Congestion and its aftereffects, Root cause of congestion and congestion control algorithms

1. Congestión Control
Presented By:
Naveen Kr. Dubey
NITTTR, Chandigarh

2. Congestion…?
When t oo
many packet s
are present in
(a part of ) t he
subnet ,
perf ormance
degrades . This
sit uat ion is
called Congestion.

3. Network Congestion(Cont’d)
 Packet dumped by hosts on
subnet within the carrying
capacity.
 Almost 100% delivery
 No of delivered packet is
proportional to no of packet sent.
 Packet dumped on subnet to far
from carrying capacity
 Router are no longer able to cope.
 Packet losing starts.
 At very high traffic , performance
collapse completely and almost
no packet delivered.

4. Network Congestion(Cont’d)

5. Factors Influencing ?
 If all of sudden, stream of packet begin arriving on three or four
input lines and all needs the same output line, A queue will build
up.
 If there is insufficient memory to hold all of them, packets
will get lost.
Nagle J: “Congestion control in TCP/IP Internetworks,” Computer
Commun. Rev. , vol 14, pp. 11-17, April 1987
 If the router have infinite amount of memory, congestion gets
worse, not better, because by the time packet get to the front of
the queue, they have already timed out. Duplicate have been
sent which further increase traffic load.

6. Factors Influencing ?(Cont’d)
 Slow processor can also cause congestion .
 If CPU is slow in book keeping task
(like queuing buffers, updating tables etc.)
 Queue build up even there is excess line capacity.
 Low bandwidth line also cause congestion.

7. Congestion leads to feed upon itself
 Congestion tends to feed upon itself to get even worse. Routers
respond to overloading by dropping packets. When these packets
contain TCP segments, the segments don't reach their destination,
and they are therefore left unacknowledged.
 which eventually leads to timeout and retransmission.
 The major cause of congestion is often the bursty nature of traffic.
 If the hosts could be made to transmit at a uniform rate, then
congestion problem will be less common and all other causes will
not even led to congestion
 because other causes just act as an enzyme which boosts up the
congestion when the traffic is bursty

8. Flow Control & Congestion Control
 There is subtle relation between Congestion control and Flow
control.
 Objective of Congestion control is to ensure the subnet is
able to carry the offered traffic.
 It is a global issue involving the behavior of all hosts, all the
routers, the store and forward processing in routers and all other
parameters tending to diminish the capacity of subnet.
 Flow control is concerned about point-to-point link between
sender and receiver .
 Its job is to make sure that a fast sender can not transmit data
faster than the receiver can absorb it.
 It always involve some direct feedback .

9.  The Congestion control and Flow control are often confused
because some congestion control algorithms operate by sending
message back to the various sources telling them to slow down
when the network gets into trouble.
 Thus a host get a “slow down” message either because the
receiver can not handle the load, or because network can’t
handle it.

10. Principles of Congestion Control
 Congestion control refers to technique and mechanism that can
either prevent congestion, before it happens, or remove
congestion, after it has happened.
 The presence of congestion means that the load is (temporarily)
greater than the resource (in part of system) can handle it.
 Two solution comes in mind: increase the resource or decrease
the load.
Many congestion control algorithms are known

11. Yang and Reddy “Taxonomy for congestion control Algorithms in
packet switching network”IEEE network magazine vol.9 pp.34-45,
Jul/Aug 1995
Forouzan/ Datat comm networks / Fig 24.5
Control theory viewpoint: dividing all the solutions to two groups

12. Principles of Congestion Control
(Open Loop)
 Tools for doing open loop control include
deciding.
 When to accept new traffic ?
 When to discard packet and which one ?
 Making scheduling decision at various point in the network .
All of these have in common the fact that
they make decision without regard to
the current state of the network.

13. Principles of Congestion Control
(Closed loop)
Closed loop solutions are based on the concept of
a feedback loop.
 This approach includes three parts when applied to congestion
control
 Monitor the system when and where the congestion occurs .
 Pass this information to place where action can be taken.
 Adjust system operation to correct the problem.

