WO2021051419A1 - Network node for performance measurement - Google Patents

Network node for performance measurement Download PDF

Info

Publication number
WO2021051419A1
WO2021051419A1 PCT/CN2019/107129 CN2019107129W WO2021051419A1 WO 2021051419 A1 WO2021051419 A1 WO 2021051419A1 CN 2019107129 W CN2019107129 W CN 2019107129W WO 2021051419 A1 WO2021051419 A1 WO 2021051419A1
Authority
WO
WIPO (PCT)
Prior art keywords
network node
performance
packet
performance metric
field
Prior art date
Application number
PCT/CN2019/107129
Other languages
French (fr)
Inventor
Tal Mizrahi
Feili QIAN
Xiaoping Zhu
Dani LIU
Ronen Hyatt
Original Assignee
Huawei Technologies Co., Ltd.
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 Huawei Technologies Co., Ltd. filed Critical Huawei Technologies Co., Ltd.
Priority to PCT/CN2019/107129 priority Critical patent/WO2021051419A1/en
Priority to CN201980096092.9A priority patent/CN113812119B/en
Publication of WO2021051419A1 publication Critical patent/WO2021051419A1/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0852Delays
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0805Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
    • H04L43/0817Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0823Errors, e.g. transmission errors
    • H04L43/0829Packet loss
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/16Threshold monitoring
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08Configuration management of networks or network elements
    • H04L41/0803Configuration setting
    • H04L41/0813Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/0816Configuration setting characterised by the conditions triggering a change of settings the condition being an adaptation, e.g. in response to network events
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/10Active monitoring, e.g. heartbeat, ping or trace-route
    • H04L43/106Active monitoring, e.g. heartbeat, ping or trace-route using time related information in packets, e.g. by adding timestamps

