QoS Presentation2

IP Quality of Service
Objectives
 Upon completing this module, you will be able to:

 Describe the need for IP QoS
 Describe the Integrated Services model
 Describe the Differentiated Services model
 Describe the building blocks of IP QoS mechanisms (classification,
marking, metering, policing, shaping, dropping, forwarding, queuing)
Inner Mongolia University

Introduction
Objectives
 Upon completing this lesson, you will be able to:

 Describe different types of applications and services that have
special resource requirements
 List the network components that affect the throughput, delay,
and jitter in IP networks
 List the benefits of deploying QoS mechanisms in IP networks
 Describe typical enterprise and service provider networks and
their QoS-related requirements

Why IP QoS?
 Application X is slow.
 Video broadcast occasionally stalls.
 Phone calls over IP are no better than over satellite.
 Phone calls can have very bad voice quality.
 ATMs (the money-dispensing type) are
nonresponsive.
5 Inner Mongolia University

Because ...
 Application X is slow!
(not enough bandwidth)
 Video broadcast occasionally stalls!
(delay temporarily increases – jitter)
 Phone calls over IP are no better than over satellite! (too
much delay)
 Phone calls can have very bad voice quality!
(too many phone calls – admission control)
 ATMs (the money-dispensing type) are nonresponsive!
(too many drops)

What Causes ...
 Lack of bandwidth?: Multiple flows are contesting for

a limited amount of bandwidth.
 Too much delay?: Packets have to traverse many
network devices and links.
 Variable delay?: Sometimes there is a lot of other
traffic, which results in more delay.
 Drops?: Packets have to be dropped when a link is
congested.

Available Bandwidth
IP IP IP IP
512 kbps
256 kbps
10 Mbps
100 Mbps
BWmax = min(10M, 256k, 512k, 100M)=256 kbps

BWavail = BWmax /Flows
 Maximum available bandwidth equals the bandwidth of the
weakest link.
 Multiple flows are competing for the same bandwidth, resulting in much less
bandwidth being available to one single application.

End-to-End Delay
IP IP IP IP
Propagation Propagation Propagation

Delay (P1) Delay (P2) Delay (P3) Propagation
Delay (P4)
Processing and Processing and Processing and
Queuing Delay (Q1) Queuing Delay (Q2) Queuing Delay (Q3)
Delay = P1 + Q1 + P2 + Q2 + P3 + Q3 + P4 = X ms
 End-to-end delay equals a sum of all propagation, processing, and
queuing delays in the path.
 Propagation delay is fixed; processing and queuing delays are
unpredictable in best-effort networks.

Processing, Queuing, and
Propagation Delay
Forwarding
Bandwidth
IP IP IP IP
Processing Delay Queuing Delay

Propagation Delay
 Processing delay is the time it takes for a router to take the packet from an input
interface and put it into the output queue of the output interface.
 Queuing delay is the time a packet resides in the output queue of a router.
 Propagation or serialization delay is the time it takes to transmit a packet.

Packet Loss
Forwarding
IP IP IP IP IP
Tail-drop
 Tail-drops occur when the output queue is full. These are the most common
drops which happen when a link is congested.
 There are also many other types of drops (input queue drop, ignore, overrun, no
buffer, etc), which are not as common and which may require a hardware
upgrade. These drops are usually a result of router congestion.

How to Increase Available
Bandwidth?
TCP Header Compression
RTP Header Compression
cTCP Data
Compress
the Headers
IP TCP Data FIFO queuing

Fancy Queuing
Compress
the Payload
Priority Queuing (PQ)
Custom Queuing (CQ)
Stacker Modified Deficit Round Robin (MDRR)
Compressed Packet
Predictor Class-Based Weighted Fair Queing (CBWFQ)
 Upgrade the link—the best solution but also the most expensive.
• Take some bandwidth from less important applications.
• Compress the payload of Layer 2 frames.
• Compress the header of IP packets.
How to Reduce Delay?
TCP Header Compression

RTP Header Compression
cRTP Data
Compress
the Headers
IP UDP RTP Data FIFO queuing

Fancy Queuing
Compress
the Payload
Priority Queuing (PQ)
Custom Queuing (CQ)
Stacker Strict Priority MDRR
Compressed Packet IP RTP Prioritization
Predictor
Class-Based Low-Latency Queuing (CBLLQ)
• Forward the important packets first.
• Compress the payload of Layer-2 frames (it takes time).
• Compress the header of IP packets.
How to Prevent Packet Loss?
Weighted Random Early Detection (WRED)
IP Data Dropper Fancy Queuing

FIFO queuing
Custom Queuing (CQ)

Modified Deficit Round Robin (MDRR)
Class-Based Weighted Fair Queuing (CBWFQ)
• Guarantee enough bandwidth to sensitive packets.
• Prevent congestion by randomly dropping less important packets
before congestion occurs.