14. Principles of Congestion Control
 How? Good design
 Make sure the problem does
not occur
 Tools
 Decide when to accept traffic
 Decide when to discard
packets and which ones
 Make scheduling decisions
in the subnet
 The Open loop algorithms
further divided into
 One that act at source
 One that act at destination
 Once Communication process
starts, then open loop can not
handle any new born
congestion problem .
 Monitor: where and when congestion?
 % packets discarded
 average queue length
 number of packets that time out
 average packet delay
Rising no. Indicate growing
congestion
 Pass collected info to places where
actions can be taken = source of traffic
 explicit
 implicit feedback
Adjust system operation
 Increase resources: bandwidth
 Decrease load: deny, degrade service
Control theory viewpoint: dividing all the solutions to two groups
open loop closed loop: feedback loop

15. Cont’d…
 In implicitalgorithms the source deduce the existence of
congestion by making local observations, such as time needed for
acknowledgment to come back.
(when router detect this is congested state it fills in the field of all
outgoing packets to warn the neighbors )
 In explicit algorithms host or router send probe packets out
periodically to ask about congestion
( For some radio station helicopter flying around the city to
update traffic info)

16. Principles of Congestion Control
(Time Scale Adjustment )
 Time scale must be adjusted carefully, to work well some kind
of averaging needed but getting the time constant right is a
non trivial matter .
Example:
Suppose there is certain set of instruction
1. When two packet comes in a row:
Router yells STOP.
2. Every time router idle for 20 µs : yells GO
(System will oscillate wildly and never converge )
On the other hand : If waits for 30 minute to make sure before
saying anything
(system will react sluggishly for any real use)

17. Congestion: prevention Policies
 open loop solutions: Minimize congestion, they try to achieve there goals by
using appropriate policies at various levels
Layer Policies
Transport  Retransmission policy
 Out-of-order caching policy
 Acknowledgement policy
 Flow control policy
 Timeout determination ( transit time over the network is hard to predict)
Network  Virtual circuits <> datagrams in subnet( many cog. Control algo work only with VC)
 Packet queueing and service policy ( 1 Q / input/output line and round robin)
 Packet discard policy
 Routing algorithm ( spreading traffic over all lines)
 Packet lifetime management
Data link  Retransmission policy( Go back N will put heavy load than Selective Reject)
 Out-of-order caching policy ( Selective repeat is better )
 Acknowledgement policy( piggyback onto reverse traffic )
 Flow control policy ( small window reduce traffic and thus congestion )

18. Types of Congestion Control
 Preventive
 The hosts and routers attempt to prevent congestion before it
can occur
 Reactive
 The hosts and routers respond to congestion after it occurs and
then attempt to stop it
 Preventive Techniques:
 Resource reservation
 Leaky/Token bucket
 Reactive Techniques:
 Load shedding
 Choke packets

19. Traffic-Aware Routing
 To make the most of the existing network capacity, routes can be
tailored to traffic patterns that change during the day as network
users wake and sleep in different time zones.
 For example:
routes may be changed to shift traffic away from heavily
used paths by changing the shortest path weights.
 This is called traffic-aware routing. Splitting traffic across multiple
paths is also helpful.

20. Traffic-Aware Routing
 The routing schemes we looked at in used fixed link weights.
 These schemes adapted to changes in topology, but not to changes
in load.
 The goal in taking load into account when computing routes is to
shift traffic away from hotspots that will be the first places in the
network to experience congestion.
 The most direct way to do this is to set the link weight to be a
function of the(fixed) link bandwidth and propagation delay plus
the (variable) measured load or average queuing delay. Least-
weight paths will then favour paths that are more lightly loaded, all
else being equal.

21. Traffic-Aware Routing (cont’d)
 Traffic-aware routing was used in the early Internet.
Khanna and Zinky J, “The revised ARPANET Routing Metric”, Proc.
SIGCOMM ‘89 Conf, ACM pp 45-56, 1989
For Example : Suppose A network shown in figure

22. Traffic-Aware Routing (cont’d)
 To avoid routing oscillation: Two technique can contribute to
successful solution.
1. The first is multipath routing, in which there can be multiple paths
from a source to a destination. In our example this means that the
traffic can be spread across both of the East to West links.
2. The second one is for the routing scheme to shift traffic across
routes slowly enough that it is able to converge.
Given these difficulties, in the Internet routing protocols do not generally
adjust their routes depending on the load. Instead, adjustments are made
outside the routing protocol by slowly changing its inputs. This is called
traffic engineering.