Definitions

  • the present disclosure relates to the field of computer networks and to performance measurement in such networks. Specifically, the present disclosure related to the network node for performance measurement using a packet header with a constant size.
  • Performance measurement and network telemetry are critical capabilities in communication networks. Monitoring the network performance is important for detecting failures, network congestion or anomalies. Moreover, performance measurement and telemetry are key capabilities for autonomous networks, which has been a quickly evolving technology over the last few years.
  • Performance measurement typically requires overhead on the network resources, since measurement and telemetry require at least one of the following: either using dedicated packets for measuring the network, or piggybacking measurement-related information onto data packets (two examples of this approach are in-band network telemetry (INT) and in situ operations, administration and maintenance (IOAM) ) .
  • INT in-band network telemetry
  • IOAM in situ operations, administration and maintenance
  • multi-hop measurement where multiple nodes along a network path are measured, typically an approach is used, where each hop along the network path pushes measurement-related data onto data packets.
  • embodiments of the present invention aim to improve the conventional solution for performance measurement.
  • the measurement overhead may in particular not depend on the number of hops.
  • a first aspect of the present invention provides a network node for performance measurement of a network, wherein the network node is configured to obtain a packet, wherein the packet is being transferred in the network and comprises a performance measurement header with a performance metric field of a fixed length; update the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field; and forward the packet to a next hop of the network node.
  • the network node for performance measurement may also be called transit node, or forwarding node.
  • the network node may however also be called ingress node or egress node.
  • the network node may be a switch, a router, a middle box, a network interface controller, NIC, or any device that can send, receive, and/or forward packets.
  • the performance metric field may comprise a value that represents a performance along a path.
  • the performance metric field may be updated by each node along the path to reflect an updated performance.
  • a value of the performance metric field may relate to a latency of the current packet in the current network node (i.e. the network node for performance measurement) .
  • a value of the performance metric field may relate to a fill level of a queue, in which the current packet is enqueued.
  • the network node is further configured to update the performance metric field based on a present value of the performance metric field and based on a performance metric of the packet measured by the network node, or based on the present value of the performance metric field and based on a performance metric of the network node.
  • the performance metric of the network node can be a current state of a queue of the network node.
  • the network node is further configured to update the performance metric field based on a result of at least one function that is applied to the present value of the performance metric field and/or to the performance metric of the packet measured by the network node.
  • the function may be an identity function, a scaling function, a moving average function, a weighted moving average function, or an exponentially weighted moving average function.
  • the updating, by the network node, the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field includes that the length of the performance metric field after the updating is identical to the length of the performance metric field before the updating.
  • a predefined number of bits may be allocated to the performance metric field of fixed length.
  • updating the performance metric field without changing the length of the performance metric field means that a number of bits allocated by the performance metric field after the updating is identical to a number of bits allocated to the performance metric field before the updating.
  • the performance measurement header further comprises a sequence number field, and/or a timestamp field.
  • the network node is further configured to provide the performance measurement header to an evaluation device.
  • an external evaluation device (which may be called detection device) can be used in combination with the network node.
  • this can be done for a current packet in the network node, before and/or after updating the performance metric field.
  • the network node may be configured to provide the performance measurement header to the evaluation device together with the packet.
  • the network node may be configured to provide the performance measurement header to the evaluation device together with a truncated part of the packet.
  • the network node is further configured to remove the performance measurement header from the packet and forward the packet without the performance measurement header.
  • the packet may be forwarded to a destination indicated in the packet.
  • the network node is further configured to receive the packet without the performance measurement header, and to add the performance measurement header to the received packet to obtain the packet, which comprises the performance measurement header.
  • the network node is further configured to evaluate a performance of the packet based on the performance measurement header.
  • network node can also implement the function of an evaluation device (which can be called detection device) itself.
  • the evaluation of the performance of the network packet may include comparing a value of the performance metric field to a predefined threshold value.
  • the evaluation of the performance of the network packet may include comparing a current value of the performance metric field to at least one previous value of the performance metric field.
  • comparing the current value of the performance metric field to at least one previous value of the performance metric field includes applying a dynamic classification, e.g. based on machine learning techniques.
  • a result of the evaluation is provided to a management device.
  • the performance measurement header further comprises a sequence number field, and/or a timestamp field
  • the network node is further configured to evaluate the performance of the packet based on the sequence number field, and/or the timestamp field.
  • evaluation can also be based on the sequence number field, and/or the timestamp field.
  • the evaluation device may be further configured to detect packet loss, based on the sequence number field.
  • a second aspect of the present invention provides a method for performance measurement of a network, wherein the method includes the steps of obtaining, by a network node, a packet, wherein the packet is being transferred in the network and comprises a performance measurement header with a performance metric field of a fixed length; updating, by the network node, the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field; forwarding, by the network node, the packet to a next hop of the network node.
  • the method further includes updating, by the network node, the performance metric field based on a present value of the performance metric field and based on a performance metric of the packet measured by the network node, or based on the present value of the performance metric field and based on a performance metric of the network node.
  • the method further includes updating, by the network node, the performance metric field based on a result of at least one function that is applied to the present value of the performance metric field and/or to the performance metric of the packet measured by the network node.
