CNv6 Chapter6

Chapter 6: Quality of Service
CCNA Routing and Switching

Connecting Networks v6.0
Chapter 1 - Sections & Objectives
 6.1 QoS Overview
• Explain the purpose and characteristics of QoS.
• Explain how network transmission characteristics impact quality.
• Describe minimum network requirements for voice, video, and data traffic.
• Describe the queuing algorithms used by networking devices.
 6.2 QoS Mechanisms
• Explain how networking devices implement QoS.
• Describe the different QoS models.
• Explain how QoS uses mechanisms to ensure transmission quality.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 2
6.1 QoS Overview
Network Transmission Quality
 QoS or Quality of Service, allows the
Video Tutorial – The Purpose of QoS network administrator to prioritize
certain types of traffic over others.
 Video traffic and voice traffic require
greater resources, such as bandwidth,
from the network than other types of
traffic.
 Financial transactions are time
sensitive and have greater needs than
an FTP transfer or web traffic (HTTP).
 Packets are buffered at the router and
three priority queues have been
established:
• High Priority Queue
• Medium Priority Queue
• Low Priority Queue
Prioritizing Traffic  QoS is an ever increasing
requirement of networks today thanks
to new applications available to users
such as voice and live video
transmissions which create higher
expectations for quality delivery.
 Congestion occurs when multiple
communication lines aggregate onto a
single device, such as a router, and
then much of that data is placed on
fewer outbound interfaces or onto a
slower interface.
 When the volume of traffic is greater
than what can be transported across
the network, devices queue, or hold,
the packets in memory until resources
become available to transmit them.
Prioritizing Traffic (Cont.)
 Queuing packets causes delay
because new packets cannot be
transmitted until previous packets
have been processed.
 Packets will be dropped when
memory fills up.
 One QoS technique that can help with
this problem is to classify data into
multiple queues as shown in the figure
to the left.
 It is important to note that a device
should implement QoS only when it is
experiencing congestion.
Bandwidth, Congestion, Delay, and Jitter
 Network bandwidth is measured in
the number of bits that can be
transmitted in one second (bps).
 Network congestion causes delay.
An interface experiences congestion
when it is presented with more traffic
than it can handle.
 Delay or latency refers to the time it
takes for a packet to travel from the
source to the destination.
• Fixed delay
• Variable delay
 Jitter is the variation in delay of
received packets.
 Without any QoS mechanisms in place,
Packet Loss packets are processed in the order in
which they are received.
• When congestion occurs, network
devices will drop packets.
• This includes time-sensitive video and
audio packets.
 For example, when a router receives a
digital audio stream for VoIP, it must
compensate for the jitter that is
encountered.
• The mechanism that handles this
function is the playout delay buffer.
• The playout delay buffer must buffer
these packets and then play them out in
a steady stream.
• The digital packets are later converted
back to an analog audio stream.
 If the jitter is so large that it causes
Packet Loss (Cont.) packets to be received out of the range of
this buffer, the out-of-range packets are
discarded and dropouts are heard in the
audio.
 For losses as small as one packet, the
digital signal processor (DSP) interpolates
what it thinks the audio should be and no
problem is audible to the user.
 However, when jitter exceeds what the
DSP can handle, audio problems are
heard.
 In a properly designed network, voice
packet loss should be zero
 Network engineers use QoS mechanisms
to classify voice packets for zero packet
loss.
Traffic Characteristics
Video Tutorial – Traffic Characteristics
 Voice and video traffic place a greater
demand on the network and are two of the
main reasons for QoS.
 There are some differences between voice
and video:
• Voice packets do not consume a lot of
resources because they are not very large
and they are fairly steady. Voice traffic
requires at least 30 kilobits per second of
bandwidth with no more than 1% packet
loss.
• Video traffic is more demanding. The
packets are more bursty and greedy. It
consumes a lot more resources. Video
traffic requires at least 384 kilobits per
second in ©bandwidth with no more than .1 to
2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 10
1% packet loss.
 In the early 2000s, the predominant types
Network Traffic Trends of IP traffic were voice and data.
 Voice traffic has a predictable bandwidth
need and known packet arrival times.
 Data traffic is not real-time and has an
unpredictable bandwidth need.
 More recently, video traffic has become
increasingly important to business
communications and operations.
 According to the Cisco Visual Networking
Index (VNI), video traffic represented 67%
of all traffic in 2014. By 2019, video will
represent 80% of all traffic.
 The type of demands that voice, video,
and data traffic place on the network are
very different.
 Voice traffic is predictable and smooth.
Voice  However, voice traffic is very sensitive to
delay and dropped packets; there is no
reason to retransmit voice if packets are lost.
 Voice packets must receive a higher priority
than other types of traffic.
 Cisco products use the RTP port range
16384 to 32767 to prioritize voice traffic.
 Voice can tolerate a certain amount of
latency, jitter, and loss without any noticeable
effects.
 Latency should be no more than 150 ms.
 Jitter should be no more than 30 ms.
 Voice packet loss should not exceed 1%.
 Without QoS and a significant amount
Video of extra bandwidth capacity, video
quality typically degrades.
 The picture appears blurry, jagged, or
in slow motion. The audio portion
may become unsynchronized with the
video.
 Video Traffic Characteristics:
• Video – Bursty, greedy, drop sensitive,
delay sensitive, UDP priority
• One-Way Requirements:
• Latency <= 200 - 400 ms
 Compared to voice, video is less resilient to loss and has a
higher volume of data per packet as shown above. • Jitter <= 30 – 50ms
• Loss <= 0.1 – 1%
• Notice how voice packets arrive every 20 ms and are 200 bytes.
• Bandwidth (384 Kb/s – 20+ Mb/s)
• In contrast, the number and size of video packets varies every
33 ms based on the content of the video.
Traffic Characteristics  Most applications use either TCP or UDP. Unlike UDP, TCP
Data performs error recovery.
 Data applications that have no tolerance for data loss, such as
email and web pages, use TCP to ensure packets will be
resent in the event they are lost.
 Some TCP applications, such as FTP, can be very greedy,
consuming a large portion of network capacity.
 Although data traffic is relatively insensitive to drops and
delays compared to voice and video, a network administrator
still needs to consider the quality of the user experience.
 Two factors that need to be determined:

