1、 I n t e r n a t i o n a l T e l e c o m m u n i c a t i o n U n i o n ITU-T G.1030 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (02/2014) SERIES G: TRANSMISSION SYSTEMS AND MEDIA, DIGITAL SYSTEMS AND NETWORKS Multimedia Quality of Service and performance Generic and user-related aspects Estimati
2、ng end-to-end performance in IP networks for data applications Recommendation ITU-T G.1030 ITU-T G-SERIES RECOMMENDATIONS TRANSMISSION SYSTEMS AND MEDIA, DIGITAL SYSTEMS AND NETWORKS INTERNATIONAL TELEPHONE CONNECTIONS AND CIRCUITS G.100G.199 GENERAL CHARACTERISTICS COMMON TO ALL ANALOGUE CARRIER-TR
3、ANSMISSION SYSTEMS G.200G.299 INDIVIDUAL CHARACTERISTICS OF INTERNATIONAL CARRIER TELEPHONE SYSTEMS ON METALLIC LINES G.300G.399 GENERAL CHARACTERISTICS OF INTERNATIONAL CARRIER TELEPHONE SYSTEMS ON RADIO-RELAY OR SATELLITE LINKS AND INTERCONNECTION WITH METALLIC LINES G.400G.449 COORDINATION OF RAD
4、IOTELEPHONY AND LINE TELEPHONY G.450G.499 TRANSMISSION MEDIA AND OPTICAL SYSTEMS CHARACTERISTICS G.600G.699 DIGITAL TERMINAL EQUIPMENTS G.700G.799 DIGITAL NETWORKS G.800G.899 DIGITAL SECTIONS AND DIGITAL LINE SYSTEM G.900G.999 MULTIMEDIA QUALITY OF SERVICE AND PERFORMANCE GENERIC AND USER-RELATED AS
5、PECTS G.1000G.1999 TRANSMISSION MEDIA CHARACTERISTICS G.6000G.6999 DATA OVER TRANSPORT GENERIC ASPECTS G.7000G.7999 PACKET OVER TRANSPORT ASPECTS G.8000G.8999 ACCESS NETWORKS G.9000G.9999 For further details, please refer to the list of ITU-T Recommendations. Rec. ITU-T G.1030 (02/2014) i Recommenda
6、tion ITU-T G.1030 Estimating end-to-end performance in IP networks for data applications Summary Recommendation ITU-T G.1030 provides a framework of tools to obtain IP network performance, estimate the performance of user applications, and apply perceptual models to gauge user satisfaction with the
7、end-to-end performance. The user-perceived performance of data applications on packet networks is dependent on many factors, including the end-to-end performance of the packet network, the applications dependency on the communications network, the performance of the terminals and other devices beyon
8、d the purview of the network operator(s), and the users task and the extent of user interaction with the application. Network designers take these factors into account to assure user satisfaction. Once the application performance has been estimated, perceptual models can be applied to interpret the
9、level of end-to-end performance attained. This Recommendation assumes that the reader will be able to provide at least some level of detail about each of the key factors above, and will then use the framework of tools to estimate end-to-end performance. History Edition Recommendation Approval Study
10、Group Unique ID* 1.0 ITU-T G.1030 2005-11-29 12 11.1002/1000/8672 2.0 ITU-T G.1030 2014-02-13 12 11.1002/1000/12122 _ * To access the Recommendation, type the URL http:/handle.itu.int/ in the address field of your web browser, followed by the Recommendations unique ID. For example, http:/handle.itu.
