WO2018049649A1 - Procédé et dispositif de mesure de performance de réseau - Google Patents

Procédé et dispositif de mesure de performance de réseau Download PDF

Info

Publication number
WO2018049649A1
WO2018049649A1 PCT/CN2016/099211 CN2016099211W WO2018049649A1 WO 2018049649 A1 WO2018049649 A1 WO 2018049649A1 CN 2016099211 W CN2016099211 W CN 2016099211W WO 2018049649 A1 WO2018049649 A1 WO 2018049649A1
Authority
WO
WIPO (PCT)
Prior art keywords
measurement
path
performance
matrix
network
Prior art date
Application number
PCT/CN2016/099211
Other languages
English (en)
Chinese (zh)
Inventor
李刚
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2016/099211 priority Critical patent/WO2018049649A1/fr
Priority to CN201680086452.3A priority patent/CN109314652B/zh
Publication of WO2018049649A1 publication Critical patent/WO2018049649A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4633—Interconnection of networks using encapsulation techniques, e.g. tunneling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/12—Discovery or management of network topologies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0823—Errors, e.g. transmission errors
    • H04L43/0829—Packet loss
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0852—Delays
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0852—Delays
    • H04L43/087—Jitter

Definitions

  • the present invention relates to communication technologies, and in particular, to a network performance measurement method and apparatus.
  • network performance measurement it refers to measurement of network delay, jitter, packet loss rate, throughput, and so on.
  • a tomographic mapping method is proposed to solve network performance. Specifically, firstly, the delay and packet loss rate performance of all services are measured, and then the packet loss rate is converted into a transmission success rate, and the delay is A unified model of transmission success rate is combined with the application of the tomographic mapping method, the minimum overall error target, and a heuristic method is applied to solve the performance of the link.
  • the embodiments of the present invention provide a network performance measurement method and device, which are used to solve the problem of high network performance calculation complexity and excessive overhead in the prior art.
  • a first aspect of the embodiments of the present invention provides a network performance measurement method, which is described from the perspective of a distribution controller.
  • the distribution controller first determines a measurement path set according to the performance measurement target, the network topology information, and the measurement capability information of the node in the network, where the measurement path set includes at least one path to be measured in the network, each of which is to be measured.
  • the measurement path passes through at least one link in the network, and the performance measurement target is used to specify the target and requirement of the measurement, and the measurement capability information of the node in the network includes at least the measurement type supported by the node in the network;
  • the distribution controller measures to the measurement
  • the start node and the end node of the path to be measured in the path set respectively send a measurement request message, where the measurement request message includes at least a measurement object, where the measurement object is used to specify a start node, an end node, and a measurement path identifier of the path to be measured;
  • the distribution controller receives the performance measurement result sent by the start node or the end node, and determines according to the received performance measurement result.
  • Link performance of links in the network is
  • the measurement and analysis server determines the measurement path set according to the performance measurement target, the network topology information, and the measurement capability information of the nodes in the network, and the path in the measurement path set is a partial path in the network, and performance measurement is performed on the paths. All required link performance results in the network are available. Therefore, the method obtains all link performance data in the network through partial measurement, thereby reducing computational complexity and reducing network overhead.
  • a specific method by which the distribution controller determines the set of measurement paths is:
  • the determining a measurement path set from the path matrix according to the performance measurement target and the measurement capability information of a node in the network including:
  • the first node Determining, according to measurement capability information of the node in the network, the first node, where the first node is a node that does not support the measurement type;
  • the solution algorithm comprises a gradient method, a simple method or an interior point method
  • the constraint model includes: the rank of the matrix corresponding to the set of measurement paths is not less than the number of links requiring link performance, the new measurement path covers all links, and the hop count of a single measurement path is not less than 1 hop, The number of measurement paths is not less than the number of links, the start node in the measurement path, and the end node support the measurement type;
  • the target model includes at least: the total number of hops of all measurement paths is the smallest.
  • the solution algorithm further includes a heuristic method; and determining, according to the constraint model and the target model, the measurement path set by using a solution algorithm, including:
  • the constraint model includes: the rank of the matrix corresponding to the set of measurement paths is not less than the number of links requiring link performance, the new measurement path covers all links, and a single measurement path hops The number of measurement paths is not less than 1 hop, the number of measurement paths is not less than the number of links, the start node in the measurement path, and the end node support the measurement type;
  • the target model includes at least: the total number of hops of all measurement paths is the smallest.
  • the determining the link performance of the link in the network according to the performance measurement result includes:
  • the method before determining the measurement path performance matrix according to the performance measurement result, the method further includes:
  • the above method includes:
  • the determining the service performance of the service in the network according to the link performance of the link in the network includes:
  • the value 1 is subtracted from the transmission success rate of the service corresponding to the corresponding row in the path matrix, and the subtraction result is used as the packet loss rate of the service corresponding to the corresponding row in the path matrix.
  • the above method further includes:
  • a second aspect of the embodiments of the present invention provides a network performance measurement method, which is described from the perspective of a node in a network that performs performance measurement, and the method includes:
  • the network performance measurement is performed according to the above measurement request message.
  • the performing network performance measurement according to the measurement request message includes:
  • the performing network performance measurement according to the measurement request message includes:
  • a third aspect of the embodiments of the present invention provides a network performance measuring apparatus, where the apparatus includes:
  • a processing module configured to determine a measurement path set according to the performance measurement target, the network topology information, and the measurement capability information of the node in the network, where the measurement path set includes at least one path to be measured in the network, and each path to be measured passes At least one link in the network, the performance measurement target is used to specify a target and a requirement for measurement, and the measurement capability information of the node in the network includes at least a measurement type supported by a node in the network;
  • a sending module configured to send a measurement request message to the start node and the end node of the path to be measured in the measurement path set, where the measurement request message includes at least a measurement object, and the measurement object is used to specify the to-be-measured The start node, end node, and measurement path identifier of the path;
  • a receiving module configured to receive a performance measurement result sent by the starting node or the ending node
  • the processing module is further configured to determine link performance of a link in the network according to the performance measurement result.
  • the processing module includes:
  • An obtaining unit configured to acquire a service path in the network
  • a generating unit configured to generate a path matrix according to the service path in the network and the network topology information, where each row in the path matrix represents a service in the network;
  • a first determining unit configured to determine, according to the performance measurement target and measurement capability information of a node in the network, a measurement path set from the path matrix, where a rank of a matrix corresponding to the measurement path set is not less than a required link performance The number of links.
  • the first determining unit is specifically configured to:
  • Determining a measurement type according to the performance measurement target determining, according to measurement capability information of a node in the network, the first node is a node that does not support the measurement type; deleting the path matrix from the The first node is used as a path of the start node or the end node to form a first path matrix; the first path matrix is sorted according to the number of links included in each path in the first path matrix to form a second path matrix; orthogonally serializing operations on the paths in the second path matrix to form the set of measurement paths.
  • the processing module further includes:
  • a second determining unit configured to determine a measurement type according to the performance measurement target
  • a third determining unit configured to determine a constraint model and a target model according to the measurement type
  • a fourth determining unit configured to determine, according to the constraint model and the target model, the measurement path set by using a solution algorithm, where the solution algorithm comprises a gradient method, a simple method or an interior point method;
  • the constraint model includes: the rank of the matrix corresponding to the set of measurement paths is not less than the number of links requiring link performance, the new measurement path covers all links, and the hop count of a single measurement path is not less than 1 hop, The number of measurement paths is not less than the number of links, the start node in the measurement path, and the end node support the measurement type;
  • the target model includes at least: the total number of hops of all measurement paths is the smallest.
