相关标准及路由器测试.ppt

第八章第八章网络设备测试网络设备测试测试标准及路由器测试测试标准及路由器测试o目标目标：n了解相关标准了解相关标准n学习路由器的常用指标学习路由器的常用指标及测试方法及测试方法o主要内容主要内容n1相关标准相关标准n2测量方法概述测量方法概述n3路由器的性能指标路由器的性能指标n4路由器性能测试路由器性能测试n5测试实例测试实例12/14/20212/14/2022 21 11 1 相关标准相关标准oRFC2544RFC2544(Benchmarking Methodology for Network (Benchmarking Methodology for Network Interconnect Devices)Interconnect Devices)oRFC2889RFC2889（BenchmarkingBenchmarkingMethodologyMethodologyforforLANLANSwSwitchingitchingDevicesDevices）oYD/T1156-2001YD/T1156-2001路由器测试规范高端路由器路由器测试规范高端路由器oYD/T1098-2001YD/T1098-2001路由器测试规范低端路由器路由器测试规范低端路由器oYD/T1141-2001YD/T1141-2001 千兆比以太网交换机测试方法千兆比以太网交换机测试方法oYD/T1142-2001YD/T1142-2001 IP IP电话网守设备技术要求及测试方法电话网守设备技术要求及测试方法oYD/T1072-2000YD/T1072-2000 IP IP电话网关设备测试方法电话网关设备测试方法oYD/T1075-2000YD/T1075-2000 网络接入服务器（网络接入服务器（NASNAS）测试方法）测试方法oYD/T1171-2001 IPYD/T1171-2001 IP网络技术要求网络技术要求网络性能参数与指标网络性能参数与指标12/14/20212/14/2022 22 2相关标准相关标准-续续oYD/T1251.1-2003YD/T1251.1-2003 路由协议一致性测试方法路由协议一致性测试方法中间系中间系统到中间系统路由交换协议（统到中间系统路由交换协议（IS-IS)IS-IS)oYD/T1251.2-2003YD/T1251.2-2003 路由协议一致性测试方法路由协议一致性测试方法开放最开放最短路径优先协议（短路径优先协议（OSPFOSPF）oYD/T1251.3-2003YD/T1251.3-2003 路由协议一致性测试方法路由协议一致性测试方法边界网边界网关协议（关协议（BGP4BGP4）oYD/T1260-2003YD/T1260-2003 基于端口的虚拟局域网（基于端口的虚拟局域网（VLANVLAN）技术要）技术要求和测试方法求和测试方法oYD/T1033-2000YD/T1033-2000 传输性能的指标系列传输性能的指标系列oYD/T10912000YD/T10912000 56kbit/s 56kbit/s调制解调器接口及传输性能调制解调器接口及传输性能技术要求和测试方法技术要求和测试方法12/14/20212/14/2022 23 3Request for Comments:2544Request for Comments:2544oNetwork Working GroupNetwork Working Group，Benchmarking Methodology Benchmarking Methodology for Network Interconnect Devicesfor Network Interconnect Devices，S.Bradner S.Bradner，Harvard UniversityHarvard University，J.McQuaid J.McQuaid，NetScout NetScout SystemsSystems，March 1999 March 1999oThis document discusses and defines a number of This document discusses and defines a number of tests that may be used to describe the perfo-tests that may be used to describe the perfo-rmance characteristics of a network intercon-rmance characteristics of a network intercon-necting device.necting device.In addition to defining the In addition to defining the tests this document also describes specific tests this document also describes specific formats for reporting the results of the tests.formats for reporting the results of the tests.12/14/20212/14/2022 24 4RFC2544 Benchmarking tests 1RFC2544 Benchmarking tests 1：ThroughputThroughputoDefinition:Definition:The maximum rate at which none of the The maximum rate at which none of the offered frames are dropped by the offered frames are dropped by the device.device.o详见：详见：RFC 1242RFC 1242o吞吐量测试可以确定被测试设备吞吐量测试可以确定被测试设备(DUT)(DUT)或被或被测试系统测试系统(SUT)(SUT)在不丢弃包的情况下所能支在不丢弃包的情况下所能支持的吞吐速率。持的吞吐速率。o二分法二分法12/14/20212/14/2022 25 5Throughput test-ProcedureThroughput test-ProcedureoSend a specific number of frames at a Send a specific number of frames at a specific rate through the DUT and then count specific rate through the DUT and then count the frames that are transmitted by the DUT.the frames that are transmitted by the DUT.oIf the count of offered frames is equal to If the count of offered frames is equal to the count of received frames,the fewer the count of received frames,the fewer frames are received than were transmitted,frames are received than were transmitted,the rate of the offered stream is reduced the rate of the offered stream is reduced and the test is rerun.and the test is rerun.oThe throughput is the fastest rate at which The throughput is the fastest rate at which the count of test frames transmitted by the the count of test frames transmitted by the DUT is equal to the number of test frames DUT is equal to the number of test frames sent to it by the test equipment.sent to it by the test equipment.12/14/20212/14/2022 26 6RFC2544 Benchmarking