Which Applications Have Which QoS
Requirements?
Throughput Delay Loss Jitter
Interactive Not
(e.g., Telnet) Low Low Low Important
Batch (e.g., Not Not
FTP) High Important Low Important
Fragile (e.g,. Low Low None Not
SNA) Important
Voice Low Low and Low Low

Predictable
Video High Low and Low Low

Predictable
 Enterprise networks are typically focused on providing QoS to
applications.
Which Services Can Be
Implemented in a Network?
Throughput Delay Loss Jitter
Gold Guaranteed Low Low Low
No No No
Silver Guaranteed Guarantee Guarantee Guarantee
Bronze Guaranteed No No No
Limited Guarantee Guarantee Guarantee
Best Effort No No No No
Guarantee Guarantee Guarantee Guarantee
 Service provider networks typically offer services based on

source and destination addresses.
How Can QoS Be Applied?
 Best effort—no QoS is applied to packets (default

behavior)
 Integrated Services model—applications signal to the
network that they require special QoS
 Differentiated Services model—the network
recognizes classes that require special QoS

Summary
 Upon completing this lesson, you should be able
to:
 Describe different types of applications and services that have
special resource requirements
 List the network components that affect the throughput, delay,
and jitter in IP networks
 List the benefits of deploying QoS mechanisms in IP networks
 Describe typical enterprise and service provider networks and
their QoS-related requirements

Review Questions
1.What are the relevant parameters that define quality

of service?
2.What can be done to give more bandwidth to an
application?
3.What can be done to reduce delay?
4.What can be done to prevent packet loss?
5.Name the two QoS models.

Integrated Services
Model
Objectives
 Upon completing this lesson, you will be
able to:
 Describe the IntServ model
 List the key benefits and drawbacks of the IntServ
model
 List some implementations that are based on the
IntServ model
 Describe the need for Common Open Policy Service
(COPS)

Integrated Services
 The Internet was initially based on a
best-effort packet delivery service.
 Today's Internet carries many more different
applications than 20 years ago.
 Some applications have special bandwidth and delay
requirements.
 The Integrated Services model (RFC1633) was
introduced to guarantee predictable network behavior
for these applications.

IntServ Building Blocks
Local Remote Admission Local

Admission Control Admission
Control Control
Policy Enforcement
Point (PEP)
request request request request
reserve reserve reserve reserve
request
reply
Policy Decision
Point (PDP)
 Resource reservation is used to identify an application (flow) and signal

if there are enough available resources for it.
 Admission control is used to determine if the application (flow) can get
the requested resources.

Reservation and Admission
Protocols
 The Resource Reservation Protocol (RSVP) was

developed to communicate resource needs between
hosts and network devices (RFCs 2205 to 2215).
 Common Open Policy Service (COPS) was
developed to offload admission control to a central
policy server (RFCs 2748 to 2753).

RSVP-Enabled Applications
 RSVP is typically used by applications carrying voice

or video over IP networks (initiated by a host).
 RSVP with extensions is also used by MPLS Traffic
Engineering to establish MPLS/TE tunnels (initiated
by a router).