  • the solving algorithm further includes a heuristic method; the fourth determining unit is specifically configured to:
  • the processing module further includes:
  • a fifth determining unit configured to determine a measurement path performance matrix according to the performance measurement result
  • a calculating unit configured to calculate, according to the measurement path performance matrix and the path matrix, a link performance matrix corresponding to the measurement path performance matrix, where the element value in the link performance matrix is used to identify the performance of the link.
  • processing module is further configured to:
  • the processing module is further configured to determine a service performance of a service in the network according to a link performance of a link in the network.
  • the processing module further includes:
  • a first multiplying unit configured to multiply the path matrix by the link delay matrix, and use an element value in a matrix corresponding to the multiplication result as a delay of a service corresponding to the corresponding row in the path matrix value;
  • a second multiplying unit configured to multiply the path matrix by the link dither value matrix, and use an element value in a matrix corresponding to the multiplication result as a jitter value of a service corresponding to a corresponding row in the path matrix ;
  • a third multiplying unit configured to multiply the path matrix by the link transmission success matrix, and use an element value in a matrix corresponding to the multiplication result as a transmission of a service corresponding to a corresponding row in the path matrix Success rate;
  • the subtraction unit is configured to subtract the value 1 and the transmission success rate of the service corresponding to the corresponding row in the path matrix, and use the subtraction result as the packet loss rate of the service corresponding to the corresponding row in the path matrix.
  • processing module is further configured to:
  • a fourth aspect of the embodiments of the present invention provides a network performance measurement apparatus, including:
  • the receiving module is configured to receive a measurement request message, where the measurement request message includes at least a measurement object.
  • the processing module is configured to perform network performance measurement according to the foregoing measurement application message.
  • the processing module is specifically used to:
  • the measurement request message construct a test message; send a test message.
  • the processing module is also specifically used to:
  • a fifth aspect of the embodiments of the present invention provides a network performance measurement apparatus, including:
  • the memory is configured to store program instructions
  • the processor is configured to invoke program instructions in the memory to perform the following methods:
  • the measurement path set includes at least one path to be measured in the network, and each path to be measured passes through at least one of the networks.
  • the performance measurement target is used to specify a target and a requirement for measurement
  • the measurement capability information of the node in the network includes at least a measurement type supported by a node in the network;
  • the measurement request message includes at least a measurement object, and the measurement object is used to refer to Determining a start node, an end node, and a measurement path identifier of the path to be measured;
  • the link performance of the link in the network is determined.
  • the solution provided by the embodiment of the present invention achieves all link performance data in the network through partial measurement, thereby reducing computational complexity and reducing network overhead.
  • FIG. 1 is a system architecture diagram of a network performance measurement method according to an embodiment of the present invention.
  • Embodiment 1 of a network performance measurement method according to an embodiment of the present invention
  • FIG. 3 is an interaction flowchart of Embodiment 2 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 4 is an interaction flowchart of Embodiment 3 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 5 is an interaction flowchart of Embodiment 4 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 7 is a schematic flowchart of Embodiment 5 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 8 is an interaction flowchart of Embodiment 6 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 9 is a schematic diagram of a loop measurement path
  • Embodiment 7 of a network performance measurement method according to an embodiment of the present invention
  • FIG. 11 is a schematic flowchart of Embodiment 8 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 12 is an interaction flowchart of Embodiment 9 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 13 is an interaction flowchart of Embodiment 10 of a network performance measurement method according to an embodiment of the present disclosure
  • FIG. 14 is a block diagram of a first embodiment of a network performance measurement apparatus according to an embodiment of the present invention.
  • FIG. 15 is a block diagram of a second embodiment of a network performance measurement apparatus according to an embodiment of the present disclosure.
  • FIG. 16 is a block diagram of a third embodiment of a network performance measurement apparatus according to an embodiment of the present disclosure.
  • FIG. 17 is a block diagram of a fourth embodiment of a network performance measurement apparatus according to an embodiment of the present disclosure.
  • FIG. 18 is a block diagram showing a module structure of a fifth embodiment of a network performance measuring apparatus according to an embodiment of the present disclosure
  • FIG. 19 is a physical block diagram of a network performance measurement apparatus according to an embodiment of the present invention.
  • FIG. 20 is a block diagram of a first embodiment of a network node according to an embodiment of the present invention.
  • FIG. 1 is a system architecture diagram of a network performance measurement method according to an embodiment of the present invention.
  • the system architecture includes a measurement analysis server and a plurality of node devices, where multiple node devices are referred to as nodes below.
  • the measurement and analysis server may be a module in a centralized controller in the network, for example, may be a module in a Software Defined Network (SDN), and the measurement analysis server may also be one of a device in the network.
  • SDN Software Defined Network
  • Centralized analysis module, measurement and analysis server can also exist as a distributed cluster to improve the data size and efficiency of the analysis. Therefore, the embodiment of the present invention does not limit the specific form of the measurement and analysis server.
  • the node may be a device with performance measurement function in the network, for example, the node may specifically be a router or a measurement probe device.
  • the nodes are physically connected to each other, as shown by the solid line in FIG. 1 .
  • the measurement and analysis server is logically connected to multiple nodes, as shown by the dotted line in FIG. 1 , so that the measurement application server sends the measurement request to the node.
  • the performance measurement is performed by the node, and the measurement result is reported to the measurement analysis server, and then the measurement analysis server completes the measurement analysis.
  • the router or measurement probe device can perform performance measurement through the internal measurement agent module.
  • routing may also be referred to as “routing”.
  • traffic path may be referred to as “service routing”
  • path matrix may be referred to as “route”. Routing matrix”.
  • Embodiment 1 of a network performance measurement method according to an embodiment of the present invention. As shown in FIG. 2, the method includes:
  • the measurement and analysis server determines the measurement path set according to the performance measurement target, the network topology information, and the measurement capability information of the node in the network.
  • the measurement path set includes at least one path to be measured in the network, and each path to be measured passes through at least one link in the network.
  • the above performance measurement target is used to specify the target and requirement of the measurement, and the measurement capability information of the node in the above network at least includes the measurement type supported by the node in the network.
  • the target and requirement of the measurement specified by the performance measurement target described above may be, for example:
  • VoIP Voice over Internet Protocol
  • the packet loss rate error is not higher than 10 ⁇ 4, and the delay error is not higher than 0.01ms.
  • the targets and requirements of the above measurements may be set by the user.
  • the setting interface or interface may be provided by the measurement analysis server, and saved by the measurement analysis server after being set by the user.
  • the measurement capability information of the node in the network may include, in addition to the measurement types supported by the nodes in the network, the node identifier, the maximum number of supported nodes, and the pipeline object flow sequence corresponding to the node, that is, the route through the node.
  • the measurement types supported by the node may be: Internet Protocol Performance Measurement (IPPM), IP Protocol Performance Monitoring (IP FPM), and RFC 6375. For example, suppose there are nodes A, B, and C, nodes A and B both support IP FPM and IPPM, while node C only supports IPPM.
  • the measurement capability information of the nodes in the network can be reported to the measurement and analysis server by each node in advance, and saved by the measurement and analysis server.
  • the measurement and analysis server may update the measurement capability information of the nodes in the saved network according to a certain policy. For example, the update indication may be sent to each node in the network periodically, and then according to the new measurement capability information reported by each node. Update.
  • the measurement and analysis server may obtain network topology information from devices such as a network management system and an SDN server.
  • the measurement path set can be determined according to the information.
  • a measurement path set is a collection of multiple measurement paths. The path in the measurement path set is part of all the paths in the network, not all of them, so measuring these paths does not cause large network overhead. At the same time, it can be ensured that the number of paths in the measurement path set is not less than the number of links requiring link performance.
  • the measurement and analysis server sends a measurement request message to a start node of the path to be measured in the measurement path set.
  • the measurement analysis server sends a measurement request message to an end node of the path to be measured in the measurement path set.