• Does the data come from an interactive
application?
• Is the data mission critical?
queuing Algorithms
Video Tutorial – QoS Algorithms  If we look at the queuing strategies for
QoS, FIFO Queuing or First in First Out
Queuing, is basically the absence of QoS.
 Packets that enter the router will leave the
router in the same order.
 Compare this with Weighted Fair Queuing
or WFQ and packets that come into a
router are then classified and prioritized
based on the classification.
 A newer form of Weighted Fair Queuing is
Class Based Weighted Fair Queuing.
 In order to guarantee that voice traffic is
prioritized to the point there are no drops,
Low-Latency Queuing can be used with
CBWFQ to prioritize voice packets above
all else.
Queuing Algorithms
Queuing Overview  The QoS policy implemented by the network
administrator becomes active when
congestion occurs on the link.
 Queuing is a congestion management tool
that can buffer, prioritize, and if required,
reorder packets before being transmitted to
the destination.
 This course will focus on the following
queuing algorithms:
• First-In, First-Out (FIFO)
• Weighted Fair Queuing (WFQ)
• Class-Based Weighted Fair Queuing (CBWFQ)
• Low Latency Queuing (LLQ)
Queuing Algorithms  FIFO queuing, also known as first-come,
First In First Out (FIFO) first-served queuing, involves buffering
and forwarding of packets in the order of
arrival.
 FIFO has no concept of priority or classes
of traffic and consequently, makes no
decision about packet priority.
 There is one queue and all packets are
treated equally.
 When FIFO is used, important or time-
sensitive traffic can be dropped when
congestion occurs on the router or switch
 FIFO is effective for large links that have little delay interface.
and minimal congestion  When no other queuing strategies are
 If your link has very little congestion, FIFO queuing configured, FIFO is used on serial
may be the only queuing you need to use. interfaces at E1 (2.048 Mbps) and below.
Queuing Algorithms  WFQ is an automated scheduling method
that provides fair bandwidth allocation to all
Weighted Fair Queuing (WFQ) network traffic.
 WFQ applies priority, or weights, to identified
traffic and classifies it into conversations or
flows.
 WFQ then determines how much bandwidth
each flow is allowed relative to other flows.
 WFQ schedules interactive traffic to the front
of a queue to reduce response time. It then
shares the remaining bandwidth among
high-bandwidth flows.
 WFQ classifies traffic into different flows
based on packet header addressing,
including source/destination IP addresses,
MAC addresses, port numbers, protocols,
and type of service (ToS) values.
Queuing Algorithms
Class-Based Weighted Fair Queuing (WFQ)
 CBWFQ extends the standard WFQ
functionality to provide support for user-
defined traffic classes.
 You define traffic classes based on match
criteria including protocols, ACLs, and input
interfaces.
 When a class has been defined according to
its match criteria, you can assign it
characteristics.
• To characterize a class, you assign it
 Packets that match the criteria for a class bandwidth, weight, and maximum packet
constitute the traffic for that class. A FIFO queue limit.
is reserved for each class, and traffic belonging to • The bandwidth assigned to a class is the
a class is directed to the queue. guaranteed bandwidth delivered to the class
during congestion.
Queuing Algorithms
 The LLQ feature brings strict priority queuing
Low Latency Queuing (LLQ) (PQ) to CBWFQ which reduces jitter in voice
conversations. See the figure to the left.
 Strict PQ allows delay-sensitive data such as
voice to be sent before packets in other
queues.
 Without LLQ, CBWFQ provides WFQ based
on defined classes with no strict priority
queue available for real-time traffic.
• All packets are serviced fairly based on
weight.
• This scheme poses problems for voice traffic
that is largely intolerant of delay.
 With LLQ, delay-sensitive data is sent first,
before packets in other queues are treated.
 LLQ allows delay-sensitive data such as
voice to be© 2016
sent first giving it preferential
Cisco and/or its affiliates. All rights reserved. Cisco Confidential 20
treatment.
6.2 QoS Mechanisms
QoS Models
Video Tutorial – QoS Models  Because packets are delivered on a best-
effort basis, the best effort model is not
really an implementation of QoS.
 The integrated services model, or IntServ
model, provides a very high degree of