IntServ Implementation Options
RSVP
1) Explicit RSVP on each network node
Class of Service
or
Best Effort
2) RSVP ‘pass-through’ and CoS transport
- map RSVP to CoS at network edge
- pass-through RSVP request to egress
3) RSVP at network edges and ‘pass-through’ with
- best-effort forwarding in the core (if there is
enough bandwidth in the core)

Explicit RSVP Transport
IntServ End-to-End
RSVP
All Routers
• WFQ applied per flow
based on RSVP requests

RSVP Pass-Through
IntServ - DiffServ Integration
RSVP RSVP
Precedence
Classifier
WRED
Premium Egress Router
Standard
• RSVP protocol
sent on to destination
Ingress Router • WFQ applied to
• RSVP protocol manage egress flow
Mapped to classes
Passed through to egress Backbone
• WRED applied based
on class
Benefits and Drawbacks of the
IntServ Model
+ RSVP benefits:
• Explicit resource admission control (end-to-end)
• Per-request policy admission control (authorization object,
policy object)
• Signaling of dynamic port numbers
(for example, H.323)
– RSVP drawbacks:
• Continuous signaling due to stateless architecture
• Not scalable

Common Open Policy Service
 Common Open Policy Service (COPS) provides the
following benefits when used with RSVP:
• Centralized management of services
• Centralized admission control and authorization of RSVP
flows
 RSVP-based QoS solutions become more scalable

Summary
 Upon completing this lesson, you should
be able to:
 Describe the IntServ model
 List the key benefits and drawbacks of the IntServ
model
 List some implementations that are based on the
IntServ model
 Describe the need for Common Open Policy Service
(COPS)

Review Questions
1.What are the two building blocks of the Integrated
Services model?
2.Which protocol is used to signal QoS requirements to
the network?
3.Which protocol is used to offload admission control to
a central policy server?

Differentiated Services
Model
Objectives

 Describe the DiffServ model
 List the key benefits of the DiffServ model compared to the
IntServ model
 Describe the purpose of the DS field in IP headers
 Describe the interoperability between DSCP-based and IP-
Precedence-based devices in a network
 Describe the expedited forwarding service
 Describe the assured forwarding service

Differentiated Services Model
 TheDifferentiated Services model describes services associated
with traffic classes.
 Complex traffic classification and conditioning are performed at
network edge, resulting in a per-packet Differentiated Services
Code Point (DSCP).
 No per-flow/per-application state exists in the core.
 The core performs only simple “per-hop behaviors” on traffic
aggregates.
 The goal is scalability.

Additional DiffServ Requirements
 Wide variety of services and provisioning policies

 Decouple service and application in use
 No application modification
 No hop-by-hop signaling
 Interoperability with non-DS-compliant nodes
 Incremental deployment

DiffServ Elements
 The service defines QoS requirements and

guarantees provided to a traffic aggregate.
 The conditioning functions and per-hop behaviors are
used to realize services.
 The DS field value (DSCP) is used to mark packets
to select a per-hop behavior.
 Per-hop Behavior (PHB) is implemented using a
particular QoS mechanism.
 Provisioning is used to allocate resources to traffic
classes.

Why Is Provisioning Important?
 QoS does not create bandwidth!

 QoS manages bandwidth usage among
multiple classes.
 QoS gives better service to a well-provisioned class
with respect to another class.

Topological Terminology
DS Interior Node
DS Egress
DS Ingress Boundary Node Boundary Node
Boundary Link
Upstream
DS Domain Downstream
DS Domain
DS Region
Traffic Stream = set of flows
Behavior Aggregate (flows with the same DSCP)

Traffic Terminology
 Flow: a single instance of an application-to-application flow

of packets. A flow is identified by source address, source
port, destination address, destination port, and protocol ID.
 Traffic stream: an administratively significant set of one or
more flows that traverse a path segment. A traffic stream
may consist of a set of active flows that are selected by a
particular classifier.
 Traffic profile: a description of the temporal properties of a
traffic stream, such as average and peak rate and burst
size.

Traffic Terminology (cont.)
 A behavior aggregate (BA) is a collection of packets

with the same DSCP crossing a link in a particular
direction.
 Per-hop behavior (queuing in a node) is externally
observable forwarding behavior applied at a DiffServ-
compliant node to a DiffServ behavior aggregate.
 A PHB Mechanism is a specific algorithm or operation
(e.g., queuing discipline) that is implemented in a
node to realize a set of one or more per-hop
behaviors.

Packet Header Terminology
DSCP Field: 6 bits Unused: 2 bits
Former ToS Byte = New DS Field
 DSCP: a specific value of the DSCP portion of the DS field. The

DSCP is used to select a PHB (Per-Hop Behavior; forwarding
and queuing method)
 DS field: the IPv4 header ToS octet or the IPv6 traffic class octet
when interpreted in conformance with the definition given in RFC
2474. The bits of the DSCP field encode the DSCP, while the
remaining bits are currently unused.