tests 2RFC2544 Benchmarking tests 2：LatencyLatencyDefinition:Definition:oFor For store and forward devicesstore and forward devices:The time interval starting when the The time interval starting when the last bit of the last bit of the input frameinput frame reaches the input port and ending when the reaches the input port and ending when the first bit of the output framefirst bit of the output frame is seen on the output is seen on the output port.port.oFor For bit forwarding devicesbit forwarding devices:The time interval starting when The time interval starting when the end of the first bit the end of the first bit of the input frameof the input frame reaches the input port and ending reaches the input port and ending when when the start of the first bit of the output framethe start of the first bit of the output frame is is seen on the output port.seen on the output port.由此定义看：仅包括排队和处理时延，不包括传输时延？由此定义看：仅包括排队和处理时延，不包括传输时延？12/14/20212/14/2022 27 7延时测试延时测试o延时测试延时测试测量一台测量一台DUT DUT 在有负载条件下转发数据包所需在有负载条件下转发数据包所需的时间。的时间。SmartApplications SmartApplications 在规定时间内生成在规定时间内生成100%100%的的负载负载(或者按测试设置中规定的比例或者按测试设置中规定的比例)。在测试过程中，。在测试过程中，SmartApplications SmartApplications 测量每对端口上的每一个包的延时。测量每对端口上的每一个包的延时。o对于对于存储转发（存储转发（Store-and-forwardStore-and-forward）设备来说，测量设备来说，测量的延时是指从输入帧的最后一个比特达到输入端口的时的延时是指从输入帧的最后一个比特达到输入端口的时刻到输出帧的第一个比特出现在输出端口上的时刻的时刻到输出帧的第一个比特出现在输出端口上的时刻的时间间隔。对于间间隔。对于cut-throughout cut-throughout 设备来说，延时是指从设备来说，延时是指从输入帧的第一比特达到输入端口的时刻到输出帧的第一输入帧的第一比特达到输入端口的时刻到输出帧的第一比特出现在输出端口的时刻的间隔。测试设置中规定的比特出现在输出端口的时刻的间隔。测试设置中规定的每一种包长度都要进行延时测试。每一种包长度都要进行延时测试。12/14/20212/14/2022 28 8Latency test-Procedure:Latency test-Procedure:oSend a stream of frames at a particular frame size Send a stream of frames at a particular frame size through the DUT at the determined throughput rate to a through the DUT at the determined throughput rate to a specific destination.The stream SHOULD be at least specific destination.The stream SHOULD be at least 120 seconds in duration.120 seconds in duration.oAn identifying tag SHOULD be included in one frame An identifying tag SHOULD be included in one frame after 60 seconds with the type of tag being after 60 seconds with the type of tag being implementation dependent.The time at which this frame implementation dependent.The time at which this frame is fully transmitted is recorded(is fully transmitted is recorded(timestamp Atimestamp A).The).The receiver logic in the test equipment MUST recognize the receiver logic in the test equipment MUST recognize the tag information in the frame stream and record the time tag information in the frame stream and record the time at which the tagged frame was received(at which the tagged frame was received(timestamp Btimestamp B).).oThe latency is timestamp B minus timestamp AThe latency is timestamp B minus timestamp A。RFC 1242RFC 1242oThe test MUST be repeated at least 20 times with the The test MUST be repeated at least 20 times with the reported value being the average of the recorded reported value being the average of the recorded values.values.12/14/20212/14/2022 29 9RFC2544 Benchmarking tests 3RFC2544 Benchmarking tests 3：Frame Frame loss rateloss rateoDefinition:Definition:Percentage of frames Percentage of frames thatthat should have been should have been forwarded by a network device under steady state forwarded by a network device under steady state(constant)load(constant)load thatthat were not forwarded due to were not forwarded due to lack of resourceslack of resources.（如何理解两个如何理解两个that?that?并列修饰。并列修饰。）oRFC 1242RFC 1242o包丢失测试通过测量由于缺少资源而未转发的包的包丢失测试通过测量由于缺少资源而未转发的包的比例来显示高负载状态下比例来显示高负载状态下DUTDUT的性能。的性能。