IP Quality of Service详解：Srinivas Vegesna 著

版权申诉

50 浏览量更新于2024-07-05 收藏 5.07MB PDF 举报

《IP服务质量（IP Quality of Service, QoS）：原理与实践》是一本由Srinivas Vegesna编著的研究方案，版权归属于2001年的Cisco Press。本书深入探讨了在互联网协议（Internet Protocol, IP）网络环境中实现QoS的关键技术和策略。作者在书中强调了IP QoS的重要性，旨在帮助读者理解和优化网络中的数据传输效率，确保不同类型服务（如实时应用、多媒体通信等）得到可靠的服务质量保障。在IP QoS的概念部分，书中介绍了如何在无连接的IP网络中处理服务质量问题，包括带宽预留、流量整形、丢包控制等技术手段。这些技术对于实现VoIP（Voice over Internet Protocol，语音通话通过互联网传输）、视频会议、在线游戏等实时应用的流畅运行至关重要。通过细致的理论分析和案例研究，读者可以了解到如何在设计和配置网络基础设施时考虑到不同业务需求的优先级，以确保关键业务不受网络拥塞或延迟的影响。书中还涵盖了QoS标准，如DiffServ（区别服务）和IntServ（集成服务），它们分别提供了不同的服务模型，前者基于类划分流量，后者则更侧重于每个流的精确控制。此外，MPLS（Multiprotocol Label Switching，多协议标签交换）也被提及，作为一种提高QoS的网络技术，它通过在数据包上添加标签来实现快速转发，减少了在网络中的时延和丢包。警告与免责声明部分明确指出，尽管本书竭尽全力提供准确的信息，但并不保证其完整性和适用性。任何依赖于书中的建议或实施可能导致的损失或损害，作者、Cisco Press和Cisco Systems, Inc.均不承担法律责任。读者应根据实际情况，结合专业判断和适当的测试，谨慎地应用书中的内容。《IP Quality of Service》是一本实用的指南，适合网络管理员、网络工程师和系统架构师阅读，以提升他们对IP网络中服务质量管理的理解和实践能力。随着互联网服务的多样化和对服务质量要求的提升，这本书对于保持网络的稳定性和效率具有很高的参考价值。

End-to-End QoS

Layer 2 QoS technologies offer solutions on a smaller scope only and can't provide end-to-end QoS simply

because the Internet or any large scale IP network is made up of a large group of diverse Layer 2

technologies. In a network, end-to-end connectivity starts at Layer 3 and, hence, only a network layer protocol,

which is IP in the TCP/IP-based Internet, can deliver end-to-end QoS.

The Internet is made up of diverse link technologies and physical media. IP, being the layer providing end-to-

end connectivity, needs to map its QoS functions to the link QoS mechanisms, especially of switched

networks, to facilitate end-to-end QoS.

Some service provider backbones are based on switched networks such as ATM or Frame Relay. In this case,

you need to have ATM and Frame Relay QoS-to-IP interworking to provide end-to-end QoS. This enables the

IP QoS request to be honored within the ATM or the frame cloud.

Switched LANs are an integral part of Internet service providers (ISPs) that provide Web-hosting services and

corporate intranets. IEEE 801.1p and IEEE 802.1Q offer priority-based traffic differentiation in switched LANs.

Interworking these protocols with IP is essential to making QoS end to end. Chapter 8

discusses IP QoS

interworking with switches, backbones, and LANs in detail.

MPLS facilitates IP QoS delivery and provides extensive traffic engineering capabilities that help provide

MPLS-based VPNs. For end-to-end QoS, IP QoS needs to interwork with the QoS mechanisms in MPLS and

MPLS-based VPNs. Chapter 9

focuses on this topic.

Objectives

This book is intended to be a valuable technical resource for network managers, architects, and engineers who

want to understand and deploy IP QoS-based services within their network. IP QoS functions are

indispensable in today's scalable, IP network designs, which are intended to deliver guaranteed and

differentiated Internet services by giving control of the network resources and its usage to the network

operator.

This book's goal is to discuss IP QoS architectures and their associated QoS functions that enable end-to-end

QoS in corporate intranets, service provider networks, and, in general, the Internet. On the subject of IP QoS

architectures, this book's primary focus is on the diffserv architecture. This book also focuses on ATM, Frame

Relay, IEEE 801.1p, IEEE 801.1Q, MPLS, and MPLS VPN QoS technologies and on how they interwork with

IP QoS in providing an end-to-end service. Another important topic of this book is MPLS traffic engineering.

This book provides complete coverage of IP QoS and all related technologies, complete with case studies.

Readers will gain a thorough understanding in the following areas to help deliver and deploy IP QoS and

MPLS-based traffic engineering:

• Fundamentals and the need for IP QoS

• The diffserv QoS architecture and its enabling QoS functionality

• The Intserv QoS model and its enabling QoS functions

• ATM, Frame Relay, and IEEE 802.1p/802.1Q QoS technologies—Interworking with IP QoS

• MPLS and MPLS VPN QoS—Interworking with IP QoS

• MPLS traffic engineering

• Routing policies, general IP QoS functions, and other miscellaneous QoS information

QoS applies to any IP-based network. As such, this book targets all IP networks—corporate intranets, service

provider networks, and the Internet.