DSCP Encoding
 Three pools:
• “xxxxx0” Standard Action
• “xxxx11” Experimental/Local Use
• “xxxx01” EXP/LU (possible std action)
 Default DSCP: “000000”
 Default PHB: FIFO, tail-drop

DSCP Usage
 DSCP selects per-hop behavior (PHB)

throughout the network:
 Default PHB
 Class selector (IP Precedence) PHB
 Expedited forwarding PHB
 Assured forwarding PHB

Backward Compatibility Using the
Class Selector
 Non-DS-compliant node: node that does not interpret

the DSCP correctly or that does not support all the
standardized PHBs
 Legacy node: a non-DS-compliant node that
interprets IPv4 ToS as defined by RFC 791 and RFC
1812
 DSCP: backward compatible with IP Precedence
(class selector code point, RFC 1812) but not with the
ToS byte definition from RFC 791 (“DTR” bits)

Class Selector Code Point
 Compatibility with current IP Precedence usage (RFC

1812)
 “xxx000” DSCPs
 Differentiates Probability of Timely Forwarding (PTF)
• PTF(xyz000) >= PTF(abc000) ifxyz > abc

Expedited Forwarding
 Expedited forwarding PHB:

• Ensures a minimum departure rate
• Guarantees bandwidth—the class is guaranteed an
amount of bandwidth with prioritized forwarding
• Polices bandwidth—the class is not allowed to exceed the
guaranteed amount (excess traffic is dropped)
 DSCP value: “101110”; looks like IP Precedence 5 to non-
DS-compliant devices

IOS Expedited Forwarding PHB
Implementations
 Priority queuing
 IP RTP Prioritization
 Class-based low-latency queuing (CBLLQ)
 Strict priority queuing within modified deficit round
robin (MDRR) on GSRs

Assured Forwarding
 Assured forwarding PHB:

• Guarantees bandwidth
• Allows access to extra bandwidth if available
 Four standard classes (af1, af2, af3, and af4)
 DSCP value range: “aaadd0” where “aaa” is a binary
value of the class and “dd” is the drop probability

Assured Forwarding Encoding
Class Value Drop Value

Probability
AF1 001dd0 (dd)
Low 01
AF2 010dd0 Medium 10
AF3 011dd0 High 11
AF4 100dd0
 Each Assured Forwarding class uses three DSCP values
 Each Assured Forwarding class is independently forwarded with
its guaranteed bandwidth
 Differentiated RED is used within each class to prevent
congestion within the class

Assured Forwarding PHB Definition
 A DS node must allocate a configurable, minimum

amount of forwarding resources (buffer space and
bandwidth) per assured forwarding class.
 Excess resources may be allocated between non-idle
classes. The manner must be specified.
 Reordering of IP packets of the same flow is not
allowed if they belong to the same assured forwarding
class.

Assured Forwarding PHB
Implementation
 CBWFQ (four classes) with WRED within each class

 (M)DRR with WRED within each class
 Optional custom queuing
(does not support differentiated dropping)

Summary

to:
 Describe the DiffServ model
 List the key benefits of the DiffServ model compared to the
IntServ model
 Describe the purpose of the DS field in IP headers
 Describe the interoperability between DSCP-based and IP-
Precedence-based devices in a network
 Describe the expedited forwarding service
 Describe the assured forwarding service

Review Questions
1.What are the benefits of the DiffServ model compared

to the IntServ model?
2.What is a DiffServ code point?
3.Name the standard PHBs.
4.How was backward compatibility with IP Precedence
achieved?
5.Describe the PHB of assured forwarding.
6.Describe the PHB of expedited forwarding.