12/14/20212/14/2022 21010Frame Loss rate-test ProcedureFrame Loss rate-test ProcedureoSend a Send a specific number of framesspecific number of frames at a at a specific ratespecific rate through the DUT to be tested and count the frames that through the DUT to be tested and count the frames that are transmitted by the DUT.The frame loss rate at are transmitted by the DUT.The frame loss rate at each point is calculated using the following equation:each point is calculated using the following equation:(input_count-output_count)*100)/input_count (input_count-output_count)*100)/input_countoThe first trial SHOULD be run for the frame rate that The first trial SHOULD be run for the frame rate that corresponds to 100%of the maximum rate for the frame corresponds to 100%of the maximum rate for the frame size on the input media.Repeat the procedure for the size on the input media.Repeat the procedure for the rate that corresponds to 90%of the maximum rate used rate that corresponds to 90%of the maximum rate used and then for 80%of this rate.This sequence SHOULD be and then for 80%of this rate.This sequence SHOULD be continued(at reducing 10%intervals)until there are continued(at reducing 10%intervals)until there are two successive trials in which no frames are lost.two successive trials in which no frames are lost.12/14/20212/14/2022 21111RFC2544 Benchmarking tests 4RFC2544 Benchmarking tests 4：Back-Back-to-backto-backoDefinition:Definition:nFixed length frames presented at a rate Fixed length frames presented at a rate such that there is the minimum legal such that there is the minimum legal separation for a given medium between separation for a given medium between frames over a short to medium period of frames over a short to medium period of time,starting from an idle state.time,starting from an idle state.nMeasurement units:Number of N-octet Measurement units:Number of N-octet frames in burstframes in burstoRFC1242RFC124212/14/20212/14/2022 21212背靠背性能测试背靠背性能测试o背靠背性能测试通过背靠背性能测试通过以最大帧速率发送以最大帧速率发送突发传输流突发传输流并并测量无包丢失时的最大突发（测量无包丢失时的最大突发（BurstBurst）长度）长度(总包总包数量数量)来测试来测试DUTDUT的缓冲区容量。的缓冲区容量。oSmartApplications SmartApplications 在全负载条件下生成突发传输在全负载条件下生成突发传输流，如果所有的包都得到转发，就增加突发长度，流，如果所有的包都得到转发，就增加突发长度，并重新进行测试。但是，如果某一对端口上出现包并重新进行测试。但是，如果某一对端口上出现包丢失，丢失，SmartApplicationsSmartApplications将突发长度减少一半并将突发长度减少一半并再次进行测试。然后，再次进行测试。然后，SmartApplicationsSmartApplications利用二利用二分搜索查找无包丢失时的最大突发长度。测试设置分搜索查找无包丢失时的最大突发长度。测试设置中规定的每一种包长度都要进行背到背性能测试。中规定的每一种包长度都要进行背到背性能测试。12/14/20212/14/2022 21313Back-to-backtest procedureBack-to-backtest procedureoSend a burst of frames with minimum inter-frame gaps to Send a burst of frames with minimum inter-frame gaps to the DUT and count the number of frames forwarded by the the DUT and count the number of frames forwarded by the DUT.If the count of transmitted frames is equal to DUT.If the count of transmitted frames is equal to the number of frames forwarded the the number of frames forwarded the length of the burstlength of the burst is increased and the test is rerun.is increased and the test is rerun.oIf the number of forwarded frames is less than the If the number of forwarded frames is less than the number transmitted,the number transmitted,the length of the burstlength of the burst is reduced is reduced and the test is rerun.and the test is rerun.oThe back-to-back value is the number of frames in the The back-to-back value is the number of frames in the longest burst that the DUT will handle without the loss longest burst that the DUT will handle without the loss of any frames.of any frames.The trial length MUST be at least 2 The trial length MUST be at least 2 seconds and SHOULD be repeated at least 50 times with seconds and SHOULD be repeated at least 50 times with the average of the recorded values being reported.the average of the recorded values being reported.12/14/20212/14/2022 21414RFC2544 Benchmarking tests 