23. Admission Control
 One technique that is widely used in virtual-circuit networks to
keep congestion at bay is admission control.
 The idea is simple: do not set up a new virtual circuit unless the
network can carry the added traffic without becoming congested.
 Thus, attempts to set up a virtual circuit may fail. This is better than
the alternative, as letting more people in when the network is busy
just makes matters worse.
 By analogy, in the telephone system, when a switch gets overloaded
it practices admission control by not giving dial tones.

25. Admission Control (cont’d)
 The task is straightforward in the telephone network because of the
fixed bandwidth of calls (64 kbps for uncompressed audio).
 virtual circuits in computer networks come in all shapes and sizes.
 Thus, the circuit must come with some characterization of its traffic
if we are to apply admission control.

26. Admission Control (cont’d)
(Traffic Descriptor)
 Traffic is often described in terms of its rate and shape.
 The main focus of congestion control and quality of service isThe main focus of congestion control and quality of service is data trafficdata traffic..
In congestion control we try to avoid traffic congestion.In congestion control we try to avoid traffic congestion.
 In quality of service, we try to create an appropriate environment for theIn quality of service, we try to create an appropriate environment for the
traffic.traffic.
 So, before talking about more details, we discuss the data traffic itself.So, before talking about more details, we discuss the data traffic itself.

27. Ref :Forouzan/ DCN/ Ch-24
Traffic descriptors

28. Ref :Forouzan/ DCN/ Ch-24
Three traffic profiles

29. Admission Control (cont’d)
 The problem of how to describe it in a simple yet meaningful way is
difficult because traffic is typically bursty—the average rate is only
half the story.
 For example: traffic that varies while browsing the Web is more
difficult to handle than a streaming movie with the same long-term
throughput because the bursts of Web traffic are more likely to
congest routers in the network.
 A commonly used descriptor that captures this effect is the leaky
bucket or token bucket.

30. Admission Control (cont’d)
 Armed with traffic descriptions, the network can decide whether to
admit the new virtual circuit.
 One possibility is for the network to reserve enough capacity along
the paths of each of its virtual circuits that congestion will not occur.
 In this case, the traffic description is a service agreement for what
the network will guarantee its users.
 Even without making guarantees, the network can use traffic
descriptions for admission control.
 The task is then to estimate how many circuits will fit within the
carrying capacity of the network without congestion.
(But this task becomes bit tricky , as explained in example next slide)

31. Admission Control (cont’d)
 Suppose that virtual circuits
 that may blast traffic at rates up to 10 Mbps all pass through the
same 100-Mbps physical link. How many circuits should be
admitted?
 Clearly, 10 circuits can be admitted without risking congestion,
but this is wasteful in the normal case since it may rarely happen that all
10 are transmitting full blast at the same time.
 In real networks, measurements of past behaviour that capture the
statistics of transmissions can be used to estimate the number of
circuits to admit, to trade better performance for acceptable risk.

32. Admission Control (cont’d)
 Admission control can also be combined with traffic-aware routing
by considering routes around traffic hotspots as part of the setup
procedure. For Example :

33. Leaky Bucket
 Used in conjunction with resource reservation to police the host’s
reservation
 At the host-network interface, allow packets into the network at a
constant rate
 Packets may be generated in a bursty manner, but after they pass
through the leaky bucket, they enter the network evenly spaced

34. Leaky Bucket: Analogy
Leaky
Bucket
Network
Packets from host

35. Leaky Bucket (cont’d)
 The leaky bucket is a “traffic shaper”: It changes the characteristics
of packet stream
 Traffic shaping makes more manageable and more predictable
 Usually the network tells the leaky bucket the rate at which it may
send packets when the connection begins

36. Leaky Bucket:
Doesn’t allow bursty transmissions
• In some cases, we may want to allow short bursts of packets to enter
the network without smoothing them out
• For this purpose we use a token bucket, which is a modified leaky
bucket