  • the measurement request message sent by the measurement and analysis server to the start node and the end node includes at least a measurement object, which is used to specify a start node, an end node, and a measurement path identifier of the path to be measured.
  • the measurement analysis server can carry the measurement type and the measurement content in addition to carrying the measurement object in the measurement application message.
  • the corresponding measurement content may include: destination IP, next hop, service priority, protocol type, inbound and outbound label information, etc., and may include packet length, measurement period, number of test packets sent per measurement period, and start Measurement time, end time, immediate execution, maximum performance data/average performance data/performance data patterns including maximum and average performance data.
  • the corresponding measurement content is mainly the end-to-end (E2E) path performance that needs to be calculated under the corresponding measurement type.
  • S102 and S103 are measurement measurement servers that send measurement application messages to a path to be measured in the measurement path set.
  • the measurement measurement applications need to be sent one by one using the processes of S102 and S103.
  • the execution order of S102 and S103 may be in no particular order.
  • the initiating node performs network performance measurement according to the received measurement request message.
  • the end node performs network performance measurement according to the received measurement application message.
  • the initiating node sends a performance measurement result to the measurement and analysis server.
  • the end node sends a performance measurement result to the measurement analysis server.
  • the foregoing steps S104 and S106 are processes for performing performance measurement and transmitting performance measurement results for the initiating node, and the foregoing steps S105 and S107 are performed for the end node for performance and transmission performance measurement.
  • the start node and the end node perform the above steps independently of each other. Therefore, the execution order of the above steps S104 and S105 is not sequential, and accordingly, the execution order of the above steps S106 and S107 is in no particular order.
  • steps S106 and S107 may perform only one of the steps. That is, the performance measurement result may be directly reported to the measurement analysis server by the end node, or the measurement result may be sent to the start node by the end node, and then reported by the start node to the measurement analysis server.
  • the following procedures for reporting performance measurement results follow this principle, and will not be described later.
  • the foregoing performance measurement results may include performance data such as measurement path identifier, delay, jitter, and packet loss rate.
  • the measurement analysis server determines link performance of the link in the network according to the performance measurement result sent by the start node and the performance measurement result sent by the end node.
  • the measurement and analysis server can determine the measurement path set according to the performance measurement target, the network topology information, and the measurement capability information of the nodes in the network, and the path in the measurement path set is part of the path in the network, thereby saving network overhead.
  • the number of paths in the measurement path set is not less than the number of links requiring link performance. Therefore, as long as the performance of each measurement path in the measurement path set is obtained, all the required link performances in the network can be obtained through operation. Link performance.
  • the measurement and analysis server determines the measurement path set according to the performance measurement target, the network topology information, and the measurement capability information of the nodes in the network, and the path in the measurement path set is a partial path in the network, and performance measurement is performed on the paths. All required link performance results in the network are available. Therefore, in this embodiment, all link performance data in the network is obtained by partial measurement, and in the prior art, all the services need to be measured to obtain all link performance data in the network. Therefore, the present embodiment reduces computational complexity and reduces network overhead.
  • FIG. 3 is an interaction flowchart of Embodiment 2 of a network performance measurement method according to an embodiment of the present invention. As shown in FIG. 3, after the foregoing step S108, the method further includes:
  • the measurement and analysis server determines the service performance of the service in the network according to the link performance of the link in the network.
  • the measurement and analysis server can calculate the service performance of the service based on the link performance of the links in the network according to the link performance of the links. For example, if a service A passes through multiple links 1, 2, and 3, the delay of the service is equal to the sum of the delays of the links 1, 2, and 3.
  • FIG. 4 is an interaction flowchart of Embodiment 3 of a network performance measurement method according to an embodiment of the present invention. As shown in FIG. 4, before the step S101, the method further includes:
  • the measurement analysis server acquires performance measurement targets, network topology information, and measurement capability information of nodes in the network.
  • the present embodiment relates to a specific method of determining a measurement path set.
  • the embodiment of the present invention provides two methods, the first one is to determine the measurement path set from the existing service path, and the other is to create a new measurement path set.
  • the measurement analysis server selects one of the specific methods based on the performance measurement target or the like.
  • the specific selection method can be:
  • the first or second method is selected according to the level of measurement accuracy requirements, such as high accuracy requirements, IP FPM measurement, or network performance measurement for a certain type of service, which requires obtaining a service path.
  • the first or second method is selected according to the cost of establishing the measurement path. For example, if the cost is large, the service path needs to be obtained.
  • the first or second method is selected according to the user indication. If the user instructs to obtain the service path, the service path needs to be directly obtained. If the user does not indicate the service path, the service path needs to be obtained by default, or the user may not completely collect the service path. The principle of business path or the inability to create a new measurement path.
  • FIG. 5 is an interaction flowchart of Embodiment 4 of a network performance measurement method according to an embodiment of the present invention.
  • the foregoing first method that is, an implementation manner of the foregoing step S101, includes:
  • S1011 Obtain a service path in the network.
  • the measurement and analysis server may obtain a service path from a device such as an SDN controller or a network management device, or the measurement and analysis server may directly obtain a service path through each node in the network.
  • a service path is a path through which a service passes.
  • S1012 Generate a path matrix according to the service path and network topology information in the network.
  • each row in the path matrix represents a service in the network.
  • the link and path information in the network can be obtained, and the path matrix can be generated by combining the obtained service paths.
  • the path matrix is a two-dimensional matrix, wherein the behavior service is listed as a link in the network, and an element with a value of 1 in the path matrix indicates that the service corresponding to the changed line passes the link corresponding to the column, and the value is 0. The element indicates that the service corresponding to the change does not pass the link corresponding to the column.
  • a path matrix An example of a path matrix.
  • D->A The path of D->A is: D->B->A
  • D->B The path of D->B is: D->B
  • D->C The path of D->C is: D->C
  • E->A The path of E->A is: E->C->A
  • E->B The path of E->B is: E->D->B
  • E->C The path of E->C is: E->C
  • FIG. 6 is a path matrix corresponding to the service path.
  • the rank of the matrix corresponding to the set of measurement paths is not less than the number of links requiring link performance.
  • the path matrix is directly generated according to the service path in the network, and the measurement path set is determined from the generated path matrix. Therefore, the correctness of the determined measurement path set can be ensured.
  • FIG. 7 is a schematic flowchart of Embodiment 5 of a network performance measurement method according to an embodiment of the present invention, as shown in FIG. 7 .
  • the above step S1013 specifically includes:
  • the performance measurement targets specify the objectives and requirements for performance measurements. Based on these requirements, the measurement analysis server can determine the type of measurement for this measurement. For example, if the performance measurement target is high, select IP FPM. The type of measurement, and vice versa, the IPPM measurement type can be selected.
  • S202 Determine, according to measurement capability information of a node in the network, a first node, where the first node is a node that does not support the foregoing measurement type.
  • the measurement and analysis server can obtain the measurement energy information of each node in the network, and the measurement capability information of the node includes the measurement type. After determining the measurement type of this measurement in the foregoing steps, first find out the nodes in the network that do not support this measurement type. For example, if the determined measurement type of this performance measurement is IP FPM, and the measurement type supported by a node in the network is IPPM, it can be determined that the node is the first node.
  • the number of first nodes determined by the measurement analysis server may be multiple.
  • the remaining nodes that are the starting node or the ending node can support the performance measurement, that is, the path in the first path matrix can be guaranteed. Is a valid path.
  • the second path matrix is obtained by sorting more or less by the number of links through which the path passes. After the number of links is ranked in the front, it is possible to select the path hop count greater than 1, and accelerate the calculation of the measurement path.
  • the first row element in the second path matrix is added to the measurement path set, and then, the second row element in the second path matrix is compared with the current measurement path by using a Gram-Schmidt method or the like.
  • the set performs orthogonal serialization. If the orthogonality is not correlated, the second row is added to the measurement path set, and then the third row is selected to perform similar orthogonal calculation with the new measurement path set, and so on, until it finds the required link performance.