Building Blocks of IP
QoS Mechanisms
Objectives

 Describe different classification options in IP networks
 Describe different marking options in IP networks
 List the mechanisms that are capable of measuring the rate of traffic
 List the mechanisms that are used for traffic conditioning, shaping,
and avoiding congestion

Router Functions
Defragmentation
Decompression (payload, header) Rate limiting
Source-based QoS-label/precedence setting Random dropping
Destination-based QoS-label/precedence Shaping
setting Compression (payload, header)
Rate limiting Fragmentation
Class-based marking Queuing and scheduling
Policy-based routing ...
...
Input Output
Input I/O Forwarding Output I/O
Processing Processing
Process switching
Fast/optimum switching
Netflow switching
CEF switching
 Depending on the configuration, a router may perform a number of actions prior to

forwarding a packet (input processing)
 Depending on the configuration, a router may perform a number of actions prior to
enqueuing a packet in the hardware queue
(output processing)
IP QoS Actions
 Classification—Each class-oriented QoS mechanism has to

support some type of classification
(access lists, route maps, class maps, etc.).
 Metering—Some mechanisms measure the rate of traffic to
enforce a certain policy
(e.g., rate limiting, shaping, scheduling, etc.).
 Dropping—Some mechanisms are used to drop packets (e.g.,
random early detection).
 Policing—Some mechanisms are used to enforce a rate limit
based on the metering
(excess traffic is dropped).
 Shaping—Some mechanisms are used to enforce a rate limit
based on the metering
(excess traffic is delayed).

IP QoS Actions (cont.)
 Marking—Some mechanisms have the capability to

mark packets based on classification or metering
(e.g., CAR, class-based marking, etc.).
 Queuing—Each interface has to have a queuing
mechanism.
 Forwarding—There are several supported
forwarding mechanisms (process switching, fast
switching, CEF switching, etc.).

DiffServ Mechanisms in IOS
Meter
Classifier Marker Conditioner Queuing

Inbound
traffic Shaping Scheduling
stream Dropping Dropping
 Most traditional QoS mechanisms include extensive built-in classifiers
• Committed access rate (CAR)
• QoS Policy Propagation on BGP (QPPB)
• Route maps
• Queuing mechanisms
 Modular QoS CLI (first implemented in 12.0(5)T) separates classifiers from other
actions
• Includes all traditional classifiers and network-based application recognition
(NBAR)

DiffServ Mechanisms in IOS (cont.)
Meter

Inbound
 Token bucket model is used for metering:
• Committed access rate (CAR)
• Generic traffic shaping (GTS)
• Frame Relay traffic shaping (FRTS)
• Class-based weighted fair queuing (CBWFQ)
• Class-based low latency queuing (CBLLQ)
• Class-based policing
• Class-based shaping
• IP RTP Prioritization

Meter

Inbound
 Marker is used to set: • Marking mechanisms:

• IP Precedence
• DSCP – Comitted access rate (CAR)
• QoS group – QoS Policy Propagation on
• MPLS experimental bits BGP (QPPB)
• Frame Relay DE bit
• ATM CLP bit – Policy-based routing (PBR)
• IEEE 802.1Q or ISL CoS – Class-based marking

Comparison of Markers
Marker Preservation Value Range
IP Precedence Throughout a network 8 values, 2 reserved

(0 to 7)
DSCP Throughout a network 64 values, 32 are standard

(0 to 63)
QoS group Local to a router 100 values

(0 to 99)
MPLS experimental bits Throughout an MPLS network 8 values
(optionally throughout an
entire IP network)
Frame Relay DE bit Throughout a Frame Relay 2 values
network (0 or 1)
ATM CLP bit Throughout an ATM 2 values
network (0 or 1)
IEEE 802.1Q or ISL CoS Throughout a LAN 8 values
switched network (0 to 7)
Meter

Inbound
 Shaping mechanisms:
• Generic traffic shaping (GTS)
• Frame Relay traffic shaping (FRTS)
• Hardware shaping on ATM VC

Meter

Inbound
 Dropping mechanisms:
• Committed access rate (CAR) and class-based policing can drop
packets that exceed the contractual rate.
• Weighted random early detection (WRED) can randomly drop packets
when an interface is nearing congestion.

Meter
Classifier Marker Conditioner Forwarding Queuing

Inbound
 Cisco Express Forwarding (CEF) is recommended from

IOS 12.0.
 Some QoS features work only in combination with CEF.

Meter

Inbound
 Traditional queuing mechanisms

• FIFO, priority queuing (PQ), custom queuing (CQ)
 Weighted fair queuing (WFQ) family
• WFQ, DWFQ, CoS-based DWFQ, QoS-group DWFQ
 Advanced queuing mechanisms
• Class-based WFQ, Class-based LLQ

Meter

Inbound
 Tail drop is used for most queue congestion.