  • the updating, by the network node, the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field includes that the length of the performance metric field after the updating is identical to the length of the performance metric field before the updating.
  • the performance measurement header further comprises a sequence number field, and/or a timestamp field.
  • the method further includes providing, by the network node, the performance measurement header to an evaluation device.
  • the method further includes removing, by the network node, the performance measurement header from the packet and forward the packet without the performance measurement header.
  • the method further includes receiving, by the network node, the packet without the performance measurement header, and adding the performance measurement header to the received packet to obtain the packet, which comprises the performance measurement header.
  • the method further includes evaluating, by the network node, a performance of the packet based on the performance measurement header.
  • the performance measurement header further comprises a sequence number field, and/or a timestamp field
  • the method further includes evaluating, by the network node, the performance of the packet based on the sequence number field, and/or the timestamp field.
  • the second aspect and its implementation forms include the same advantages as the first aspect and its respective implementation forms.
  • the embodiments of the present invention provide a solution for in-band measurement that allows multi-hop measurement, but uses performance data with a constant length.
  • in-band measurement refers to a measurement method that piggybacks performance-related data onto data packets.
  • the fixed-length data is carried with the data packet along its path, and may be updated by intermediate network devices in order to reflect their performance.
  • a detection module can be used at the end of the path to detect performance degradation.
  • FIG. 1 shows a schematic view of a network node for performance measurement according to an embodiment of the present invention.
  • FIG. 2 shows a schematic view of a network node for performance measurement according to an embodiment of the present invention in more detail.
  • FIG. 3 shows an operating example according to the present invention.
  • FIG. 4 shows a schematic view of a packet according to the present invention.
  • FIG. 5 shows a schematic view of packets according to the present invention.
  • FIG. 6 shows a schematic view of a method according to an embodiment of the present invention.
  • FIG. 1 shows a schematic view of a network node 100 according to an embodiment of the present invention.
  • the network node 100 is for performance measurement in a network.
  • the network node 100 obtains a packet 101.
  • the packet 101 comprises a performance measurement header 102 and a performance metric field 103 of a fixed length. That is, the length of the performance metric field 103 does not change during the whole processing which is applied to it for performance measurement.
  • the packet 101 can be obtained from another network node prior in the path of the packet 101.
  • the other network node e.g. has added the performance measurement header 102 and the performance metric field 103 of a fixed length to the packet 101. Therefore, this network node may also be called ingress node.
  • the network node 100 then updates the performance metric field 103 based on a performance measurement 104 performed in the network node 100 without changing the length of the performance metric field 103. That means, the length of the performance metric field 103 is the same before and after the updating. In particular, the bit size of the metric field 103 does not change due to the updating, i.e. is the same before the updating and after the updating. Thereby a constant header size is obtained and increasing overhead is avoided.
  • the network node 100 forwards the packet 101 to another network node, e.g. along the intended path of the packet 101.
  • this node may also be called egress node.
  • the network node 100 may also be called transit node.
  • the packet 101 generally can be any kind of network packet.
  • the packet101 can be a data packet, i.e. a packet that is sent from a source host to a destination host.
  • the packet101 can be a control packet, which is sent between network devices.
  • FIG. 2 shows a schematic view of a network node 100 according to an embodiment of the present invention in more detail.
  • the network node 100 of FIG. 2 builds on the network node 100 of FIG. 1 and therefore includes all of its features and functions.
  • the network node 100 optionally may update the performance metric field 103 based on a present value of the performance metric field 103 and based on a performance metric 201 of the packet 101 measured by the network node 100. That is, the network node 100 can read the present value of the performance metric field 103, can obtain a performance metric 201 (e.g. a latency value) of the packet 101 and can then calculates a new value of the performance metric field 103 which replaces the old one.
  • a length of the new value is identical to the length of the old one, i.e. the new value may include the same number of bits as the old value.
  • the network node 100 may update the performance metric field 103 based on the present value of the performance metric field 103 and based on a performance metric 202 of the network node 100. That is, the network node 100 can read the present value of the performance metric field 103, can obtain a performance metric 202 (e.g. a latency value, such as a fill level of a packet queue) of the network node 100 and can then calculates a new value of the performance metric field 103 which replaces the old one.
  • a length of the new value is identical to the length of the old one, i.e. the new value may include the same number of bits as the old value.
  • the network node 100 may update the performance metric field 103 based on a result of at least one function that is applied to the present value of the performance metric field 103 and to the performance metric 201 or to the performance metric 202.
  • the function in particular ensures that the length of the performance metric field 103 remains unchanged.
  • the network node 100 in particular may implement an update algorithm, according to which a node along the path (that is, a transit node, and optionally also an ingress node or an egress node) can update the performance metric field 103.
  • An example of a possible performance metric 201 is latency of the current packet in the current device.
  • An example of a possible performance metric 202 is a fill level of the queue that the current packet is queued in.
  • the performance metric 201 can be the accumulated delay along the path, i.e. the sum of the delays of the network nodes 100 along the path.
  • f (x) or g (x) can be anyone of a an identity function, a scaling function, a moving average function, a weighted moving average function, or an exponentially weighted moving average function.
  • functions f (x) and g (x) are:
  • a performance metric is simply a sum of the metric values along the hops of the packet 101.
  • the performance measurement header 102 optionally may comprises a sequence number field 203, and/or a timestamp field 204.
  • the sequence number field 203, and/or a timestamp field 204 can assist in performance measurement and evaluation, as it is going to be described below.
  • the network node 100 can provide the performance measurement header 102 to an evaluation device. That is, the network node can directly forward information which is required for performance measurement to an evaluation device, without the need for a further device to do so.