  • the number of links is up, that is, when the rank of the matrix corresponding to the measurement path set is not less than the number of links requiring link performance.
  • the rank of the matrix is a concave problem, and it is necessary to transform the trace of the matrix into a convex problem.
  • S201 and subsequent steps may be re-executed, that is, the measurement type of the current performance measurement is re-determined, and the measurement path set is determined based on the new measurement type.
  • the measurement path set that satisfies the requirements is not obtained through the above operation, if a new measurement path can be deployed, a small number of measurement paths may be calculated according to the principle that the uncovered link is not related to the measurement path set to be added to the measurement path. concentrated.
  • the deployment of the newly-built measurement paths can be implemented by using a Label-Switched Path (LSP), Segment Routing, or static IP routing to cover certain specific links. the goal of.
  • LSP Label-Switched Path
  • Segment Routing Segment Routing
  • static IP routing to cover certain specific links. the goal of.
  • FIG. 8 is an interaction flowchart of Embodiment 6 of the network performance measurement method according to the embodiment of the present invention.
  • the foregoing second method that is, another implementation manner of the foregoing step S101, includes:
  • the constraint model includes: the newly-built measurement path covers all links, the metric hop of a single measurement path is not less than 1 hop (it needs to pass through three different device nodes including the original destination and above), and the number of measurement paths is not less than the link.
  • the number, the start node and the end node in all measurement paths support the above measurement types, the measurement path number balance of each node in the measurement path, and the like.
  • the target model includes at least: the minimum total number of hops for all measurement paths.
  • the measurement path set may be determined by using a gradient algorithm, a simple method, an interior point method, a heuristic method, or the like.
  • the method for determining the measurement path set according to the constraint model and the target model is: first selecting a start node and an end node, and generating at least the start node to the end node according to the network topology information. a path; secondly, calculate the shortest hop path between the start node and the end node; then remove the path of no more than 1 hop; and further, determine whether the remaining path satisfies the condition that the number of paths is not less than the number of links requiring link performance If yes, add the remaining path to the measurement path set.
  • a new loop measurement may be performed.
  • Path mode For example, when there is only one measurement node in the network, it is impossible to cover all links by using the start node and the end node as different nodes.
  • the new loop measurement path mode can be used.
  • the constraint model and the target model, and the corresponding solution methods can all adopt the foregoing methods.
  • a simpler heuristic method can be used.
  • FIG. 9 is a schematic diagram of a loop measurement path. The loop shown in FIG. 9 is taken as an example.
  • the method is specifically: first selecting a node that can form a measurement loop, and then designing a measurement loop for each node. Among them, when selecting a node, it is also necessary to select according to the type of measurement, and reference may be made to the foregoing embodiment.
  • the specific process for designing a measurement loop for each node is:
  • loop 1 does not satisfy the condition of passing 3 different nodes, while loop 2 satisfies the condition of passing 3 different nodes.
  • the method for calculating the measurement path from the foregoing loop is similar to the foregoing method for determining the measurement path, and details are not described herein again.
  • FIG. 10 is an interaction flowchart of Embodiment 7 of a network performance measurement method according to an embodiment of the present invention.
  • the foregoing step S108 specifically includes:
  • S1082 Calculate a link performance matrix corresponding to the measurement path performance matrix according to the measurement path performance matrix and the path matrix.
  • the element value in the link performance matrix is used to identify the performance of the link.
  • the measurement path performance matrix is a performance matrix of each measurement path in the measurement path set.
  • the measurement path of each measurement path can be calculated.
  • Performance wherein the performance of each measurement path may include a delay value, a packet loss rate value, and a jitter value.
  • the measurement path performance matrix may specifically include: a measurement path delay matrix, a measurement path loss rate matrix, and a measurement Path jitter value matrix, etc.
  • the measurement path delay matrix can be represented by X d
  • the measurement path loss rate matrix can be represented by X l
  • the measurement path jitter value matrix can be represented by X j .
  • the link performance matrix represents the performance matrix of the link in the network that requires link performance.
  • the link performance matrix may specifically include: a link delay matrix, a link transmission loss rate matrix, a link transmission success rate matrix, and a chain. Road jitter value matrix, etc. If Y is used to represent the link performance matrix, the link delay matrix can be represented by Y d , the link transmission loss rate matrix can be represented by Y ls , and the link transmission success rate matrix can be represented by Y sc , link jitter The value matrix can be represented by Y j . Assume that the number of links in the network that require link performance is m. The dimensions of the above-mentioned several link performance matrices are shown in Table 2.
  • the link performance matrix may be generated after the measurement analysis server obtains the network topology information, and the value of each element in the generated matrix is an invalid value.
  • the calculation is performed according to the measurement path matrix and the path matrix.
  • the measurement analysis server before the step S1081 described above, the measurement analysis server further needs to first determine whether the link is faulty.
  • the measurement and analysis server determines whether there is a faulty link in the measurement path set according to the performance measurement result, and if so, deletes the path through the faulty link from the measurement path set, and selects a new one from the second path matrix by orthogonal serialization operation The path is added to the measurement path set.
  • the measurement analysis server receives the measurement results sent by the start node and the end node of all the measurement paths, the corresponding result is parsed, and the performance data of each measurement path is obtained, if the packet loss rate is equal to 100% or the delay is greater than the pre- If the threshold is set (for example, 1 second), it indicates that there is link congestion or link failure, and the measurement and analysis server identifies the faulty measurement path according to the performance data, and then identifies the link congestion by determining the same link set through which the measurement path passes. Or faulty object.
  • the threshold for example, 1 second
  • the measurement analysis server measures the D->C->E path.
  • the packet loss rate is 100%, the delay is 1 second; the packet loss rate of the A->C->E path is 100%, the delay is 1 second; and the B->D->C path packet loss rate is 0%.
  • the delay is 0.1 milliseconds; the packet loss rate of the A->C->D path is 0%, and the delay is 0.1 milliseconds. Based on the performance data of this path, it can be identified that the C->E link is faulty or the link is heavily congested.
  • the measurement and analysis server measures the measurement path corresponding to the faulty link in the centralized path, and then identifies the orthogonal uncorrelated measurement path from the second path matrix to join the measurement path set according to the foregoing method. The measurement is then re-measured for the new set of measurement paths, or the performance is measured for the path newly added to the measurement path set.
  • FIG. 11 is a schematic flowchart of Embodiment 8 of a network performance measurement method according to an embodiment of the present invention. As shown in FIG. 11 , in step S109 of FIG. 3, determining service performance of a service in a network according to link performance of a link in the network.
  • An alternative implementation is:
  • the link and/or service performance in the network can be predicted by performing the above network performance measurement process multiple times.
  • the foregoing network performance measurement is performed multiple times to obtain a plurality of performance measurement results.
  • the above network performance measurement can be performed periodically.
  • multiple performance measurement results are analyzed.
  • a vector basis, a neural network, or the like may be used for analysis, thereby predicting links and/or links in the network according to the analysis results.
  • each node in the network may change during the actual running process, for example, a node in a certain service path fails, or the network optimizes the service path, the network The business path has changed. In this case, it is necessary to update the above measurement path set.
  • the measurement and analysis server monitors whether the service path corresponding to the measurement path set changes by means of real-time monitoring, etc., if it is determined that the service path changes, it is determined whether the changed measurement path set can cover the link requiring the link performance. If yes, continue to use the original measurement path set. If not, update the measurement path set by using a new service path or a new service path in the network, and perform network performance measurement according to the updated measurement path set.
  • the method of updating the measurement path set also needs to satisfy the principle that the rank of the matrix corresponding to the foregoing measurement path set is not less than the number of links requiring link performance.
  • FIG. 12 is an interaction flowchart of Embodiment 9 of the network performance measurement method according to the embodiment of the present invention, as shown in FIG.
  • the specific process of the network performance measurement of the starting node is as follows:
  • S1041 Construct a test message according to the measurement request message.
  • the initiating node after receiving the measurement request message, the initiating node first parses the measurement request message, and further constructs a test packet for the measurement path information in the message and the corresponding destination IP, the next hop, and other test packet header feature information. . A timestamp can be added to the test message.
  • S1042 Send the test packet.