37. Token Bucket
 The bucket holds tokens instead of packets
 Tokens are generated and placed into the token bucket at a constant
rate
 When a packet arrives at the token bucket, it is transmitted if there
is a token available. Otherwise it is buffered until a token becomes
available.
 The token bucket has a fixed size, so when it becomes full,
subsequently generated tokens are discarded

38. Token Bucket
Network
Packets from host
Token Generator
(Generates a token
once every T seconds)

39. Token Bucket vs. Leaky Bucket
Case 1: Short burst arrivals
6543210
Arrival time at bucket
Departure time from a leaky bucket
Leaky bucket rate = 1 packet / 2 time units
Leaky bucket size = 4 packets
6543210
6543210
Departure time from a token bucket
Token bucket rate = 1 tokens / 2 time units
Token bucket size = 2 tokens

40. Token Bucket vs. Leaky Bucket
Case 2: Large burst arrivals
6543210
Arrival time at bucket
Departure time from a leaky bucket
Leaky bucket rate = 1 packet / 2 time units
Leaky bucket size = 2 packets
6543210
6543210
Departure time from a token bucket
Token bucket rate = 1 token / 2 time units
Token bucket size = 2 tokens

41. Contents
 Traffic Throttling
 Choke Packets
Explicit Congestion Notification (ECN)
Hope-by-Hope Backpressure

42. Traffic Throttling
 In the Internet and many other computer networks,
senders adjust their transmissions to send as much
traffic as the network can readily deliver.
 In this setting, the network aims to operate just before
the onset of congestion.
 When congestion is imminent, it must tell the senders to
throttle back their transmissions and slow down.
 There are some approaches to throttling traffic that can
be used in both datagram networks and virtual-circuit
networks.

43. Traffic Throttling (cont’d)
 Each approach must solve two problems.
 First, routers must determine when congestion is
approaching, ideally before it has arrived.
 To do so, each router can continuously monitor
the resources it is using.
 Three possibilities are: the utilization of the output
links, the buffering of queued packets inside the
router, and the number of packets that are lost due to
insufficient buffering.

44. Traffic Throttling (cont’d)
 The second one is the most useful.
 Averages of utilization do not directly account for the
burstiness of most traffic—a utilization of 50% may be
low for smooth traffic and too high for highly variable
traffic.
 The queueing delay inside routers directly captures any
congestion experienced by packets.

45. Traffic Throttling (cont’d)
To maintain a good estimate of the queueing delay d, a
sample of the instantaneous queue length s, can be made
periodically and d updated according to
dnew = αdold + (1 − α)s
where the constant α determines how fast the router
forgets recent history. This is called an EWMA
(Exponentially Weighted Moving Average).
It smoothes out fluctuations and is equivalent to a low-
pass filter. Whenever d moves above the threshold, the
router notes the onset of congestion.

46. Traffic Throttling (cont’d)
 The second problem is that routers must deliver timely
feedback to the senders that are causing the congestion.
 To deliver feedback, the router must identify the
appropriate senders. It must then warn them carefully,
without sending many more packets into the already
congested network.
 Different schemes use different feedback mechanisms ,
Like
• Choke Packets
• Explicit Congestion Notification
• Hope-by-Hope Backpressure

47. Choke Packets Approach
 The most direct way to notify a sender of congestion is to
tell it directly.
 In this approach, the router selects a congested packet
and sends a choke packet back to the source host, giving
it the destination found in the packet.
 The original packet may be tagged (a header bit is
turned on) so that it will not generate any more choke
packets further along the path and then forwarded in the
usual way.
 To avoid increasing load on the network during a time of
congestion, the router may only send choke packets at a
low rate.

48. Forouzan/DCN/ CH.24
Choke packet

49. Choke Packets Approach
 When the source host gets the choke packet, it is
required to reduce the traffic sent to the specified
destination, for example, by 50%.
 For the same reason, it is likely that multiple choke
packets will be sent to a given host and destination.
 The host should ignore these additional chokes for the
fixed time interval until its reduction in traffic takes
effect. After that period, further choke packets indicate
that the network is still congested.
 The modern Internet uses an alternative notification
design (Explicit congestion notification).