  • the network node 100 also allows for per-hop exporting. That is, each network node 100 along a path, upon receiving a packet 101 with a performance measurement header 102 can export a set of detailed performance related fields to a detection module, and/or canattach a sequence number 203 or timestamp 204 to the performance measurement header 102. The detection module then can use the detailed performance information for further performance analysis. The sequence number 203 or the timestamp 204 can be used for correlating information from the same packet 101 received at different network nodes 100.
  • the network node 100 can remove the performance measurement header 102 from the packet 101 and forward the packet 101 without the performance measurement header 102. That is, the packet 101 can be forwarded to its intended destination without the performance measurement header 102.
  • the network node 100 can also implement the function of an egress node, which removes performance data from the packet and forwards the packet towards its destination. That is, an egress node optionally also can update the performance metric field and export the performance data to a detection module.
  • the network node 100 can receive a packet 101 without the performance measurement header 102, and add the performance measurement header 102 to the received packet 101 to obtain the packet 101, which comprises the performance measurement header 102.
  • the network node 100 can also implement the function of an ingress node, which prepares a packet 101 for performance measurement, by adding a performance measurement header 102 to a received packet 101 which does not yet comprise a performance measurement header 102.
  • an ingress node can push the following performance data into all or a subset of the packets 101 that are forwarded through it: a performance measurement header 102, a performance metric field 103 (which is going to be updated on a hop-by-hop basis and is a value that represents the performance along the path, and may be updated by each node along the path to reflect the updated performance) , a sequence number field (which is in particular assigned by the ingress node and unchanged by other following nodes) , a timestamp field.
  • the network node 100 can evaluate a performance of the packet 101 based on the performance measurement header 102. That is, the network node 100 can also implement the function of an evaluation module (which may also be called detection module) without the need for a further device. In other words, a detection module may be used for detecting performance issues or performance degradation.
  • the module may be an external node, or may be an internal module in the network node 100.
  • the detection module may be used for detecting packet loss by comparing the sequence number of the current packet to a previous sequence number. If no packets were lost the difference in the sequence numbers should be one (i.e. the difference is the number of packets that were lost plus one) .
  • the detection module may detect performance degradation by processing the performance metric field 103 using a detection function in order to detect performance degradation.
  • the detection function may be one of the following: comparing a performance metric field 103 to a configurable threshold that indicates a problem; or comparing a current value to previous values and apply a dynamic classification (for example using machine learning) .
  • the detection module may report to a user or to a management system, or may trigger fine-grained measurement to detect the location of the problem, e.g., IOAM, or may take a corrective action, such as a reconfiguration of a network path.
  • IOAM fine-grained measurement to detect the location of the problem
  • the network node 100 optionally can also evaluate the performance of the packet 101 based on the sequence number field 203, and/or the timestamp field 204.
  • the sequence number field 203, and/or the timestamp field 204 do not necessarily be comprised by the performance measurement header 102. They however can be separate parts of the performance data along with the header (i.e. separate fields in the performance data) .
  • FIG. 3 shows an example scenario of operating several instances of a network node 100.
  • the network node 100 A implements a transit node
  • the network node 100 B implements an ingress node
  • the network node 100 C implements an egress node
  • the network node 100 D implements a detection module.
  • performance measurement is performed between an ingress node 100 B and an egress node 100 C.
  • the ingress node 100 B is the first node along the path that takes part in the measurement
  • the egress node 100 C is the last node along the path that takes part in the measurement.
  • the ingress node 100 C may push measurement data into packets 101.
  • the egress node 100 C may remove the measurement data pushed by the ingress node 100 B and forwards data packets towards a destination.
  • Transit nodes 100 A are nodes along the path that may push additional data or update the existing data (although only one transit node 100 A is shown, there can be multiple transit nodes 100 A along the path) .
  • the detection module 100 D is a module that is used for detecting performance degradation.
  • the network devices of FIG. 3 may be switches, routers, middle boxes, NICs, or any device that forwards packets.
  • the egress node 100 C can provide information to the detection module 100 D
  • any other node along the path e.g. the ingress node 100 B or the transit node 100 A
  • FIG. 4 shows a packet 101 with performance data, wherein the performance data comprises a performance measurement header 102 and a performance metric field 103.
  • the performance data shown in FIG. 4 also includes the optional sequence number field 203, and the optional timestamp field 204.
  • FIG. 5 shows packets which include performance data (which inter alia includes the performance measurement header 102) pushed by an ingress node.
  • This data may be pushed with a tunnel header (such as VXLAN-GPE) , as shown in the second line in the figure, or can be pushed as an extension to an existing header in the packet, such as an IPv6 extension header, as shown in the third line.
  • the first line shows a regular packet without the performance data.
  • FIG. 6 shows a method 600 according to an embodiment of the present invention.
  • the method 600 is for performance measurement and comprises a step of obtaining 601, by a network node 100, a packet 101, wherein the packet 101 comprises a performance measurement header 102 with a performance metric field 103 of a fixed length.
  • the method further comprises a step of updating 602, by the network node 100, the performance metric field 103 based on a performance measurement 104 performed in the network node 100 without changing the length of the performance metric field 103.
  • the method further comprises a step of forwarding 603, by the network node 100, the packet 101 to a network node.
  • present invention provides multi-hop in-band measurement data while keeping a constant overhead on packets, independent of a number of hops in a network.
  • present invention allows for using a generic per-hop update function so that the multi-hop information is represented by a fixed-length field.
  • present invention uses an innovative combination of two components while keeping a constant overhead:
  • the present invention further provides a detection module which can be used as a trigger for fine-grained measurement using higher overhead.
  • the measurement data according to the present invention can e.g. be extended to include more types of information, while keeping a fixed length.
  • Another possible implementation of the present invention is to only use the multi-hop measurement data, without using the single-hop measurement data (sequence number or timestamp) .