11、int/11.1002/1000/11830-en. ii Rec. ITU-T G.1030 (02/2014) FOREWORD The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T)
12、is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. The World Telecommunication Standardization Assembly (WTSA), which meets every four years
13、, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. In some areas of information technology which fall within ITU-Ts purview, the necess
14、ary standards are prepared on a collaborative basis with ISO and IEC. NOTE In this Recommendation, the expression “Administration“ is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. Compliance with this Recommendation is voluntary. However,
15、 the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words “shall“ or some other obligatory language such as “must“ and the negative equ
16、ivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. INTELLECTUAL PROPERTY RIGHTSITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a c
17、laimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. As of the date of approval of this Recommendation, ITU ha
18、d not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at http:/www.itu.in
19、t/ITU-T/ipr/. ITU 2014 All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. Rec. ITU-T G.1030 (02/2014) iii Table of Contents Page 1 Scope . 1 2 References . 1 3 Definitions 1 3.1 Terms defined elsewhere 1 3.2 Terms
20、 defined in this Recommendation . 1 4 Abbreviations and acronyms 1 5 Conventions 2 6 End-to-end performance estimation . 2 6.1 Network performance assessment 2 6.2 Application performance assessment . 3 6.3 Perceptual models . 3 6.4 Framework for models in the end-to-end performance assessment proce
21、ss 3 7 Evaluation of end-user QoS 4 Appendix I Estimates of network performance with limited information 5 I.1 Introduction 5 I.2 Reference connection . 5 I.3 Concatenation of packet transfer performance values 5 I.4 Bottleneck bandwidth . 6 I.5 Handshaking time . 8 I.6 Application example: HTTP tra
22、nsaction 10 I.7 Summary . 12 Appendix II Opinion model for a simple web search task 13 II.1 Scope 13 II.2 Introduction 13 II.3 Subjective web browse quality experiment and results 14 II.4 Modelling results 17 II.5 Perceived quality of one-page web browse sessions and single timing events 20 Bibliogr
23、aphy. 22 Rec. ITU-T G.1030 (02/2014) 1 Recommendation ITU-T G.1030 Estimating end-to-end performance in IP networks for data applications 1 Scope This Recommendation provides a framework of tools and the process for estimating end-to-end performance of applications operating on IP networks. 2 Refere
24、nces The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users
25、 of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommenda
26、tion does not give it, as a stand-alone document, the status of a Recommendation. ITU-T G.1010 Recommendation ITU-T G.1010 (2001), End-user multimedia QoS categories. ITU-T G.1031 Recommendation ITU-T G.1031 (2014), QoE factors in web-browsing. ITU-T G.1040 Recommendation ITU-T G.1040 (2006), Networ
27、k contribution to transaction time. ITU-T Y.1540 Recommendation ITU-T Y.1540 (2011), Internet protocol data communication service IP packet transfer and availability performance parameters. 3 Definitions 3.1 Terms defined elsewhere None. 3.2 Terms defined in this Recommendation None. 4 Abbreviations
28、 and acronyms This Recommendation uses the following abbreviations and acronyms: HPT Host Processing Time IPLR IP Packet Loss Ratio MOS Mean Opinion Score MSS Maximum Segment Size NAT Network Address Translation QoS Quality of Service RTT Round-Trip Time TCP Transmission Control Protocol UNI User Ne
29、twork Interface 2 Rec. ITU-T G.1030 (02/2014) 5 Conventions None. 6 End-to-end performance estimation This Recommendation covers the process of estimating end-to-end performance of applications operating on IP networks, using: performance of the IP network of interest, based on relevant measurements
30、 or network modelling results; specifications of the application of interest, in terms of its governing protocol(s) with specified options, or a model of the application using network performance and customer appliance performance as input and producing a key metric of application performance (e.g.,
31、 file download time) as a result; a perceptual model intended for the applications of interest, to interpret the application performance as an estimate of the quality experienced by a typical population of users. Figure 6-1 illustrates the general process used to develop an end-to-end performance es
32、timate Figure 6-1 Process to obtain end-to-end performance estimate The steps to obtain network and application performance may be combined in some cases, such as when a simulation provides a means to measure the performance of a particular session, or set of sessions. 6.1 Network performance assess
33、ment Network performance may be assessed in terms of the packet transfer performance parameters defined in ITU-T Y.1540 and other relevant standards (e.g., RFCs developed by the IETF IP Performance Metrics Working Group). Besides the usual one-way transfer performance metrics, the most straightforwa
