for sale the ensuing facts
(I.E. Excellent route choice) to all RS's peer routers.
● Also referred to as Multi-lateral BGP peer periods.
● How does a router use the BGP selection manner to choose which routes to import? -
ANS-The router compares routes by means of going via a listing of attributes and chooses
the path with the bottom amount of hops, in the simplest situation. The router makes use of a
choice tool known as LocalPref attribute that's the preferred course discovered via a
particular AS over different ASes. An AS ranks routes via first who prefer the routes
discovered from its clients, then the routes learned from its friends, and finally from the
routes found out from its vendors. Another characteristic that may have an effect on routing
choices is the MED (Multi-Exit Discriminator) characteristic. This is utilized by ASes linked
through more than one links to designate which of those links are desired for inbound
visitors.
● How does a TCP sender restriction the sending price? - ANS-TCP makes use of a
congestion window that's much like the obtain window used for drift manage. It represents
the maximum number of unacknowledged data that a sending host can have in transit (sent
but not but acknowledged).
TCP uses a probe-and-adapt technique in adapting the congestion window. Under ordinary
situations, TCP increases the congestion window looking to reap the available throughput.
Once it detects congestion then the congestion window is decreased.
In the cease, the quantity of unacknowledged facts that a sender may have is the minimum
of the congestion window and the receive window.
● How does an AS decide what policies to import/export? - ANS-AS commercial enterprise
relationships force an AS' routing regulations and have an impact on which routes an AS
wishes to import or export.
There are 3 selective transit routes: Transit client routes, Transit Provider Routes, Peer
Routes. The major rule is monetary incentive. Transit purchaser routes > peer routes >
transit company routes.
● Describe the relationships among ISPs, IXPs, and CDNs. - ANS-The Internet is a complex
environment. It is built with a community of networks. These networks encompass (ISPs)
Internet Service Providers, (IXPs) Internet Exchange Points, (CDNs) Content Delivery
Networks. In 2019, there were apx. 500 IXPs round the arena.
Huge scale Tier-1 ISPs that function at a worldwide scale, and basically they shape the
"spine" community over which smaller networks can join
IXPs are interconnection infrastructures, which offer the physical infrastructure, wherein a
couple of networks (eg ISPs and CDNs) can interconnect and alternate site visitors
regionally.
CDNs are networks which can be created with the aid of content providers with the goal of
getting extra manage of how the content is delivered to the give up-users, and also to lessen
connectivity expenses. Some example CDNs encompass Google and Netflix. These
networks have more than one facts centers - and each certainly one of them can be housing
masses of servers - which are disbursed internationally.
● Explain Additive Increase/Multiplicative Decrease (AIMD) inside the context of TCP. -
ANS-TCP decreases the window when the level of congestion goes by halving the window
length, and it will increase the window whilst the extent of congestion goes down by adding
to the window size. This causes convergence to the gold standard bandwidth by means of
quickly cutting use in times of congestion at the same time as slowly increasing usage while
the congestion clears.
The concept in the back of additive boom is to increase the window via one packet each
RTT (Round Trip Time).
Once TCP Reno detects congestion, it reduces the price at which the sender transmits. So,
while the TCP sender detects that a timeout occurred, then it sets the CongestionWindow
(cwnd) to half of of its previous price.
● Explain how TCP CUBIC works. - ANS-CUBIC makes use of a cubed polynomial as its
increase characteristic. To maintain the TCP equity it uses a multiplicative lower and
decreases the window to half of.
TCP CUBIC is fair regardless of RTT due to the fact the calculation that is used relies upon
on the elapsed time from two congestion events, rather than being dependent on the RTT of
a connection.
● Explain TCP throughput calculation. - ANS-The calculation is:
P = Probability of packet loss
MSS = Maximum Segment Size
RTT = Round-Trip Times
BW = information consistent with cycle / time consistent with cycle
BW = MSS/RTT * C / sqrt(P)
, ● How does a bunch infer congestion? - ANS-The host infer congestion from the network
behavior particularly via 2 alerts:
First is the packet postpone. As the community receives congested, the queues in the router
buffers increase. This ends in extended packet delays. Thus, an increase inside the
spherical-trip time, which can be expected primarily based on ACKs, may be a trademark of
congestion within the network. However, it turns out that packet put off in a network tends to
be variable, making delay-based totally congestion inference quite complex.
Another signal for congestion is packet loss. As the network gets congested, routers begin
dropping packets. Note that packets can also be misplaced due to different reasons which
include routing mistakes, hardware failure, TTL expiry, mistakes inside the hyperlinks, or go
with the flow control issues, although it is uncommon.
● How does a course server work? - ANS-A path server:
● Collects and shares routing statistics from its peers or members that connects with (i.E.
IXP individuals that connect with the RS).
● Executes its personal BGP decision system and also re-put it up for sale the ensuing facts
(I.E. Excellent route choice) to all RS's peer routers.
● Also referred to as Multi-lateral BGP peer periods.
● How does a router use the BGP selection manner to choose which routes to import? -
ANS-The router compares routes by means of going via a listing of attributes and chooses
the path with the bottom amount of hops, in the simplest situation. The router makes use of a
choice tool known as LocalPref attribute that's the preferred course discovered via a
particular AS over different ASes. An AS ranks routes via first who prefer the routes
discovered from its clients, then the routes learned from its friends, and finally from the
routes found out from its vendors. Another characteristic that may have an effect on routing
choices is the MED (Multi-Exit Discriminator) characteristic. This is utilized by ASes linked
through more than one links to designate which of those links are desired for inbound
visitors.
● How does a TCP sender restriction the sending price? - ANS-TCP makes use of a
congestion window that's much like the obtain window used for drift manage. It represents
the maximum number of unacknowledged data that a sending host can have in transit (sent
but not but acknowledged).
TCP uses a probe-and-adapt technique in adapting the congestion window. Under ordinary
situations, TCP increases the congestion window looking to reap the available throughput.
Once it detects congestion then the congestion window is decreased.
In the cease, the quantity of unacknowledged facts that a sender may have is the minimum
of the congestion window and the receive window.
● How does an AS decide what policies to import/export? - ANS-AS commercial enterprise
relationships force an AS' routing regulations and have an impact on which routes an AS
wishes to import or export.
There are 3 selective transit routes: Transit client routes, Transit Provider Routes, Peer
Routes. The major rule is monetary incentive. Transit purchaser routes > peer routes >
transit company routes.