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

The present disclosure relates to the field of computer networks and to performance measurement in such networks. The performance measurement is performed by a network node using packet headers with a constant size. The present invention thus provides a network node for performance measurement, wherein the network node is configured to obtain a packet, wherein the packet comprises a performance measurement header with a performance metric field of a fixed length; update the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field; and forward the packet to a network node.

Description

NETWORK NODE FOR PERFORMANCE MEASUREMENT TECHNICAL FIELD
The present disclosure relates to the field of computer networks and to performance measurement in such networks. Specifically, the present disclosure related to the network node for performance measurement using a packet header with a constant size.
BACKGROUND
Performance measurement and network telemetry are critical capabilities in communication networks. Monitoring the network performance is important for detecting failures, network congestion or anomalies. Moreover, performance measurement and telemetry are key capabilities for autonomous networks, which has been a quickly evolving technology over the last few years.
Performance measurement typically requires overhead on the network resources, since measurement and telemetry require at least one of the following: either using dedicated packets for measuring the network, or piggybacking measurement-related information onto data packets (two examples of this approach are in-band network telemetry (INT) and in situ operations, administration and maintenance (IOAM) ) . In multi-hop measurement, where multiple nodes along a network path are measured, typically an approach is used, where each hop along the network path pushes measurement-related data onto data packets.
It is a problem of this approach that the amount of overhead on each data packet increases with the number of hops along the network path, yielding significant overhead along a path with a large number of hops.
SUMMARY
In view of the above-mentioned problem, embodiments of the present invention aim to improve the conventional solution for performance measurement.
It is an objective to provide a measurement solution that performs multi-hop measurement with a low measurement overhead. The measurement overhead may in particular not depend on the number of hops.
The objective is achieved by the embodiments of the invention as described in the enclosed independent claims. Advantageous implementations of the embodiments of the invention are further defined in the dependent claims.
A first aspect of the present invention provides a network node for performance measurement of a network, wherein the network node is configured to obtain a packet, wherein the packet is being transferred in the network and comprises a performance measurement header with a performance metric field of a fixed length; update the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field; and forward the packet to a next hop of the network node.
This ensures that the overhead, which is allocated for performance measurement, remains constant.
In particular, the network node for performance measurement may also be called transit node, or forwarding node. The network node may however also be called ingress node or egress node.
In particular, the network node may be a switch, a router, a middle box, a network interface controller, NIC, or any device that can send, receive, and/or forward packets.
In particular, the performance metric field may comprise a value that represents a performance along a path. In particular, the performance metric field may be updated by each node along the path to reflect an updated performance.
In particular, a value of the performance metric field may relate to a latency of the current packet in the current network node (i.e. the network node for performance measurement) . In particular, a value of the performance metric field may relate to a fill level of a queue, in which the current packet is enqueued.
In an implementation form of the first aspect, the network node is further configured to update the performance metric field based on a present value of the performance metric field and based on a performance metric of the packet measured by the network node, or based on the present value of the performance metric field and based on a performance metric of the network node.
This is beneficial as either a state of the packet in the network node can be considered for performance measurement, as well as a state of the network node itself.
In particular, the performance metric of the network node can be a current state of a queue of the network node.
In a further implementation form of the first aspect, the network node is further configured to update the performance metric field based on a result of at least one function that is applied to  the present value of the performance metric field and/or to the performance metric of the packet measured by the network node.
This ensures that functions can be used which provide output values of a same size as the corresponding input values.
In particular, the function may be an identity function, a scaling function, a moving average function, a weighted moving average function, or an exponentially weighted moving average function.
In a further implementation form of the first aspect, the updating, by the network node, the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field includes that the length of the performance metric field after the updating is identical to the length of the performance metric field before the updating.
In particular, a predefined number of bits may be allocated to the performance metric field of fixed length. In particular, updating the performance metric field without changing the length of the performance metric field means that a number of bits allocated by the performance metric field after the updating is identical to a number of bits allocated to the performance metric field before the updating.
In a further implementation form of the first aspect, the performance measurement header further comprises a sequence number field, and/or a timestamp field.
This ensures that further information can be used for performance measurement.
In a further implementation form of the first aspect, the network node is further configured to provide the performance measurement header to an evaluation device.
This ensures, that an external evaluation device (which may be called detection device) can be used in combination with the network node.
In particular, this can be done for a current packet in the network node, before and/or after updating the performance metric field.
In particular, the network node may be configured to provide the performance measurement header to the evaluation device together with the packet.
In particular, the network node may be configured to provide the performance measurement header to the evaluation device together with a truncated part of the packet.
In a further implementation form of the first aspect, the network node is further configured to remove the performance measurement header from the packet and forward the packet without the performance measurement header.
This ensures that the network node can be used to implement an egress node which terminates performance measurement.
In particular, the packet may be forwarded to a destination indicated in the packet.
In a further implementation form of the first aspect, the network node is further configured to receive the packet without the performance measurement header, and to add the performance measurement header to the received packet to obtain the packet, which comprises the performance measurement header.
This ensures that the network node can be used to implement an ingress node which starts performance measurement.
In a further implementation form of the first aspect, the network node is further configured to evaluate a performance of the packet based on the performance measurement header.
This ensures that the network node can also implement the function of an evaluation device (which can be called detection device) itself.
In particular, the evaluation of the performance of the network packet may include comparing a value of the performance metric field to a predefined threshold value.
In particular, the evaluation of the performance of the network packet may include comparing a current value of the performance metric field to at least one previous value of the performance metric field.
In particular, comparing the current value of the performance metric field to at least one previous value of the performance metric field includes applying a dynamic classification, e.g. based on machine learning techniques.
In particular, after evaluating the performance of a network packet, a result of the evaluation is provided to a management device.
In a further implementation form of the first aspect, the performance measurement header further comprises a sequence number field, and/or a timestamp field, and the network node is further configured to evaluate the performance of the packet based on the sequence number field, and/or the timestamp field.
This ensures that evaluation can also be based on the sequence number field, and/or the timestamp field.
In particular, the evaluation device may be further configured to detect packet loss, based on the sequence number field.
A second aspect of the present invention provides a method for performance measurement of a network, wherein the method includes the steps of obtaining, by a network node, a packet, wherein the packet is being transferred in the network and comprises a performance  measurement header with a performance metric field of a fixed length; updating, by the network node, the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field; forwarding, by the network node, the packet to a next hop of the network node.