QoS to IP packets with guaranteed
delivery.
 It uses a signaling process known as
RSVP, or resource reservation protocol.
 The differentiated services model, or
DiffServ model, is a highly scalable and
flexible implementation of QoS. It works
off manually configured traffic classes
which need to be configured on routers
throughout the network.
QoS Models
Selecting an Appropriate QoS Policy Model
 How can QoS be implemented in a
network? The three models for
implementing QoS are these:
• Best-effort model
• Integrated services (IntServ)
• Differentiated Services (DiffServ)
 The table in the figure to the left
summarizes these three models.
 QoS is implemented in a network using
either or both of these:
• IntServ – provides the highest guarantee
of QoS, but is resource-intensive
• DiffServ – less resource intensive and
more scalable
QoS Models
Best-Effort  The basic design of the Internet, which is
still applicable today, provides for best-
effort packet delivery and provides no
guarantees.
 The best-effort model treats all network
packets the same way.
 Without QoS, the network cannot tell the
difference between packets. A voice call
will be treated the same as an email with
a digital photograph attached.
 The best effort-model is similar in concept
to sending a letter using standard postal
mail. All letters are treated the same and
in some cases will never arrive.
QoS Models
Integrated Services  The needs of real-time applications, such as
remote video, multimedia conferencing,
visualization, and virtual reality, motivated
the development of the IntServ architecture
model in 1994.
 IntServ provides a way to deliver end-to-end
Qos that real-time applications require by
explicitly managing network resources to
provide QoS to specific user packet
streams.
 It uses resource reservation and an
admission-control mechanism as building
blocks to establish and maintain QoS.
 IntServ uses a connection-oriented
approach inherited from telephony network
design.
QoS Models
Integrated Services (Cont.)  In the IntServ model, the application
requests a specific kind of service from the
network before sending the data.
 The application informs the network of its
traffic profile and requests a particular kind
of service that can encompass its
bandwidth and delay requirements.
 IntServ uses the Resource Reservation
Protocol (RSVP) to signal the QoS needs of
an application’s traffic along devices in the
end-to-end path through the network.
 If the network devices along the path can
reserve the necessary bandwidth, the
originating application can begin
transmitting – otherwise, no data is sent.
QoS Models
 The differentiated services (DiffServ) QoS
Differentiated Services model:
• Specifies a simple and scalable
mechanism for classifying and managing
network traffic.
• Provides QoS guarantees on modern IP
networks.
• DiffServ can provide low-latency
guaranteed service to critical network
traffic such as voice or video.
 The DiffServ design overcomes the
limitations of both the best-effort and
IntServ models.
 DiffServ can provide an “almost
guaranteed” QoS while still being cost-
effective and scalable.
QoS Models
 DiffServ is not an end-to-end QoS strategy
Differentiated Services (Cont.) because it cannot enforce end-to-end
guarantees. However, it is a more scalable
approach to implementing QoS.
 In the figure to the left, a host forwards
traffic to a router, the router classifies the
flows into aggregates (classes) and
provides the appropriate QoS policy for the
classes.
 DiffServ enforces and applies QoS
mechanisms on a hop-by-hop basis
uniformly applying global meaning to each
traffic class to provide both flexibility and
scalability.
 DiffServ divides network traffic into classes
based on business requirements. Each
class can then be assigned a different level
of service.© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 28
QoS Implementation Techniques
Video Tutorial – QoS Implementation Techniques
 QoS implementation tools can be
categorized into three main categories:
• Classification and marking tools – Session
traffic is classified into different priority
groupings and packets are marked.
• Congestion avoidance tools – Traffic classes
are allotted network resources and some
traffic may be selectively dropped, delayed
or remarked to avoid congestion.
• Congestion management tools – During
congestion, traffic is queued to await the
availability of those resources; tools include