  • FIG. 13 is an interaction flowchart of the tenth embodiment of the network performance measurement method according to the embodiment of the present invention, as shown in FIG.
  • a specific method for ending node performance measurement is:
  • S1052 Calculate network performance according to the foregoing test packet, and obtain performance measurement results.
  • the end node After receiving the test packet, the end node calculates performance results such as delay and jitter based on the information such as the receiving time of the packet.
  • the end node sends the performance measurement result to the analysis server.
  • the end node may not analyze the received test packet, but directly send the received test packet to the measurement analysis server, and the measurement analysis server uniformly analyzes the method. Performance measurement results.
  • the end node may also analyze the performance measurement result, and send the performance measurement result to the start node, and the start node uniformly sends the measurement result to the measurement analysis server.
  • the foregoing program may be stored in a computer readable storage medium, and when executed, the program includes the steps of the foregoing method embodiment; and the foregoing storage medium includes: ROM, RAM A variety of media that can store program code, such as a disk or a disc.
  • FIG. 14 is a block diagram of a first embodiment of a network performance measurement apparatus according to an embodiment of the present invention. As shown in FIG. 14, the apparatus includes:
  • the processing module 501 is configured to determine, according to the performance measurement target, the network topology information, and the measurement capability information of the node in the network, the measurement path set, where the measurement path set includes at least one path to be measured in the network, and each path to be measured passes through At least one link in the network, the performance measurement target is used to specify the target and requirement of the measurement, and the measurement capability information of the node in the network includes at least the measurement type supported by the node in the network.
  • the sending module 502 is configured to separately send a measurement request message to the start node and the end node of the path to be measured in the measurement path set, where the measurement request message includes at least a measurement object, and the measurement object is used to specify a start node of the path to be measured, and the end Node and measurement path ID.
  • the receiving module 503 is configured to receive a performance measurement result sent by the starting node or the ending node.
  • the processing module 501 is further configured to determine link performance of the link in the network according to the performance measurement result.
  • the device is used to implement the foregoing method embodiments, and the implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 15 is a block diagram of a second embodiment of a network performance measurement apparatus according to an embodiment of the present invention.
  • the processing module 501 includes:
  • the obtaining unit 5011 is configured to acquire a service path in the network.
  • the generating unit 5012 is configured to generate a path matrix according to the service path and the network topology information in the network, where each row in the path matrix represents a service in the network.
  • the first determining unit 5013 is configured to determine, according to the performance measurement target and the measurement capability information of the node in the network, the measurement path set, where the rank of the matrix corresponding to the measurement path set is not less than the link that requires the link performance. number.
  • the first determining unit 5013 is specifically configured to:
  • Determining a measurement type according to the performance measurement target determining, according to the measurement capability information of the node in the network, the first node, the first node being a node that does not support the measurement type; deleting the first node as a starting node from the path matrix or End the path of the node to form the first path matrix; The number of links included in each path in the first path matrix, sorting the first path matrix to form a second path matrix; performing orthogonal serialization operations on the paths in the second path matrix to form a measurement path set .
  • FIG. 16 is a block diagram of a third embodiment of a network performance measurement apparatus according to an embodiment of the present invention. As shown in FIG. 16, the processing module 501 further includes:
  • the second determining unit 5014 is configured to determine a measurement type according to the performance measurement target.
  • the third determining unit 5015 is configured to determine the constraint model and the target model according to the measurement type.
  • the fourth determining unit 5016 is configured to determine a measurement path set according to the constraint model and the target model, where the solution algorithm includes a gradient method, a simple method, or an interior point method.
  • the foregoing constraint model includes at least: the number of the matrix corresponding to the measurement path set is not less than the number of links requiring link performance, the new measurement path covers all links, the metric hop count of a single measurement path is not less than 1 hop, and the number of measurement paths Not less than the number of links, the starting node in the measurement path, and the ending node support the above measurement types.
  • the above target model includes at least: the total number of hops of all measurement paths is the smallest.
  • the above solution algorithm further includes a heuristic method; the fourth determining unit 5016 is specifically configured to:
  • FIG. 17 is a block diagram of a fourth embodiment of a network performance measurement apparatus according to an embodiment of the present invention. As shown in FIG. 17, the processing module 501 further includes:
  • the fifth determining unit 5017 is configured to determine a measurement path performance matrix according to the performance measurement result.
  • the calculating unit 5018 is configured to calculate, according to the measurement path performance matrix and the path matrix, a link performance matrix corresponding to the measurement path performance matrix, where the element value in the link performance matrix is used to identify the performance of the link.
  • processing module 501 is further configured to:
  • processing module 501 is further configured to: determine service performance of services in the network according to link performance of links in the network.
  • FIG. 18 is a block diagram showing a module structure of a network performance measuring apparatus according to Embodiment 5 of the present invention. As shown in FIG. 18, the processing module further includes:
  • the first multiplying unit 5019 is configured to multiply the path matrix by the link delay matrix, and use the element value in the matrix corresponding to the multiplication result as the delay value of the service corresponding to the corresponding row in the path matrix.
  • the second multiplying unit 50110 is configured to multiply the path matrix by the matrix of the link jitter value, and use the element value in the matrix corresponding to the multiplication result as the jitter value of the service corresponding to the corresponding row in the path matrix.
  • the third multiplying unit 50111 is configured to multiply the path matrix by the link transmission success matrix, and use the element value in the matrix corresponding to the multiplication result as the transmission success rate of the service corresponding to the corresponding row in the path matrix.
  • the subtraction unit 50112 is configured to subtract the transmission success rate of the service corresponding to the corresponding row in the value matrix and the path matrix, and use the subtraction result as the packet loss rate of the service corresponding to the corresponding row in the path matrix.
  • processing module 501 is further configured to:
  • FIG. 19 is a physical block diagram of a network performance measurement apparatus according to an embodiment of the present invention. As shown in FIG. 19, the apparatus includes:
  • the memory 601 is used to store program instructions, and the processor 602 is configured to call program instructions in the memory 601 to perform the following methods:
  • the performance measurement target is used to specify the target and requirement of the measurement.
  • the measurement capability information of the node in the network includes at least the measurement type supported by the nodes in the network.
  • the message, the measurement request message includes at least a measurement object, and the measurement object is used to specify a start node, an end node, and a measurement path identifier of the path to be measured.
  • processor 602 is further configured to:
  • processor 602 is further configured to:
  • the first node Determining, according to measurement capability information of the node in the network, the first node, where the first node is a node that does not support the measurement type;
  • processor 602 is further configured to:
  • the solution algorithm comprises a gradient method, a simple method or an interior point method
  • the constraint model includes: the rank of the matrix corresponding to the set of measurement paths is not less than the number of links requiring link performance, the new measurement path covers all links, and the hop count of a single measurement path is not less than 1 hop, The number of measurement paths is not less than the number of links, the start node in the measurement path, and the end node support the measurement type;
  • the target model includes at least: the total number of hops of all measurement paths is the smallest.
  • the above solution algorithm further includes a heuristic method, and correspondingly, the processor 602 is further configured to:
  • processor 602 is further configured to:
  • processor 602 is further configured to:
  • processor 602 is further configured to:
  • the value 1 is subtracted from the transmission success rate of the service corresponding to the corresponding row in the path matrix, and the subtraction result is used as the packet loss rate of the service corresponding to the corresponding row in the path matrix.
  • processor 602 is further configured to:
  • FIG. 20 is a block diagram of a first embodiment of a network node according to an embodiment of the present invention.
  • the network node may be the foregoing start node or end node. As shown in FIG. 20, the network node includes:
  • the receiving module 701 is configured to receive a measurement request message, where the measurement request message includes at least a measurement object.
  • the processing module 702 is configured to perform network performance measurement according to the foregoing measurement request message.
  • processing module 702 is specifically configured to:
  • the measurement request message construct a test message; send a test message.
  • processing module 702 is further specifically configured to:

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Environmental & Geological Engineering (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

L'invention concerne un procédé et un dispositif de mesure de performance de réseau. Le procédé comprend les étapes suivantes : déterminer un ensemble de trajets de mesure selon une cible de mesure de performance, des informations topologiques de réseau et des informations de performance de mesure concernant un nœud dans un réseau, l'ensemble de trajets de mesure comprenant au moins un trajet à mesurer dans le réseau, et chaque trajet à mesurer passant par au moins une liaison dans le réseau ; envoyer respectivement un message d'application de mesure à un nœud de départ et un nœud final du trajet à mesurer dans l'ensemble de trajets de mesure ; et recevoir un résultat de mesure de performance envoyé par le nœud de départ ou le nœud final, et déterminer la performance de liaison d'une liaison dans le réseau en fonction du résultat de mesure de performance reçu. Dans le procédé, toutes les données de performance de liaison dans un réseau sont obtenues grâce à une mesure partielle, de sorte que la complexité de calcul et le temps système de réseau soient réduits.
PCT/CN2016/099211 2016-09-18 2016-09-18 Procédé et dispositif de mesure de performance de réseau WO2018049649A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2016/099211 WO2018049649A1 (fr) 2016-09-18 2016-09-18 Procédé et dispositif de mesure de performance de réseau
CN201680086452.3A CN109314652B (zh) 2016-09-18 2016-09-18 网络性能测量方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/099211 WO2018049649A1 (fr) 2016-09-18 2016-09-18 Procédé et dispositif de mesure de performance de réseau