34、rd assessment of application performance will sometimes be possible with round-trip metrics instead. There are two principal sources of network performance information: measurement and modelling. Network measurements allow the assessor to treat the network as a black-box, and produce information tha
35、t may be useful in the remaining steps of the modelling chain. However, there are several important considerations for the measurement design, including: 1) The sending discipline needs to match the application of interest in some cases. For example, transmission control protocol (TCP) flow control
36、responds to network conditions and tends to fill the queue at the bottleneck, increasing delay beyond that which would otherwise be measured. 2) The non-measurement load must be similar in size and character to the conditions where the performance estimates are to be applicable. Measurements on an u
37、nloaded network are not particularly useful. Rec. ITU-T G.1030 (02/2014) 3 Network modelling can provide the needed performance characterization when the network is not yet fully constructed (e.g., nodes and links are in place, but not running a critical protocol), or when the key considerations for
38、 measurement are not attainable. There are many choices of modelling tools, including commercial products and public domain research tools. Modelling tools require a substantial degree of expertise and information about the network of interest to be used effectively. As an alternative to this fairly
39、 precise network modelling, Appendix I provides simplified methods which may be used for estimating end-to-end performance in an IP network. However, the accuracy of this method will depend greatly on the accuracy of the information provided. 6.2 Application performance assessment Application models
40、 take the estimates of network performance and information describing application device performance as inputs, and produce one or more key metrics of application performance as outputs. One such application performance model has been specified in ITU-T G.1040 for conversational packet exchanges typ
41、ical of credit card and other point-of-sale transactions. For long-lived file transfers using TCPs reliable byte transfer service, the models described in Appendix I provide accurate results, assuming the inputs are correct. 6.3 Perceptual models This Recommendation includes a limited perceptual mod
42、el for a simple web search task in Appendix II. It should be noted that this model cannot be applied to general web browsing as experienced by the end-user as it does not consider the dynamics of real web browsing. For the more complex case of real web browsing, ITU-T G.1031 defines perceptual event
43、s that have to be considered for creating a perceptual model and goes beyond the events considered in the model described in Appendix II. 6.4 Framework for models in the end-to-end performance assessment process Figure 6-2 illustrates the various alternatives in the process to estimate the end-to-en
44、d performance of applications on IP networks. As indicated in this figure, there are many options available to complete the process, although in practice the assessor must combine options that are consistent with the goal of an end-to-end estimate (and with one another). 4 Rec. ITU-T G.1030 (02/2014
45、) Inputs (beyond application of interest) Network performance assessment Application performance assessment Perceptual performance assessment Access to live network for testing Measurements Ping Unbiased sample Application stream Bulk transfer tool Model for long-lived TCP flows (see Appendix I) Mod
46、el for web browsing (see Appendix II) End-to-end performance estimate Direct assessment with application tool Network design (detailed) Modelling tools Commercial products Research tools Embedded in network modelling tool Future model for video teleconferencing Network design (limited info) Simplifi
47、ed estimation (see Appendix I) ITU-T G.1040 (Note) Other models NOTE ITU-T G.1040 provides the network contribution to transaction application performance, and does not supply the full application performance estimation, yet it gives an application-oriented view of network performance. Figure 6.2 Fr
48、amework for developing end-to-end IP performance estimate 7 Evaluation of end-user QoS After evaluating end-to-end physical performance, it is essential for network planners to evaluate the end-user quality of service (QoS) taking into account perceptual aspects. ITU-T G.1010 provides QoS categories
49、 for data applications as well as voice and video applications. In addition, Appendix II provides an opinion model for estimating users perceived quality of a simple web search task. This enables diagnostic evaluation based on measured/planned end-to-end delay values in terms of customers opinion. Rec. ITU-T G.1030 (02/2014) 5 Appendix I Estimates of network performance with limited information (This appendix does not form an integral part of this Recommendation.) I.1