class based weighted fair queuing, and low
latency queuing.
QoS Implementation Techniques  Packet loss is usually the result of
Avoiding Packet Loss congestion on an interface.
 Most TCP applications experience
slowdown because TCP automatically
adjusts to network congestion.
• Some applications do not use TCP and
cannot handle drops (fragile flows).
 The following approaches can prevent
drops in sensitive applications:
• Increase link capacity to ease or prevent
congestion.
• Guarantee enough bandwidth and increase
buffer space to accommodate bursts of
traffic from fragile flows – WFQ, CBWFQ
and LLQ.
• Prevent congestion by dropping lower-
priority packets before congestion occurs –
weighted random early detection (WRED).
QoS Implementation Techniques  There are three categories of QoS tools:
QoS Tools • Classification and marking tools
• Congestion avoidance tools
• Congestion management tools
 Ingress packets (gray squares) are
classified and their respective IP header is
marked (colored squares). To avoid
congestion, packets are then allocated
resources based on defined policies.
 Packets are then queued and forwarded
out the egress interface based on their
defined QoS shaping and policing policy.
 Classification and marking can be done on
ingress or egress, whereas other QoS
actions such as queuing and shaping are
usually done on egress.
QoS Implementation Techniques  A packet has to be classified before it can have
a QoS policy applied to it.
Classification and Marking
 Classification and marking allows us to identify,
or “mark” types of packets.
 Classification determines the class of traffic to
which packets or frames belong. Policies can
not be applied unless the traffic is marked.
 Methods of classifying traffic flows at Layer 2
 The table in the figure describes some of the marking and 3 include using interfaces, ACLs, and
fields used in various technologies. Consider the class maps.
following points when deciding to mark traffic at Layers
2 or 3:  Marking requires the addition of a value to the
• Layer 2 marking of frames can be performed for non-IP packet header and devices that receive the
traffic. packet look at this field to see if it matches a
• Layer 2 marking of frames is the only QoS option defined policy.
available for switches that are not “IP aware”.
• Layer 3 marking will carry the QoS information end-to-  Marking should be done as close to the source
end. as possible and this establishes the trust
boundary.
Marking at Layer 2  802.1Q is the IEEE standard that
supports VLAN tagging at Layer 2 on
Ethernet networks.
 When 802.1Q is implemented, two
fields are added to the Ethernet Frame
and are inserted following the source
MAC address field as shown in the
figure to the left.
 The 802.1Q standard includes the QoS
prioritization scheme known as IEEE
802.1p. The standard uses the first
three bits in the Tag Control Information
(TCI) field and identifies the CoS
markings.
 These three bits allow eight levels of
priority (0-7).
 IPv4 and IPv6 specify an 8-bit field in
Marking at Layer 3 their packet headers to mark packets.
• IPv4 – Type of Service (ToS) field
• IPv6 – Traffic Class field
 These fields are used to carry the
packet marking assigned by the QoS
classification tools. Forwarding
devices refer to this field and forward
the packets based on the QoS policy.
 RFC 2474 redefines the ToS field by
renaming and extending the IPP field.
The new filed has 6-bits allocated for
QoS called the differentiated services
code point (DSCP) field.
 These six bits offer a maximum of 64
possible classes of service.
 The 64 DSCP values are organized into three
Marking at Layer 3 (Cont.) categories:
• Best-Effort (BE) – Default for all IP packets.
The DSCP value is 0.
• Expedited Forwarding (EF) – The DSCP value
is 46. At layer 3, Cisco recommends that EF
should only be used to mark voice packets.
• Assured Forwarding (AF) – Uses the 5 most
significant DSCP bits to indicate queues and
drop preference. As shown in the figure, the
first 3 most significant bits are used to
designate the class.
• Class 4 is the best queue and Class 1 is the worst
queue.
• The 4th and 5th most significant bits are used to
designate the drop preference.
• The 6th most significant bit is set to zero.
• The AFxy formula shows how the AF values are
calculated.
QoS Implementation Techniques  Where should markings occur?
Trust Boundaries  Traffic should be classified and marked