5RFC2544 Benchmarking tests 5：System recoverySystem recoveryoObjective:Objective:To characterize the speed at which a To characterize the speed at which a DUT recovers from an overload condition.DUT recovers from an overload condition.oProcedure:Procedure:First determine the throughput for a First determine the throughput for a DUT at each of the listed frame sizes.Send a DUT at each of the listed frame sizes.Send a stream of frames at a rate 110%of the recorded stream of frames at a rate 110%of the recorded throughput rate or the maximum rate for the throughput rate or the maximum rate for the mediamedia（线速）（线速）,whichever is lower,for at least,whichever is lower,for at least 60 seconds.At 60 seconds.At Timestamp ATimestamp A reduce the frame reduce the frame rate to 50%of the above rate and record the rate to 50%of the above rate and record the time of the last frame lost (time of the last frame lost (Timestamp BTimestamp B).).The The system recovery time is determined by system recovery time is determined by subtracting Timestamp B from Timestamp A.subtracting Timestamp B from Timestamp A.12/14/20212/14/2022 21515RFC2544 Benchmarking tests 6RFC2544 Benchmarking tests 6：ResetResetoObjective:Objective:To characterize the speed at which a DUT To characterize the speed at which a DUT recovers from a device or software reset.recovers from a device or software reset.oProcedure:Procedure:First determine the throughput for the DUT First determine the throughput for the DUT for the minimum frame size on the media used in the for the minimum frame size on the media used in the testing.Send a continuous stream of frames at the testing.Send a continuous stream of frames at the determined throughput rate for the minimum sized frames.determined throughput rate for the minimum sized frames.Cause a reset in the DUT.Monitor the output until Cause a reset in the DUT.Monitor the output until frames begin to be forwarded and record the time that the frames begin to be forwarded and record the time that the last frame(Timestamp A)of the initial stream and the last frame(Timestamp A)of the initial stream and the first frame of the new stream(Timestamp B)are received.first frame of the new stream(Timestamp B)are received.oThe reset value is obtained by subtracting Timestamp A The reset value is obtained by subtracting Timestamp A from Timestamp B.from Timestamp B.oHardware and software resets,as well as a power Hardware and software resets,as well as a power interruption SHOULD be tested.interruption SHOULD be tested.12/14/20212/14/2022 21616转发能力测试实例转发能力测试实例o思博伦（思博伦（SpirentSpirent）通）通信的信的SmartBits 6000BSmartBits 6000Bo依据依据RFC 2544RFC 2544完成完成n吞吐率吞吐率n延迟延迟n帧丢失率帧丢失率nBack to BackBack to Back测试测试 12/14/20212/14/2022 217173 3 路由器的测试指标路由器的测试指标o1 1、吞吐量、吞吐量：吞吐量是指路由器的包转发能力。吞吐量与路由器的：吞吐量是指路由器的包转发能力。吞吐量与路由器的端口数量、端口速率、数据包长度、数据包类型、路由计算模式端口数量、端口速率、数据包长度、数据包类型、路由计算模式（分布或集中）以及测试方法有关，一般泛指处理器处理数据包（分布或集中）以及测试方法有关，一般泛指处理器处理数据包的能力。高速路由器的包转发能力至少应在的能力。高速路由器的包转发能力至少应在2020包包/秒以上。秒以上。o端口吞吐量端口吞吐量：端口吞吐量是路由器在某端口上的包转发能力。通常采端口吞吐量是路由器在某端口上的包转发能力。通常采用两个相同速率测试接口。一般测试接口可能与接口位置及关系用两个相同速率测试接口。一般测试接口可能与接口位置及关系相关，例如同一插卡上端口间测试的吞吐量可能与不同插卡上端相关，例如同一插卡上端口间测试的吞吐量可能与不同插卡上端口间吞吐量值不同。口间吞吐量值不同。整机吞吐量整机吞吐量：整机吞吐量是指设备整机的包转发能力，是设备性能的整机吞吐量是指设备整机的包转发能力，是设备性能的重要指标。路由器的工作在于根据重要指标。路由器的工作在于根据IPIP包头或者包头或者MPLSMPLS（多协议标签（多协议标签交换）标记选路，因此性能指标是指每秒转发包的数量。整机吞交换）标记选路，因此性能指标是指每秒转发包的数量。整机吞吐量通常小于路由器所有端口吞吐量之和。吐量通常小于路由器所有端口吞吐量之和。12/14/20212/14/2022 21818路由器的测试指标路由器的测试指标o2 2、背板能力、背板能力：背板指输入与输出端口间的物理通：背板指输入与输出端口间的物理通路。路。o背板能力是路由器的内部实现，传统路由器采用共背板能力是路由器的内部实现，传统路由器采用共享背板，但是应用于高性能路由器时不可避免的会享背板，但是应用于高性能路由器时不可避免的会遇到拥塞问题，其次也很难设计出高速的共享总线，遇到拥塞问题，其次也很难设计出高速的共享总线，所以现有的高速路由器一般采用可交换式背板的设所以现有的高速路由器一般采用可交换式背板的设计。计。o背板能力通常大于依据吞吐量和测试包长所计算的背板能力通常大于依据吞吐量和测试包长所计算的值。但是背板能力只能在设计中体现，一般无法测值。但是背板能力只能在设计中体现，一般无法