In an implementation form of the second aspect, the method further includes updating, by the network node, the performance metric field based on a present value of the performance metric field and based on a performance metric of the packet measured by the network node, or based on the present value of the performance metric field and based on a performance metric of the network node.
In a further implementation form of the second aspect, the method further includes updating, by the network node, the performance metric field based on a result of at least one function that is applied to the present value of the performance metric field and/or to the performance metric of the packet measured by the network node.
In a further implementation form of the second aspect, the updating, by the network node, the performance metric field based on a performance measurement performed in the network node without changing the length of the performance metric field includes that the length of the performance metric field after the updating is identical to the length of the performance metric field before the updating.
In a further implementation form of the second aspect, the performance measurement header further comprises a sequence number field, and/or a timestamp field.
In a further implementation form of the second aspect, the method further includes providing, by the network node, the performance measurement header to an evaluation device.
In a further implementation form of the second aspect, the method further includes removing, by the network node, the performance measurement header from the packet and forward the packet without the performance measurement header.
In a further implementation form of the second aspect, the method further includes receiving, by the network node, the packet without the performance measurement header, and adding the performance measurement header to the received packet to obtain the packet, which comprises the performance measurement header.
In a further implementation form of the second aspect, the method further includes evaluating, by the network node, a performance of the packet based on the performance measurement header.
In a further implementation form of the second aspect, the performance measurement header further comprises a sequence number field, and/or a timestamp field, and the method further includes evaluating, by the network node, the performance of the packet based on the sequence number field, and/or the timestamp field.
The second aspect and its implementation forms include the same advantages as the first aspect and its respective implementation forms.
In summary, the embodiments of the present invention provide a solution for in-band measurement that allows multi-hop measurement, but uses performance data with a constant length. In this context “in-band measurement” refers to a measurement method that piggybacks performance-related data onto data packets. The fixed-length data is carried with the data packet along its path, and may be updated by intermediate network devices in order to reflect their performance. A detection module can be used at the end of the path to detect performance degradation.
It has to be noted that all devices, elements, units and means described in the present application could be implemented in the software or hardware elements or any kind of combination thereof. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements, or any kind of combination thereof.
BRIEF DESCRIPTION OF DRAWINGS
The above-described aspects and implementation forms of the present invention will be explained in the following description of specific embodiments in relation to the enclosed drawings, in which
FIG. 1 shows a schematic view of a network node for performance measurement according to an embodiment of the present invention.
FIG. 2 shows a schematic view of a network node for performance measurement according to an embodiment of the present invention in more detail.
FIG. 3 shows an operating example according to the present invention.
FIG. 4 shows a schematic view of a packet according to the present invention.
FIG. 5 shows a schematic view of packets according to the present invention.
FIG. 6 shows a schematic view of a method according to an embodiment of the present invention.
DETAILED DESCRIPTION OF EMBODIMENTS
FIG. 1 shows a schematic view of a network node 100 according to an embodiment of the present invention. The network node 100 is for performance measurement in a network.
For performance measurement, the network node 100 obtains a packet 101. The packet 101 comprises a performance measurement header 102 and a performance metric field 103 of a fixed length. That is, the length of the performance metric field 103 does not change during the whole processing which is applied to it for performance measurement. The packet 101 can be obtained from another network node prior in the path of the packet 101. The other network node e.g. has added the performance measurement header 102 and the performance metric field 103 of a fixed length to the packet 101. Therefore, this network node may also be called ingress node.
The network node 100 then updates the performance metric field 103 based on a performance measurement 104 performed in the network node 100 without changing the length of the performance metric field 103. That means, the length of the performance metric field 103 is the same before and after the updating. In particular, the bit size of the metric field 103 does not change due to the updating, i.e. is the same before the updating and after the updating. Thereby a constant header size is obtained and increasing overhead is avoided.
Then, the network node 100 forwards the packet 101 to another network node, e.g. along the intended path of the packet 101. In this other network node, further processing can be applied to the packet 101 (e.g. the performance measurement can be terminated at this node) . Therefore, this node may also be called egress node. As the network 100 forwards the packet  101 from an ingress node to an egress node, the network node 100 may also be called transit node.
The packet 101 generally can be any kind of network packet. In particular the packet101 can be a data packet, i.e. a packet that is sent from a source host to a destination host. In particular, the packet101 can be a control packet, which is sent between network devices.
FIG. 2 shows a schematic view of a network node 100 according to an embodiment of the present invention in more detail. The network node 100 of FIG. 2 builds on the network node 100 of FIG. 1 and therefore includes all of its features and functions.
As it is illustrated in FIG. 2, the network node 100 optionally may update the performance metric field 103 based on a present value of the performance metric field 103 and based on a performance metric 201 of the packet 101 measured by the network node 100. That is, the network node 100 can read the present value of the performance metric field 103, can obtain a performance metric 201 (e.g. a latency value) of the packet 101 and can then calculates a new value of the performance metric field 103 which replaces the old one. A length of the new value is identical to the length of the old one, i.e. the new value may include the same number of bits as the old value.
Additionally or alternatively, the network node 100 may update the performance metric field 103 based on the present value of the performance metric field 103 and based on a performance metric 202 of the network node 100. That is, the network node 100 can read the present value of the performance metric field 103, can obtain a performance metric 202 (e.g. a latency value, such as a fill level of a packet queue) of the network node 100 and can then calculates a new value of the performance metric field 103 which replaces the old one. A length of the new value is identical to the length of the old one, i.e. the new value may include the same number of bits as the old value.
Further optionally, the network node 100 may update the performance metric field 103 based on a result of at least one function that is applied to the present value of the performance metric field 103 and to the performance metric 201 or to the performance metric 202. The function in particular ensures that the length of the performance metric field 103 remains unchanged.
The network node 100 in particular may implement an update algorithm, according to which a node along the path (that is, a transit node, and optionally also an ingress node or an egress node) can update the performance metric field 103. The performance metric field 103 can be updated from OldValue to NewValue according to the following algorithm: NewValue =f (OldValue) + g (m) , where m is the value of the performance metric 201 or the performance metric 202, and where f () and g () are two functions that are used in the processing.
An example of a possible performance metric 201 is latency of the current packet in the current device. An example of a possible performance metric 202 is a fill level of the queue that the current packet is queued in.
For example, the performance metric 201 can be the accumulated delay along the path, i.e. the sum of the delays of the network nodes 100 along the path. In this case the update algorithm in each hop is: NewValue = OldValue + CurrentHopLatency, where: f (x) =x and g (x) =x.
f (x) or g (x) however also can be anyone of a an identity function, a scaling function, a moving average function, a weighted moving average function, or an exponentially weighted moving average function. Further examples of functions f (x) and g (x) are:
Identity function:
f (x) =x, g (x) =x
In this case, a performance metric is simply a sum of the metric values along the hops of the packet 101.