Publications (1)

Publication Number Publication Date
WO2018049649A1 true WO2018049649A1 (fr) 2018-03-22

Family

ID=61618598

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/099211 WO2018049649A1 (fr) 2016-09-18 2016-09-18 Procédé et dispositif de mesure de performance de réseau

Country Status (2)

Country Link
CN (1) CN109314652B (fr)
WO (1) WO2018049649A1 (fr)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111682986A (zh) * 2020-06-08 2020-09-18 平安科技（深圳）有限公司 全量链路质量探测方法、装置、计算机设备及存储介质
US11165677B2 (en) 2018-10-18 2021-11-02 At&T Intellectual Property I, L.P. Packet network performance monitoring
CN113630171A (zh) * 2021-08-05 2021-11-09 台州学院 一种基于QoS的卫星网络k端可靠性分析方法
CN113709605A (zh) * 2021-08-31 2021-11-26 烽火通信科技股份有限公司 一种光传输网络规划中互斥组业务路由计算方法和装置
US20220158867A1 (en) * 2015-01-28 2022-05-19 Umbra Technologies Ltd. System and method for a global virtual network
WO2022199164A1 (fr) * 2021-03-26 2022-09-29 华为技术有限公司 Procédé de mesure de réseau basé sur un échantillonnage par blocs
US11630811B2 (en) 2016-04-26 2023-04-18 Umbra Technologies Ltd. Network Slinghop via tapestry slingshot
US11681665B2 (en) 2015-12-11 2023-06-20 Umbra Technologies Ltd. System and method for information slingshot over a network tapestry and granularity of a tick
US11711346B2 (en) 2015-01-06 2023-07-25 Umbra Technologies Ltd. System and method for neutral application programming interface
US11750419B2 (en) 2015-04-07 2023-09-05 Umbra Technologies Ltd. Systems and methods for providing a global virtual network (GVN)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113067740B (zh) * 2020-01-02 2023-09-05 中国移动通信有限公司研究院 随路性能检测方法、装置、设备及计算机可读存储介质
CN113704963B (zh) * 2021-07-19 2024-04-26 深圳优地科技有限公司 机器人测试方法、系统及机器人