as close to its source as possible.
 This defines the trust boundary as
shown in the figure.
• Trusted endpoints have the capabilities
and intelligence to mark application
traffic to the appropriate Layer 2 CoS or
Layer 3 DSCP values. Examples of
trust endpoints include IP phones,
wireless access points, and
videoconferencing systems.
• Secure endpoints can have traffic
marked at the Layer 2 switch.
• Traffic can also be marked at Layer 3
switches and routers.
 Re-marking of traffic is typically
necessary.
Congestion Avoidance  Congestion avoidance tools monitor
network traffic loads in an effort to
anticipate and avoid congestion at
common network bottlenecks before
congestion becomes a problem.
 Congestion avoidance is achieved
through packet dropping.
 These tools monitor the average depth
of the queue.
• For example, when the queue fills up
to the maximum threshold, a small
percentage of packets are dropped.
• When the maximum threshold is
passed, all packets are dropped.
Congestion Avoidance (Cont.)  The Cisco IOS includes weighted random
early detection (WRED) as a possible
congestion avoidance solution.
• WRED is a congestion avoidance technique
that allows for preferential treatment of which
packets will get dropped.
• The WRED algorithm allows for congestion
avoidance on network interfaces by providing
buffer management and allowing TCP traffic to
decrease, or throttle back, before buffers are
exhausted.
• Using WRED helps avoid tail drops and
maximizes network use and TCP-application
performance.
 There is no congestion avoidance for UDP
traffic – such as voice traffic.
QoS Implementation Techniques  Traffic shaping and policing are two
Shaping and Policing mechanisms provided by the Cisco IOS
QoS software to prevent congestion.
 Traffic shaping retains excess packets in a
queue and then schedules the excess for
later transmission over increments of time.
• The result of traffic shaping is a smoothed
packet output rate as shown in the figure.
• Shaping requires sufficient memory.
 Shaping is used on outbound traffic.
 Policing is commonly implemented by

service providers to enforce a contracted
customer information rate (CIR).
 Policing either drops or remarks excess
traffic.
 Policing is often applied to inbound traffic.
6.3 Summary
Summary
Conclusion
• Explain the purpose and characteristics of QoS.
• Explain how networking devices implement QoS.

CNv6 Chapter6

Uploaded by

Copyright:

Available Formats

CNv6 Chapter6

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CNv6 Chapter6

Uploaded by

Copyright:

Available Formats

Chapter 6: Quality of Service

CCNA Routing and Switching

 Jitter should be no more than 30 ms.

 Voice packet loss should not exceed 1%.

 Two factors that need to be determined:

Trust Boundaries  Traffic should be classified and marked

 Policing is commonly implemented by

You might also like