Audience

The book is written for internetworking professionals who are responsible for designing and maintaining IP

services for corporate intranets and for service provider network infrastructures. If you are a network engineer,

architect, planner, designer, or operator who has a rudimentary knowledge of QoS technologies, this book will

provide you with practical insights on what you need to consider to design and implement varying degrees of

QoS in the network.

This book also includes useful information for consultants, systems engineers, and sales engineers who

design IP networks for clients. The information in this book covers a wide audience because incorporating

some measure of QoS is an integral part of any network design process.

Scope and Limitations

Although the book attempts to comprehensively cover IP QoS and Cisco's QoS functionality, a few things are

outside this book's scope. For example, it doesn't attempt to cover Cisco platform architecture information that

might be related to QoS. Although it attempts to keep the coverage generic such that it applies across the

Cisco platforms, some features relevant to specific platforms are highlighted because the current QoS

offerings are not truly consistent across all platforms.

One of the goals is to keep the coverage generic and up-to-date so that it remains relevant for the long run.

However, QoS in general and Cisco QoS features in particular, are seeing a lot of new developments, and

there is always some scope for a few details to change here and there as time passes.

The case studies in this book are designed to discuss the application and provide some configuration details

on enabling QoS functionality to help the reader implement QoS in his network. It is not meant to replace the

general Cisco documentation. Cisco documentation is still the best resource for complete details on a

particular QoS configuration command.

The case studies in this book are based on a number of different IOS versions. In general, most case studies

are based on 12.0(6)S or a more recent 12.0S IOS version unless otherwise noted. In case of the MPLS case

studies, 12.0(8)ST or a more recent 12.0ST IOS version is used.

Organization

This book consists of four parts: Part I, "IP QoS," focuses on IP QoS architectures and the QoS functions

enabling them. Part II, "Layer 2, MPLS QoS—Interworking with IP QoS,"

lists the QoS mechanisms in

ATM, Frame Relay, Ethernet, MPLS, and MPLS VPN and discusses how they map with IP QoS. Part III,

"Traffic Engineering," discusses traffic engineering using MPLS. Finally, Part IV, "Appendixes,"

discusses the modular QoS command-line interface and miscellaneous QoS functions and provides some

useful reference material.

Most chapters include a case study section to help in implementation, as well as a question and answer

section.

Part I

This part of the book discusses the IP QoS architectures and their enabling functions. Chapter 2 introduces

the two IP QoS architectures: diffserv and intserv, and goes on to discuss the diffserv architecture.

Chapters 3,

4, 5, and 6 discuss the different functions that enable diffserv architecture. Chapter 3, for

instance, discusses the QoS functions that condition the traffic at the network boundary to facilitate diffserv

within the network. Chapters 4

and 5 discuss packet scheduling mechanisms that provide minimum

bandwidth guarantees for traffic. Chapter 6

focuses on the active queue management techniques that

proactively drop packets signaling congestion. Finally, Chapter 7

discusses the RSVP protocol and its two

integrated service types.

Part II

This section of the book, comprising Chapters 8 and 9, discusses ATM, Frame Relay, IEEE 801.1p, IEEE

801.1Q, MPLS, and MPLS VPN QoS technologies and how they interwork to provide an end-to-end IP QoS.

Part III

Chapter 10, the only chapter in Part III, talks about the need for traffic engineering and discusses MPLS

traffic engineering operation.

Part IV

This part of the book has useful information that didn't fit well with previous sections but still is relevant in

providing IP QoS.

Appendix A, "Cisco Modular QoS Command-Line Interface,"

details the new user interface that

enables flexible and modular QoS configuration.

Appendix B, "Packet Switching Mechanisms,"

introduces the various packet-switching mechanisms

available on Cisco platforms. It compares the switching mechanisms and recommends CEF, which also is a

required packet-switching mechanism for certain QoS features.

Appendix C, "Routing Policies,"

discusses QoS routing, policy-based routing, and QoS Policy Propagation

using Border Gateway Protocol (QPPB).

Appendix D, "Real-Time Transport Protocol (RTP),"

talks about the transport protocol used to carry real-

time packetized audio and video traffic.

Appendix E, "General IP Line Efficiency Functions,"

talks about some IP functions that help improve

available bandwidth.

Appendix F, "Link Layer Fragmentation and Interleaving,"

discusses fragmentation and interleaving

functionality with the Multilink Point-to-Point protocol.

Appendix G, "IP Precedence and DSCP Values,"

tabulates IP precedence and DSCP values. It also

shows how IP precedence and DSCP values are mapped to each other.