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1260656A (zh) * 1998-10-09 2000-07-19 朗迅科技公司 测试网络性能的方法和装置以及应激反应分析
CN101056217A (zh) * 2006-04-14 2007-10-17 华为技术有限公司 一种网络性能测量方法及系统
CN101599870A (zh) * 2009-06-30 2009-12-09 西北工业大学 网络链路性能测量方法
CN101784058A (zh) * 2009-12-30 2010-07-21 北京市天元网络技术股份有限公司 一种选择网络性能测量点的方法和装置
CN101815002A (zh) * 2009-12-24 2010-08-25 中国人民解放军理工大学 一种基于多测度离差最大化的网络整体性能评价方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7111073B1 (en) * 2000-05-30 2006-09-19 Cisco Technology, Inc. Apparatus for estimating delay and jitter between network routers
CN101854268B (zh) * 2009-04-04 2013-06-05 华为技术有限公司 Ip网络性能测量、服务质量控制的方法、装置和系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1260656A (zh) * 1998-10-09 2000-07-19 朗迅科技公司 测试网络性能的方法和装置以及应激反应分析
CN101056217A (zh) * 2006-04-14 2007-10-17 华为技术有限公司 一种网络性能测量方法及系统
CN101599870A (zh) * 2009-06-30 2009-12-09 西北工业大学 网络链路性能测量方法
CN101815002A (zh) * 2009-12-24 2010-08-25 中国人民解放军理工大学 一种基于多测度离差最大化的网络整体性能评价方法
CN101784058A (zh) * 2009-12-30 2010-07-21 北京市天元网络技术股份有限公司 一种选择网络性能测量点的方法和装置

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11711346B2 (en) 2015-01-06 2023-07-25 Umbra Technologies Ltd. System and method for neutral application programming interface
US11881964B2 (en) * 2015-01-28 2024-01-23 Umbra Technologies Ltd. System and method for a global virtual network
US20220158867A1 (en) * 2015-01-28 2022-05-19 Umbra Technologies Ltd. System and method for a global virtual network
US11799687B2 (en) 2015-04-07 2023-10-24 Umbra Technologies Ltd. System and method for virtual interfaces and advanced smart routing in a global virtual network
US11750419B2 (en) 2015-04-07 2023-09-05 Umbra Technologies Ltd. Systems and methods for providing a global virtual network (GVN)
US11681665B2 (en) 2015-12-11 2023-06-20 Umbra Technologies Ltd. System and method for information slingshot over a network tapestry and granularity of a tick
US11630811B2 (en) 2016-04-26 2023-04-18 Umbra Technologies Ltd. Network Slinghop via tapestry slingshot
US11743332B2 (en) 2016-04-26 2023-08-29 Umbra Technologies Ltd. Systems and methods for routing data to a parallel file system
US11789910B2 (en) 2016-04-26 2023-10-17 Umbra Technologies Ltd. Data beacon pulser(s) powered by information slingshot
US11165677B2 (en) 2018-10-18 2021-11-02 At&T Intellectual Property I, L.P. Packet network performance monitoring
CN111682986A (zh) * 2020-06-08 2020-09-18 平安科技（深圳）有限公司 全量链路质量探测方法、装置、计算机设备及存储介质
CN111682986B (zh) * 2020-06-08 2023-11-07 平安科技（深圳）有限公司 全量链路质量探测方法、装置、计算机设备及存储介质
WO2022199164A1 (fr) * 2021-03-26 2022-09-29 华为技术有限公司 Procédé de mesure de réseau basé sur un échantillonnage par blocs
CN113630171A (zh) * 2021-08-05 2021-11-09 台州学院 一种基于QoS的卫星网络k端可靠性分析方法
CN113709605B (zh) * 2021-08-31 2023-06-09 烽火通信科技股份有限公司 一种光传输网络规划中互斥组业务路由计算方法和装置
CN113709605A (zh) * 2021-08-31 2021-11-26 烽火通信科技股份有限公司 一种光传输网络规划中互斥组业务路由计算方法和装置

Also Published As

Publication number Publication date
CN109314652B (zh) 2021-01-05
CN109314652A (zh) 2019-02-05

Similar Documents

Publication Publication Date Title
WO2018049649A1 (fr) Procédé et dispositif de mesure de performance de réseau
US20200236038A1 (en) Dynamic Deployment of Network Applications Having Performance and Reliability Guarantees in Large Computing Networks
US10374956B1 (en) Managing a hierarchical network
Shafiee et al. A simple congestion-aware algorithm for load balancing in datacenter networks
EP1481510B1 (fr) Procede et systeme d'optimisation des flux de trafic sur la base de contraintes
US7903573B2 (en) Method and system for network traffic matrix analysis
JP4658135B2 (ja) データネットワークにおける往復遅延の遠隔推定
US20140043987A1 (en) Passive network latency monitoring
US6912203B1 (en) Method and apparatus for estimating delay and jitter between many network routers using measurements between a preferred set of routers
US10164864B2 (en) Network tomography through selection of probing paths
CN113938407B (zh) 基于带内网络遥测系统的数据中心网络的故障检测方法及装置
JP2007074711A (ja) 注釈付きネットワーク・トポロジを生成する方法、システム、及びプログラム
CN113408139B (zh) 电力通信网络路径控制仿真验证方法、装置、设备和介质
WO2018102992A1 (fr) Procédé de mesure de performances de réseau et dispositif de détection
JP6084583B2 (ja) フロー経路変更計算装置およびフロー経路変更計算システム
JP2006245874A (ja) オーバーレイネットワークにおける部分測定を利用した通信経路計算システムおよび通信経路計算方法ならびにそのためのプログラム
US7802012B2 (en) Estimating traffic values or intervals for network topology and network behaviour
Abdallah et al. Network convergence in SDN versus OSPF networks
US11368380B1 (en) Estimating end-to-end network packet loss
Syaifuddin et al. Comparison analysis of multipath routing implementation in software defined network
Tri et al. Locating deteriorated links by network-assisted multicast proving on OpenFlow networks
Shin et al. An Efficient Local Repair‐Based Multi‐Constrained Routing for Congestion Control in Wireless Mesh Networks
CN111901237B (zh) 源路由选路方法及系统、相关设备及计算机可读存储介质
Huo et al. A software‐defined networks‐based measurement method of network traffic for 6G technologies
US11388073B1 (en) Estimating end-to-end loss distributions in computer networks

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16916029

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16916029

Country of ref document: EP

Kind code of ref document: A1