 WFQ has an improved tail-drop scheme.
 WRED randomly drops packets when nearing
congestion.

Summary

to:
 Describe different classification options in IP networks
 Describe different marking options in IP networks
 List the mechanisms that are capable of measuring the rate of
traffic
 List the mechanisms that are used for traffic conditioning,
shaping, and avoiding congestion

Review Questions
1.Name the QoS building blocks.
2.What is the purpose of classification?
3.What is the purpose of marking?
4.Which parameters can be used to mark packets?
5.Which mechanisms can classify and mark packets?
6.Which mechanisms have the ability to measure the rate of
traffic?
7.How, when, and where do routers drop packets?

Enterprise Network
Case Study
Objectives

 Describe the typical structure of an enterprise network
 Describe the need for QoS in enterprise networks
 List typical QoS requirements in enterprise networks
 List the QoS mechanisms that are typically used in enterprise
networks

Traditional
Enterprise Networks
Core
(central sites
and
data centers)
X.25 (ancient), Frame Relay (old),

ATM (newer)
Distribution
(regional centers)
X.25 (ancient), Frame Relay (old),

ATM (newer)
Access
(branch offices)
 Traditional enterprise networks use a hub-and-spoke topology.

 Redundant connections are used to improve resilience.
 A partial mesh can be used between the core sites and the distribution sites.

Modern
Enterprise Networks
Core
(central sites
and
data centers)
MPLS/VPN (new)
Access
(branch offices)
 Modern enterprise networks use a full mesh topology provided by an MPLS/VPN backbone.
 Redundant connections to the backbone can be used to improve resilience
 The MPLS/VPN backbone uses redundant connections and a partial mesh to improve
resilience.

QoS in Enterprise Networks
 Typical enterprise networks have a large number of

different applications.
 Some applications are business-critical and require
some guarantees (bandwidth, delay).
 The network should provide enough resources to
these business-critical applications.
 Applications are usually identified based on TCP or
UDP port numbers.

Case Study
 Typical line speeds:

• Core to Distribution < 2 Mbps
• Distribution to Branch 64 kbps - 256 kbps
 Typical protocols:
• SNA, NetBIOS, desktop protocols (IPX), some TCP/IP,
voice, multimedia
 Typical QoS requirements:
• SNA and voice are high priority
• Guaranteed bandwidth for some applications
• Rest of the traffic is best-effort

Case Study
Implementation #1
 Core to Distribution:
• Custom queuing
 Distribution to Branch:
• Priority queuing or
• Custom queuing with a priority queue
 Options:
• Traffic shaping
• Adaptation to Frame Relay congestion notification

Case Study
Implementation #2
 Core to Distribution:
• Class-based low-latency queuing (CBLLQ)
 Distribution to Branch:
• Class-based low-latency queuing (CBLLQ)
 Options:
• Adaptation to Frame Relay congestion notification
• Class-based policing
• Weighted random early detection (WRED)

Summary

to:
 Describe the typical structure of an enterprise network
 Describe the need for QoS in enterprise networks
 List typical QoS requirements in enterprise networks
 List the QoS mechanisms that are typically used in
enterprise networks

Review Questions
1.What is the typical enterprise network topology?

2.How is resilience achieved?
3.Based on what information do typical enterprise
networks apply QoS?

Services Provider
Case Study
Objectives
 Upon completing this lesson, you will be

able to:
 Describe the typical structure of a service provider
network
 Describe the need for QoS in service provider
networks
 List typical QoS requirements in service provider
networks
 List the QoS mechanisms that can be used in service
provider networks

Typical
Service Provider Networks
ATM, SONET/SDH, DPT, GE, ...

Partial mesh Core
Rings
Redundant connections
ATM, SONET/SDH, DPT, GE, ... Rings
Distribution
(regional POPs)
Single connections
Frame Relay, ATM, leased line (analog, TDM), Optional redundant connections
dial-up (PSTN, ISDN, GSM), xDSL, (fast) Ethernet, ... Dial backup
Access
(customers)
 Typical service provider networks use a high-speed partially meshed core (backbone).
 Regional POPs use two or more connections to the core.
 There may be another layer of smaller POPs connected to distribution-layer POPs.
 Customers are usually connected to the service provider via a single point-to-point link (a secondary
link or a dial line can be used to improve resilience).