Shift right by 4 bits:
f (x) =x, g (x) = x>>4
This allows to scale a performance metric value to 2^4 = 16 hops.
Exponentially weighted moving average:
f (x) = x –x>>5, g (x) =x>>5
NewValue = (1-2^-5) *OldValue + 2^-5*NewValue
As it is also illustrated in FIG. 2, the performance measurement header 102 optionally may comprises a sequence number field 203, and/or a timestamp field 204. The sequence number field 203, and/or a timestamp field 204 can assist in performance measurement and evaluation, as it is going to be described below.
Further optionally, the network node 100 can provide the performance measurement header 102 to an evaluation device. That is, the network node can directly forward information which is required for performance measurement to an evaluation device, without the need for a further device to do so.
In other words, the network node 100 also allows for per-hop exporting. That is, each network node 100 along a path, upon receiving a packet 101 with a performance measurement header 102 can export a set of detailed performance related fields to a detection module, and/or canattach a sequence number 203 or timestamp 204 to the performance measurement header 102. The detection module then can use the detailed performance information for further performance analysis. The sequence number 203 or the timestamp 204 can be used for correlating information from the same packet 101 received at different network nodes 100.
Further optionally, the network node 100 can remove the performance measurement header 102 from the packet 101 and forward the packet 101 without the performance measurement header 102. That is, the packet 101 can be forwarded to its intended destination without the performance measurement header 102.
In other words, the network node 100 can also implement the function of an egress node, which removes performance data from the packet and forwards the packet towards its destination. That is, an egress node optionally also can update the performance metric field and export the performance data to a detection module.
Further optionally, the network node 100 can receive a packet 101 without the performance measurement header 102, and add the performance measurement header 102 to the received packet 101 to obtain the packet 101, which comprises the performance measurement header 102.
In other words, the network node 100 can also implement the function of an ingress node, which prepares a packet 101 for performance measurement, by adding a performance measurement header 102 to a received packet 101 which does not yet comprise a performance measurement header 102. In other words, an ingress node, can push the following performance data into all or a subset of the packets 101 that are forwarded through it: a performance measurement header 102, a performance metric field 103 (which is going to be updated on a hop-by-hop basis and is a value that represents the performance along the path, and may be updated by each node along the path to reflect the updated performance) , a sequence number field (which is in particular assigned by the ingress node and unchanged by other following nodes) , a timestamp field.
Further optionally, the network node 100 can evaluate a performance of the packet 101 based on the performance measurement header 102. That is, the network node 100 can also implement the function of an evaluation module (which may also be called detection module) without the need for a further device. In other words, a detection module may be used for detecting performance issues or performance degradation. The module may be an external node, or may be an internal module in the network node 100.
The detection module may be used for detecting packet loss by comparing the sequence number of the current packet to a previous sequence number. If no packets were lost the difference in the sequence numbers should be one (i.e. the difference is the number of packets that were lost plus one) . The detection module may detect performance degradation by processing the performance metric field 103 using a detection function in order to detect performance degradation. The detection function may be one of the following: comparing a performance metric field 103 to a configurable threshold that indicates a problem; or comparing a current value to previous values and apply a dynamic classification (for example using machine learning) .
Upon detecting packet loss or performance degradation (or both) , the detection module may report to a user or to a management system, or may trigger fine-grained measurement to detect the location of the problem, e.g., IOAM, or may take a corrective action, such as a reconfiguration of a network path.
Moreover, as the performance measurement header 102 can comprise a sequence number field 203, and/or a timestamp field 204, the network node 100 optionally can also evaluate the performance of the packet 101 based on the sequence number field 203, and/or the timestamp field 204. The sequence number field 203, and/or the timestamp field 204 do not necessarily be comprised by the performance measurement header 102. They however can be separate parts of the performance data along with the header (i.e. separate fields in the performance data) .
FIG. 3 shows an example scenario of operating several instances of a network node 100. In FIG. 3, the network node 100 A implements a transit node, the network node 100 B implements an ingress node, the network node 100 C implements an egress node and the network node 100 D implements a detection module.
As shown in FIG. 3, performance measurement is performed between an ingress node 100 B and an egress node 100 C. The ingress node 100 B is the first node along the path that takes part in the measurement, and the egress node 100 C is the last node along the path that takes part in the measurement. The ingress node 100 C may push measurement data into packets 101. The egress node 100 C may remove the measurement data pushed by the ingress node 100 B and forwards data packets towards a destination. Transit nodes 100 A are nodes along the path that may push additional data or update the existing data (although only one transit node 100 A is shown, there can be multiple transit nodes 100 A along the path) . The detection module 100 D is a module that is used for detecting performance degradation. It may run on a remote server, or may run locally, e.g. as a module inside one of the nodes 100 A, 100 B, 100C. The network devices of FIG. 3 may be switches, routers, middle boxes, NICs, or any device that forwards packets. Although it is only shown that the egress node 100 C can provide information to the detection module 100 D, any other node along the path (e.g. the ingress node 100 B or the transit node 100 A) can also provide information to the detection module 100 D.
FIG. 4 shows a packet 101 with performance data, wherein the performance data comprises a performance measurement header 102 and a performance metric field 103. The performance data shown in FIG. 4 also includes the optional sequence number field 203, and the optional timestamp field 204.
FIG. 5 shows packets which include performance data (which inter alia includes the performance measurement header 102) pushed by an ingress node. This data may be pushed with a tunnel header (such as VXLAN-GPE) , as shown in the second line in the figure, or can be pushed as an extension to an existing header in the packet, such as an IPv6 extension header, as shown in the third line. The first line shows a regular packet without the performance data.
FIG. 6 shows a method 600 according to an embodiment of the present invention. The method 600 is for performance measurement and comprises a step of obtaining 601, by a network node 100, a packet 101, wherein the packet 101 comprises a performance measurement header 102 with a performance metric field 103 of a fixed length. The method further comprises a step of updating 602, by the network node 100, the performance metric field 103 based on a performance measurement 104 performed in the network node 100 without changing the length of the performance metric field 103. The method further comprises a step of forwarding 603, by the network node 100, the packet 101 to a network node.
The main difference between the present invention and prior art solutions is that that present invention provides multi-hop in-band measurement data while keeping a constant overhead on packets, independent of a number of hops in a network. The present invention allows for using a generic per-hop update function so that the multi-hop information is represented by a fixed-length field. Moreover, the present invention uses an innovative combination of two components while keeping a constant overhead:
single-hop measurement data (sequence number or timestamp) and multi-hop measurement data (the performance metric) . The present invention further provides a detection module which can be used as a trigger for fine-grained measurement using higher overhead. The measurement data according to the present invention can e.g. be extended to include more types of information, while keeping a fixed length. Another possible implementation of the present invention is to only use the multi-hop measurement data, without using the single-hop measurement data (sequence number or timestamp) .
The present invention has been described in conjunction with various embodiments as examples as well as implementations. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed invention, from the studies of the drawings, this disclosure and the independent claims. In the claims as well as in the description the word “comprising” does not exclude other elements or steps and the  indefinite article “a” or “an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutual different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation.

