ITU-T Y 1560-2003 Parameters for TCP connection performance in the presence of middleboxes SERIES Y GLOBAL INFORMATION INFRASTRUCTURE INTERNET PROTOCOL ASPECTS AND NEXT GENERATION of.pdf

《ITU-T Y 1560-2003 Parameters for TCP connection performance in the presence of middleboxes SERIES Y GLOBAL INFORMATION INFRASTRUCTURE INTERNET PROTOCOL ASPECTS AND NEXT GENERATION of.pdf》由会员分享，可在线阅读，更多相关《ITU-T Y 1560-2003 Parameters for TCP connection performance in the presence of middleboxes SERIES Y GLOBAL INFORMATION INFRASTRUCTURE INTERNET PROTOCOL ASPECTS AND NEXT GENERATION of.pdf（18页珍藏版）》请在麦多课文档分享上搜索。

1、 INTERNATIONAL TELECOMMUNICATION UNION ITU-T Y.1560TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (09/2003) SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE, INTERNET PROTOCOL ASPECTS AND NEXT GENERATION NETWORKS Internet protocol aspects Quality of service and network performance Parameters for TCP con

2、nection performance in the presence of middleboxes ITU-T Recommendation Y.1560 ITU-T Y-SERIES RECOMMENDATIONS GLOBAL INFORMATION INFRASTRUCTURE, INTERNET PROTOCOL ASPECTS AND NEXT GENERATION NETWORKS GLOBAL INFORMATION INFRASTRUCTURE General Y.100Y.199 Services, applications and middleware Y.200Y.29

3、9 Network aspects Y.300Y.399 Interfaces and protocols Y.400Y.499 Numbering, addressing and naming Y.500Y.599 Operation, administration and maintenance Y.600Y.699 Security Y.700Y.799 Performances Y.800Y.899 INTERNET PROTOCOL ASPECTS General Y.1000Y.1099 Services and applications Y.1100Y.1199 Architec

4、ture, access, network capabilities and resource management Y.1200Y.1299 Transport Y.1300Y.1399 Interworking Y.1400Y.1499 Quality of service and network performance Y.1500Y.1599 Signalling Y.1600Y.1699 Operation, administration and maintenance Y.1700Y.1799 Charging Y.1800Y.1899 NEXT GENERATION NETWOR

5、KS Frameworks and functional architecture models Y.2000Y.2099 Quality of Service and performance Y.2100Y.2199 Service aspects: Service capabilities and service architecture Y.2200Y.2249 Service aspects: Interoperability of services and networks in NGN Y.2250Y.2299 Numbering, naming and addressing Y.

6、2300Y.2399 Network management Y.2400Y.2499 Network control architectures and protocols Y.2500Y.2599 Security Y.2700Y.2799 Generalized mobility Y.2800Y.2899 For further details, please refer to the list of ITU-T Recommendations. ITU-T Rec. Y.1560 (09/2003) i ITU-T Recommendation Y.1560 Parameters for

7、 TCP connection performance in the presence of middleboxes Summary This Recommendation defines the end-to-end Transmission Control Protocol (TCP) performance in terms of speed, accuracy, and dependability in an IP-based network with middleboxes, which are nodes terminating TCP connections. Source IT

8、U-T Recommendation Y.1560 was approved by ITU-T Study Group 13 (2001-2004) under the ITU-T Recommendation A.8 procedure on 13 September 2003. Keywords Availability, Internet Protocol (IP), Transmission Control Protocol (TCP), TCP performance. ii ITU-T Rec. Y.1560 (09/2003) FOREWORD The International

9、 Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on

10、 them with a view to standardizing telecommunications on a worldwide basis. The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. The approval of I

11、TU-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 necessary standards are prepared on a collaborative basis with ISO and IEC. NOTE In this Recommendation, the expression “Administration“ is u

12、sed for conciseness to indicate both a telecommunication administration and a recognized operating agency. Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure e.g. interoperability or applicability) and compliance with the

13、 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 equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of

14、any party. INTELLECTUAL PROPERTY RIGHTS ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Pro

15、perty Rights, whether asserted by ITU members or others outside of the Recommendation development process. As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, i

16、mplementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. ITU 2004 All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. ITU-T R