QoS in Service Provider Networks
 Service providers extend their service offerings by introducing

quality.
 Customers can get bandwidth guarantees
(like CIR in Frame Relay).
 Customers can get delay guarantees
(like CBR in ATM).
 Customers can get preferential treatment in case of congestion
(Olympic service).
 QoS mechanisms have to be deployed where congestion is
likely (usually at the network edge).
 The customer traffic is identified based on source or destination
IP addresses.

Case Study
 A service provider wants to offer bronze,

silver, gold and premium services:
 Bronze gets 10% of available bandwidth
 Silver gets 20% of available bandwidth
 Gold gets 30% of available bandwidth
 Premium gets 40% of available bandwidth with a low-
delay guarantee

Case Study
Implementation
 Class-based weighted fair queuing (CBWFQ) on slow

to moderate-speed links
 Class-based low latency queuing (CBLLQ) on slow to
moderate-speed links
 Weighted random early detection (WRED) on fast links

Summary
 Upon completing this lesson, you should

be able to:
 Describe the typical structure of a service provider
network
 Describe the need for QoS in service provider
networks
 List typical QoS requirements in service provider
networks
 List the QoS mechanisms that can be used in service
provider networks

Review Questions
1.What is the typical topology of service provider

networks?
2.How is resilience achieved?
3.Based on what information do typical service provider
networks apply QoS?

Module Summary
 Upon completing this module, you should be able

to:
 Describe the need for IP QoS
 Describe the Integrated Services model
 Describe the Differentiated Services model
 Describe the building blocks of IP QoS mechanisms
(classification, marking, metering, policing, shaping, dropping,
forwarding, queuing)
 List the IP QoS mechanisms available in Cisco IOS
 Describe what QoS features are supported by different IP QoS
mechanisms

Thank You
and
Good-bye !!
Cisco’s Certification Track

QoS Presentation2

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

QoS Presentation2

Uploaded by

Copyright:

Available Formats

IP Quality of Service

 Upon completing this module, you will be able to:

Inner Mongolia University

 Upon completing this lesson, you will be able to:

Inner Mongolia University

5 Inner Mongolia University

6 Inner Mongolia University

 Lack of bandwidth?: Multiple flows are contesting for

7 Inner Mongolia University

BWmax = min(10M, 256k, 512k, 100M)=256 kbps

8 Inner Mongolia University

Propagation Propagation Propagation

9 Inner Mongolia University

Processing Delay Queuing Delay

10 Inner Mongolia University

11 Inner Mongolia University

IP TCP Data FIFO queuing

TCP Header Compression

IP UDP RTP Data FIFO queuing

Weighted Random Early Detection (WRED)

IP Data Dropper Fancy Queuing

Custom Queuing (CQ)

14 Inner Mongolia University

Throughput Delay Loss Jitter

Voice Low Low and Low Low

Video High Low and Low Low

Throughput Delay Loss Jitter

Gold Guaranteed Low Low Low

 Service provider networks typically offer services based on

 Best effort—no QoS is applied to packets (default

17 Inner Mongolia University

Inner Mongolia University

1.What are the relevant parameters that define quality

Inner Mongolia University

Inner Mongolia University

22 Inner Mongolia University

Local Remote Admission Local

reserve reserve reserve reserve

 Resource reservation is used to identify an application (flow) and signal

23 Inner Mongolia University

 The Resource Reservation Protocol (RSVP) was

24 Inner Mongolia University

 RSVP is typically used by applications carrying voice

25 Inner Mongolia University

26 Inner Mongolia University

27 Inner Mongolia University

29 Inner Mongolia University

30 Inner Mongolia University

Inner Mongolia University

Inner Mongolia University

 Upon completing this lesson, you will be able to:

Inner Mongolia University

35 Inner Mongolia University

 Wide variety of services and provisioning policies

36 Inner Mongolia University

 The service defines QoS requirements and

37 Inner Mongolia University

 QoS does not create bandwidth!

38 Inner Mongolia University

Traffic Stream = set of flows

Behavior Aggregate (flows with the same DSCP)

39 Inner Mongolia University

 Flow: a single instance of an application-to-application flow