50. Explicit Congestion Notification
 Instead of generating additional packets to warn of
congestion, a router can tag any packet it forwards (by
setting a bit in the packet’s header) to signal that it is
experiencing congestion.
 When the network delivers the packet, the destination
can note that there is congestion and inform the sender
when it sends a reply packet.
 The sender can then throttle its transmissions as before.
 This design is called ECN (Explicit Congestion
Notification) and is used in the Internet.

51. Explicit Congestion Notification
(cont’d)
• It is a refinement of early congestion signaling protocols.
• Two bits in the IP packet header are used to record
whether the packet has experienced congestion. Packets
are unmarked when they are sent, as illustrated in
Figure:

52. Explicit Congestion Notification
(cont’d)
 If any of the routers they pass through is congested, that
router will then mark the packet as having experienced
congestion as it is forwarded.
 The destination will then echo any marks back to the
sender as an explicit congestion signal in its next reply
packet.
 This is shown with a dashed line in the figure to indicate
that it happens above the IP level (e.g., in TCP). The
sender must then throttle its transmissions, as in the case
of choke packets.

53. Forouzan/DCN/ Ch.24
Backpressure method for alleviating congestion

54. Hop-by-Hop Backpressure
 At high speeds or over long distances, many new
packets may be transmitted after congestion has been
signalled because of the delay before the signal takes
effect.
 Consider, for example, a host in San Francisco (router A
in Figure) that is sending traffic to a host in New York
(router D in Figure) at the OC-3 speed of 155 Mbps.
 If the New York host begins to run out of buffers, it will
take about 40 msec for a choke packet to get back to San
Francisco to tell it to slow down.
 An ECN indication will take even longer because it is
delivered via the destination.

55. • Choke packets:
– Example showing
slow reaction
– Solution: Hop-by-
Hop choke packets

56. • Hop-by-Hop choke
packets
– Have choke packet
take effect at every
hop
– Problem: more buffers
needed in routers

57. Hop-by-Hop Backpressure
 Choke packet propagation is illustrated as the second, third, and
fourth steps in figure.
 In those 40 msec, another 6.2 megabits will have been sent. Even if
the host in San Francisco completely shuts down immediately, the
6.2 megabits in the pipe will continue to pour in and have to be dealt
with.
 Only in the seventh diagram in Fig. (a) will the New York router
notice a slower flow.
 An alternative approach is to have the choke packet take effect at
every hop it passes through, as shown in the sequence of Fig. (b).
 Here, as soon as the choke packet reaches F, F is required to reduce
the flow to D. Doing so will require F to devote more buffers to the
connection, since the source is still sending away at full blast, but it
gives D immediate relief, like a headache remedy in a television
commercial.

58. Hop-by-Hop Backpressure
 In the next step, the choke packet reaches E, which tells E
to reduce the flow to F. This action puts a greater
demand on E’s buffers but
gives F immediate relief. Finally, the choke packet
reaches A and the flow genuinely slows down.
 The net effect of this hop-by-hop scheme is to provide
quick relief at the point of congestion, at the price of
using up more buffers upstream.
 In this way, congestion can be nipped in the bud
without losing any packets. The idea is discussed in
detail by Mishra et al. (1996).

59. Load Shedding
When a router becomes inundated with packets, it simply drops some
Load Shedding

60. Load Shedding (Cont’d)
 When none of the above methods make the congestion disappear,
routers can bring out the heavy artillery: load shedding. Load
shedding is a fancy way of saying that when routers are being
inundated by packets that they cannot handle , they just throw them
away.
 The term comes from the world of electrical power generation,
where it refers to the practice of utilities intentionally blacking out
certain areas to save the entire grid from collapsing on hot summer
days when the demand for electricity greatly exceeds the supply.
 The key question for a router drowning in packets is which packets
to drop ?.

61. Load Shedding (Cont’d)
• The preferred choice may depend on the type of applications that
use the network. For a file transfer, an old packet is worth more
than a new one.
• In contrast, for real-time media, a new packet is worth more than an
old one. This is because packets become useless if they are delayed
and miss the time at which they must be played out to the user.
• The former policy (old is better than new) is often called wine and
the latter (new is better than old) is often called milk because most
people would rather drink new milk and old wine than the
alternative.