17、ec. Y.1560 (09/2003) iii CONTENTS Page 1 Scope 1 2 References. 1 3 Definitions 1 4 Abbreviations 1 5 Reference model . 2 5.1 Layered model 2 5.2 Reference configuration . 2 5.3 Reference events. 3 6 Performance parameters . 4 6.1 Speed parameters 4 6.2 Accuracy and dependability parameters. 5 6.3 Av

18、ailability parameters 6 Appendix I Middlebox 6 I.1 Introduction 6 I.2 Factors that affect TCP performance 6 I.3 References 6 Appendix II Effectiveness of middleboxes. 8 Appendix III Information on throughput 8 III.1 Issue to define TCP throughput 8 III.2 Data communication throughput 9 III.3 Overall

19、 communication throughput 9 Appendix IV Bibliography 9 ITU-T Rec. Y.1560 (09/2003) 1 ITU-T Recommendation Y.1560 Parameters for TCP connection performance in the presence of middleboxes 1 Scope This Recommendation defines the end-to-end, point-to-point Transmission Control Protocol (TCP) performance

20、 in terms of speed, accuracy, and dependability in an IP-based network with middleboxes, which are nodes terminating TCP connections. TCP performance is specified in an end-to-end timeline of TCP connections composed of three phases: connection establishment, data communication, and connection clear

21、ing. Availability performance is also one of TCP performance. The TCP connection performance that a user can experience through an application can be measured by monitoring the IP-based network. Performance interaction between the TCP and IP layers is for further study. 2 References The following IT

22、U-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 of this Recommendati

23、on 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 Recommendation does not give it

24、, as a stand-alone document, the status of a Recommendation. 1 ITU-T Recommendation I.350 (1993), General aspects of quality of service and network performance in digital networks, including ISDNs. 2 ITU-T Recommendation Y.1540 (2002) (formerly ITU-T Rec. I.380), Internet protocol data communication

25、 service IP packet transfer and availability performance parameters. 3 ITU-T Recommendation Y.1541 (2002), Network performance objectives for IP-based services. 4 IETF RFC 793 (1981) Transmission Control Protocol (TCP). 3 Definitions This Recommendation defines the following terms: 3.1 end-to-end TC

26、P performance: Performance on TCP layer from SRC to DST. 3.2 ports: The portion of a socket that specifies which logical input or output channel of a process is associated with the data. 4 Abbreviations This Recommendation uses the following abbreviations. ACK Acknowledgment DST Destination host FIN

27、 Finish FTP File Transfer Protocol HTTP HyperText Transfer Protocol 2 ITU-T Rec. Y.1560 (09/2003) IETF Internet Engineering Task Force IP Internet Protocol ISDN Integrated Services Digital Network ITU-T International Telecommunication Union Telecommunication Standardization Sector LL Lower Layers, p

28、rotocols and technology supporting the IP layer MP Measurement Point SMTP Simple Mail Transfer Protocol SRC Source host SYN Initiation flag of TCP connection 5 Reference model 5.1 Layered model Figure 1 illustrates the layered nature of the performance of a TCP connection with a middlebox, which ter

29、minates TCP connections. TCP connection is a logical communication path established between a pair of ports and initialized on each side. The performance provided to TCP service users depends on the performance of the IP layer and other layers. Y.1560_F01TCP connectionsTCP layerTCP layer TCP layerIP

30、 layer IP layer IP layer IP layerLLLL LLLink Link Router LinkHigh layerperformanceTCP connectionperformanceIP packetlayer serviceperformance(Y.1540)Lower layerperformance(3 instances)Networkcomponents(HTTP)(FTP)(SMTP)(HTTP)(FTP)(SMTP)SourceHost(SRC)Middle boxDestinationHost(DST)Figure 1/Y.1560 Layer

31、ed model of performance for TCP connection with a middlebox example 5.2 Reference configuration Figure 2 illustrates the relationship between reference events and measurement points for a TCP connection. When there is no middlebox, there are only two edge measurement points. ITU-T Rec. Y.1560 (09/20

32、03) 3 Y.1560_F02IP layerMP MPTCP connectionsIP flow(exit) (entry)MP MPTCP layer MiddleboxLayerIP layerTCP layerHigher-protocollayerIngresseventIP network cloudHigher-protocollayerLower layerEgress eventFigure 2/Y.1560 Relationship between reference events and measurement points 5.3 Reference events

33、Figure 3 illustrates the end-to-end timelines of some TCP connections. Each end-to-end timeline has a connection establishment phase, data communication phase, and connection clearing phase. The connection establishment phase is the duration between the time that the SYN packet is sent from Ingress