References

1. RFC 791: "Internet Protocol Specification," J. Postel, 1981

2. RFC 896: "Congestion Control in IP/TCP Internetworks," J. Nagle, 1984

3. RFC 1122: "Requirements for Internet Hosts—Communication Layers," R. Braden, 1989

4. RFC 2001: "TCP Slow Start, Congestion Avoidance, Fast Retransmit, and Fast Recovery Algorithms,"

W. Stevens, 1997

5. S. Floyd and V. Jacobson. "Random Early Detection Gateways for Congestion Avoidance." IEEE/ACM

Transactions on Networking, August 1993

6. A. Demers, S. Keshav, and S. Shenkar. "Design and Analysis of a Fair Queuing Algorithm."

Proceedings of ACM SIGCOMM '89, Austin, TX, September 1989

7. IETF Intserv Working Group, http://www.ietf.org/html.charters/intserv-charter.html

8. IETF DiffServ Working Group, http://www.ietf.org/html.charters/diffserv-charter.html

9. IETF MPLS Working Group, http://www.ietf.org/html.charters/mpls-charter.html

Chapter 2. Differentiated Services Architecture

The aim of IP Quality of Service (QoS) is to deliver guaranteed and differentiated services on the Internet or

any IP-based network. Guaranteed and differentiated services provide different levels of QoS, and each

represents an architectural model for delivering QoS.

This chapter primarily focuses on Differentiated Services (diffserv) architecture for delivering QoS in the

Internet. The other architectural model, Integrated Services (intserv) is introduced. Intserv is discussed in

Chapter 7, "Integrated Services: RSVP."

An operational QoS model for a network node is also

presented.

Intserv Architecture

The Internet Engineering Task Force (IETF) set up the intserv Working Group (WG) in 1994 to expand the

Internet's service model to better meet the needs of emerging, diverse voice/video applications. It aims to

clearly define the new enhanced Internet service model as well as to provide the means for applications to

express end-to-end resource requirements with support mechanisms in routers and subnet technologies. It

follows the goal of managing those flows separately that requested specific QoS.

Two services—guaranteed[1]

and controlled load[2]—are defined for this purpose. Guaranteed service

provides deterministic delay guarantees, whereas controlled load service provides a network service close to

that provided by a best-effort network under lightly loaded conditions. Both services are discussed in detail in

Chapter 7.

Resource Reservation Protocol (RSVP) is suggested as the signaling protocol that delivers end-to-end service

requirements[3].

The intserv model requires per-flow guaranteed QoS on the Internet. With the thousands of flows existing on

the Internet today, the amount of state information required in the routers can be enormous. This can create

scaling problems, as the state information increases as the number of flows increases. This makes intserv

hard to deploy on the Internet.

In 1998, the diffserv Working Group was formed under IETF. Diffserv is a bridge between intserv's guaranteed

QoS requirements and the best-effort service offered by the Internet today. Diffserv provides traffic

differentiation by classifying traffic into a few classes, with relative service priority among the traffic classes.

Diffserv Architecture

The diffserv approach[4] to providing QoS in networks employs a small, well-defined set of building blocks

from which you can build a variety of services. Its aim is to define the differentiated services (DS) byte, the

Type of Service (ToS) byte from the Internet Protocol (IP) Version 4 header and the Traffic Class byte from IP

Version 6, and mark the standardized DS byte of the packet such that it receives a particular forwarding

treatment, or per-hop behavior (PHB), at each network node.

The diffserv architecture provides a framework[5]

within which service providers can offer customers a range

of network services, each differentiated based on performance. A customer can choose the performance level

needed on a packet-by-packet basis by simply marking the packet's Differentiated Services Code Point

(DSCP) field to a specific value. This value specifies the PHB given to the packet within the service provider

network. Typically, the service provider and customer negotiate a profile describing the rate at which traffic can

be submitted at each service level. Packets submitted in excess of the agreed profile might not be allotted the

requested service level.

The diffserv architecture only specifies the basic mechanisms on ways you can treat packets. You can build a

variety of services by using these mechanisms as building blocks. A service defines some significant

characteristic of packet transmission, such as throughput, delay, jitter, and packet loss in one direction along a

path in a network. In addition, you can characterize a service in terms of the relative priority of access to

resources in a network. After a service is defined, a PHB is specified on all the network nodes of the network

offering this service, and a DSCP is assigned to the PHB. A PHB is an externally observable forwarding

剩余231页未读，继续阅读

bigtiger

粉丝: 0
资源: 920

IP Quality of Service详解：Srinivas Vegesna 著

Cisco Press： IP Quality of Service - Vegesna.pdf

ORAN-WG4.CUS.0-v01.00.pdf

ISO-IEC-14496-8. MPEG-4 Part 8. Carriage of MPEG4 contents over IP networks.pdf

Power-quality-monitor-research.rar_单片机开发_Others_

PTN技术--论文.pdf

MPLS技术架构--英文.pdf

IR.92-v11.0.pdf

H3CSE-RS-372.pdf

MQTT-V5.0-CN.pdf

cos-qos.pdf

最新资源