62. Load Shedding (Cont’d)
 An examples: packets that carry routing information. These packets
are more important than regular data packets because they establish
routes; if they are lost, the network may lose connectivity.
 Another example is that algorithms for compressing video, like
MPEG, periodically transmit an entire frame and then send
subsequent frames as differences from the last full frame.
 In this case, dropping a packet that is part of a difference is
preferable to dropping one that is part of a full frame because future
packets depend on the full frame

63. Load Shedding (Cont’d)
• More intelligent load shedding requires cooperation from the
senders.
• To implement an intelligent discard policy, when packets have to
be discarded, routers can first drop packets from the least
important class, then the next important class, and so on.
• Unless there is some significant instruction to avoid marking every
packet as VERY IMPORTANT—NEVER, EVER DISCARD, nobody
will do it.
• For example,
• the network might let senders send faster than the service they
purchased allows if they mark excess packets as low priority. Such a
strategy is actually not a bad idea because it makes more efficient
use of idle resources, allowing hosts to use them as long as nobody
else is interested, but without establishing a right to them when
times get tough.

64. Intelligent Load Shedding
 Discarding packets does not need to be done randomly
 Router should take other information into account
 Possibilities:
 Total packet dropping
 Priority discarding
 Age biased discarding

65.  Total Packet Dropping
 When the buffer fills and a packet segment is dropped, drop all the rest of the
segments from that packet, since they will be useless anyway
 Only works with routers that segment and reassemble packets
 Priority Discarding
 Sources specify the priority of their packets
 When a packet is discarded, the router chooses a low priority packet
 Requires hosts to participate by labeling their packets with priority levels.
 Age Biased Discarding
 When the router has to discard a packet, it chooses the oldest one in its buffer
 This works well for multimedia traffic which requires short delays
 This may not work so well for data traffic, since more packets will need to be
retransmitted
Load Shedding (Cont’d)

66. Random Early Detection
 Dealing with congestion when it first starts is more effective than
letting it gum up the works and then trying to deal with it.
 This observation leads to an interesting twist on load shedding,
which is to discard packets before all the buffer space is really
exhausted.
 The motivation for this idea is that most Internet hosts do not yet
get congestion signals from routers in the form of ECN.
 Instead, the only reliable indication of congestion that hosts get
from the network is packet loss.
 After all, it is difficult to build a router that does not drop packets
when it is overloaded.

67. Random Early Detection
 Transport protocols such as TCP are thus hardwired to react to loss
as congestion, slowing down the source in response.
 The reasoning behind this logic is that TCP was designed for wired
networks and wired networks are very reliable, so lost packets are
mostly due to buffer overruns rather than transmission errors
 .
 Wireless links must recover transmission errors at the link layer (so
they are not seen at the network layer) to work well with TCP.
 This situation can be exploited to help reduce congestion. By having
routers drop packets early, before the situation has become
hopeless, there is time for the source to take action before it is too
late

68. Random Early Detection
 A popular algorithm for doing this is called RED (Random Early
Detection) (Floyd and Jacobson, 1993).
 To determine when to start discarding, routers maintain a running average of
their queue lengths.
 When the average queue length on some link exceeds a threshold, the link is
said to be congested and a small fraction of the packets are dropped at random.
 Picking packets at random makes it more likely that the fastest senders will see
a packet drop; this is the best option since the router cannot tell which source is
causing the most trouble in a datagram network.

69. Random Early Detection
• The affected sender will notice the loss when there is no
acknowledgement, and then the transport protocol will slow down.
• The lost packet is thus delivering the same message as a choke
packet, but implicitly, without the router sending any explicit
signal.
• RED routers improve performance compared to routers that drop
packets only when their buffers are full, though they may require
tuning to work well.
• For example, the number of packets to drop depends on how many
senders need to be notified of congestion.
• However, ECN is the preferred option if it is available . It works in
exactly the same manner, but delivers a congestion signal explicitly
rather than as a loss; RED is used when hosts cannot receive explicit
signals.

70. Jitter Control
 In real-time interactive audio/video, people communicate with one
another in real time.
 The Internet phone or voice over IP is an example of this type of
application.
 Video conferencing is another example that allows people to
communicate visually and orally.