34、MP to middlebox, and ACK packet from middlebox is received at Egress MP. The data communication phase is the duration between the time that the first data request packet is sent from Ingress MP to middlebox, and ACK packet from Ingress MP for the last data transfer packet is received at middlebox. T

35、he connection clearing phase is the duration between the time that the FIN packet is sent from Ingress MP to middlebox, and FIN+ACK packet from middlebox is received at Egress MP. The overall communication phase is the duration between the time that the SYN packet is sent from Ingress MP to middlebo

36、x, and FIN+ACK packet from middlebox is received at Egress MP. Overall communication phase consists of communication establishment, data communication, and connection clearing phases. TCP performance reference events and parameters are defined based on these phases and flows. The end-to-end timeline

37、 of TCP connections depends on the TCP version. To define the end-to-end timeline, one should specify the TCP version. NOTE 1 TCP version means IETF RFC 793, Reno, Tahoe, and so on. NOTE 2 Figure 3 is one sample. There are some ACK handling ways in middlebox. See Appendix I. 4 ITU-T Rec. Y.1560 (09/

38、2003) Y.1560_F03data packet A for requestEgress MPdata packet A for requestMiddleboxSYN packetSYN+ACK packetACK packetFIN+ACK packetFIN+ACK packetFIN packetACK packet for packet Adata packet Cdata packet DSYN packetSYN+ACK packetACK packetFIN packetFIN+ACK packetACK packet for packet Adata packet Bd

39、ata packet CACK packet for packet Cdata packet DFIN packetACK packet for packet DFIN packetdata packet BACK packet for packet CFIN+ACK packetACK packet for packet DIngres MPconnectionestablishmentdatacommunicationOverallcommunication connectionclearingFigure 3/Y.1560 End-to-end timelines of TCP conn

40、ections 6 Performance parameters Table 1 illustrates the application of the three generic performance criteria to each of the three TCP functions. Availability performance is also one of TCP performance derived from the three generic performance parameters. 6.1 Speed parameters 6.1.1 Connection esta

41、blishment delay The connection establishment delay is the duration between the time that the SYN packet is sent from Ingress MP to middlebox, and ACK packet from middlebox is received at Egress MP. The connection establishment delay is the duration of the connection establishment phase in Figure 3.

42、ITU-T Rec. Y.1560 (09/2003) 5 6.1.2 Data communication throughput The definition of throughput parameters for data communication is for further study. NOTE See Appendix III.2. Table 1/Y.1560 Generic performance criteria for TCP connection functions Speed Accuracy Dependability Connection establishme

43、nt Connection establishment delay Connection establishment error probability Connection establishment failure probability Data communication Data communication throughput TCP connection Connection clearing Connection clearing delay Overall communication throughput Premature disconnect probability Co

44、nnection clearing failure probability 6.1.3 Connection clearing delay The connection clearing delay is the duration between the time that the FIN packet is sent from Ingress MP to middlebox, and FIN+ACK packet from middlebox is received at Egress MP. The connection clearing delay is the duration of

45、the connection clearing phase in Figure 3. The definition of clearing delay depends on the TCP version and is for further study. 6.1.4 Overall communication throughput The definition of throughput parameters for overall communication is for further study. NOTE See Appendix III.3. 6.2 Accuracy and de

46、pendability parameters 6.2.1 Connection establishment error probability The connection establishment error probability is the ratio of the number of connection establishment attempts that result in an error in connection establishment to the total number of connection establishment attempts in a mea

47、surement period. 6.2.2 Connection establishment failure probability The connection establishment failure probability is the ratio of the number of connection establishment attempts that fail to establish a connection to the total number of connection establishment attempts in a measurement period. 6

48、.2.3 Premature disconnect probability The definition of premature disconnect probability is for further study. 6.2.4 Connection clearing failure probability The connection release failure probability is the ratio of connection clearing failures to the total number of connection release attempts in a

49、 measurement period. 6 ITU-T Rec. Y.1560 (09/2003) 6.3 Availability parameters Decisions on the appropriate primary performance parameters, outage threshold and algorithms for its definition require further detailed study. TCP service unavailability may be one of availability performance parameters. Appendix I Middlebox I.1 Introduction Middleboxes are being discussed by IETF 1. The main ones are: i) packet filtering firewall, ii) application proxy and iii) Network Address Translation (NAT). i) A packet filtering fire