71. Jitter Control(Cont’d)
 Time Relationship:
 Real-time data on a packet-switched network require the
preservation of the time relationship between packets of a session.
 For example, let us assume that a real-time video server creates live
video images and sends them online.
 The video is digitized and packetized.
 There are only three packets, and each packet holds 10s of video
information.

72. Jitter Control(Cont’d)
 The first packet starts at 00:00:00, the second packet starts at 00:00: 10,
and the
 third packet starts at 00:00:20.
 Also imagine that it takes 1 s for each packet to reach the destination
(equal delay).
 The receiver can play back the first packet at 00:00:01, the second
packet at 00:00:11, and the third packet at 00:00:21.

73. Jitter Control(Cont’d)
Figure: Time relationship

74. Jitter Control(Cont’d)
 But what happens if the packets arrive with different delays?
 For example, say the first packet arrives at 00:00:01 (1-s delay), the
second arrives at 00:00: 15 (5-s delay), and the third arrives at 00:00:27 (7-s
delay).
 If the receiver starts playing the first packet at 00:00:01, it will finish at
00:00: 11.
 However, the next packet has not yet arrived; it
arrives 4 s later.

75. Jitter Control(Cont’d)
•There is a gap between the first and second packets and between the
second and the third as the video is viewed at the remote site.
• This phenomenon is called jitter.
• Jitter is introduced in real-time data by the delay between packets.

76. Jitter Control(Cont’d)
Timestamp:
• One solution to jitter is the use of a timestamp.
• If each packet has a timestamp that shows the time it was produced
relative to the first (or previous) packet, then the receiver can add this
time to the time at which it starts the playback.
• In other words, the receiver knows when each packet is to be played.

77. Jitter Control(Cont’d)
•Imagine the first packet in the previous example
has a timestamp of 0, the second has a timestamp of 10, and the third has
a timestamp of
20.
• If the receiver starts playing back the first packet at 00:00:08, the second
will be played at 00:00: 18 and the third at 00:00:28.
• There are no gaps between the packets.
• Next Figure shows the situation.

78. Jitter Control(Cont’d)
Figure: Timestamp
• To prevent jitter, we can time-stamp the packets and separate the
arrival time from the playback time.

79. Playback Buffer:
 To be able to separate the arrival time from the playback time, we need
a buffer to store
the data until they are played back.
 The buffer is referred to as a playback buffer.
 When a session begins (the first bit of the first packet arrives), the
receiver delays playing the data until a threshold is reached.
 In the previous example, the first bit of the first packet arrives at
00:00:01; the threshold is 7 s, and the playback time is 00:00:08.

80. Jitter Control(Cont’d)
• The threshold is measured in time units of data. The replay does not
start until the time units
of data are equal to the threshold value.
• Data are stored in the buffer at a possibly variable rate, but they are
extracted and played back at a fixed rate.
• Next Figure shows the buffer at different times for our example.

81. Figure: Playback buffer
• A playback buffer is required for real-time
traffic.

82. Other Characteristics
• Ordering
• Multicasting
• Translation
• Mixing

83. EXAMPLES
To better understand the concept of congestion control,To better understand the concept of congestion control,
let us give an example:let us give an example:
Congestion Control in TCP

84. Slow start, exponential increase

85. In the slow-start algorithm, the size of the congestion window increases
exponentially until it reaches a threshold.
Note

86. Figure 24.9 Congestion avoidance, additive increase

87. In the congestion avoidance algorithm, the size of the congestion window
increases additively until
congestion is detected.
Note

88. An implementation reacts to congestion detection in one of the following ways:
❏ If detection is by time-out, a new slow
start phase starts.
❏ If detection is by three ACKs, a new
congestion avoidance phase starts.

89. Congestion example

90. References
1. Andrew S. Tanenbaum, Devid J. Wetherall, “ Computer
Networks” , Pearson , 5th
Edition
2. Andrew S. Tanenbaum, Devid J. Wetherall, “ Computer
Networks” , Pearson , 3rd
Edition
3. Behrouz A Forouzan”Data Communications and
Networking” TMH,4th
Edition.
4. http://tools.ietf.org/html/rfc2581 [RFC 2581]
5. http://www.rfc-base.org/rfc-5681.html [RFC 5681]
6. www.net-seal.net