Claims (13)

  1. A network node (100) for performance measurement, wherein the network node (100) is configured to:
    - obtain a packet (101) , wherein the packet (101) comprises a performance measurement header (102) with a performance metric field (103) of a fixed length;
    - update the performance metric field (103) based on a performance measurement (104) performed in the network node (100) without changing the length of the performance metric field (103) ; and
    - forward the packet (101) to a next hop of the network node (100) .
  2. The network node (100) according to claim 1, further configured to update the performance metric field (103) based on a present value of the performance metric field (103) and one of the following information:
    a performance metric (201) of the packet (101) measured by the network node (100) , or
    a performance metric (202) of the network node (100) .
  3. The network node (100) according to claim 2, further configured to update the performance metric field (103) based on a result of at least one function that is applied to the present value of the performance metric field (103) and/or to the performance metric (201) of the packet (101) measured by the network node (100) .
  4. The network node (100) according to any of the preceding claims, wherein the updating, by the network node (100) , the performance metric field (103) based on a performance measurement (104) performed in the network node (100) without changing the length of the performance metric field (103) includes that the length of the performance metric field (103) after the updating is identical to the length of the performance metric field (103) before the updating.
  5. The network node (100) according to any of the preceding claims, wherein the performance measurement header (102) further comprises a sequence number field (203) , and/or a timestamp field (204) .
  6. The network node (100) according to any of the preceding claims, further configured to provide the performance measurement header (102) to an evaluation device.
  7. The network node (100) according to any of the preceding claims, further configured to remove the performance measurement header (102) from the packet (101) and forward the packet (101) without the performance measurement header (102) .
  8. The network node (100) according to any of the preceding claims, further configured to:
    - receive the packet (101) without the performance measurement header (102) , and
    - add the performance measurement header (102) to the received packet (101) to obtain the packet (101) , which comprises the performance measurement header (102) .
  9. The network node (100) according to any one of the preceding claims, configured to evaluate a performance of the packet (101) based on the performance measurement header (102) .
  10. The network node (100) according to claim 9, wherein the performance measurement header (102) further comprises a sequence number field (203) , and/or a timestamp field (204) , and wherein the network node (100) is further configured to evaluate the performance of the packet (101) based on the sequence number field (203) , and/or the timestamp field (204) .
  11. A method (600) for performance measurement of a network, wherein the method (600) includes the steps of:
    - obtaining (601) , by a network node (100) , a packet (101) , wherein the packet (101) is being transferred in the network and comprises a performance measurement header (102) with a performance metric field (103) of a fixed length;
    - updating (602) , by the network node (100) , the performance metric field (103) based on a performance measurement (104) performed in the network node (100) without changing the length of the performance metric field (103) ;
    - forwarding (603) , by the network node (100) , the packet (101) to a next hop of the network node.
  12. A computer program product with a program code for performing a method according to claim 11 when the computer program runs on a computer.
  13. A computer readable storage medium comprising computer program code instructions, being executable by a computer, for performing a method according to claim 11 when the computer program code instructions runs on a computer.
PCT/CN2019/107129 2019-09-21 2019-09-21 Network node for performance measurement WO2021051419A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2019/107129 WO2021051419A1 (en) 2019-09-21 2019-09-21 Network node for performance measurement
CN201980096092.9A CN113812119B (en) 2019-09-21 2019-09-21 Network node for performance measurement

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/107129 WO2021051419A1 (en) 2019-09-21 2019-09-21 Network node for performance measurement

Publications (1)

Publication Number Publication Date
WO2021051419A1 true WO2021051419A1 (en) 2021-03-25

Family

ID=74883104

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/107129 WO2021051419A1 (en) 2019-09-21 2019-09-21 Network node for performance measurement

Country Status (2)

Country Link
CN (1) CN113812119B (en)
WO (1) WO2021051419A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116781499A (en) * 2022-03-11 2023-09-19 华为技术有限公司 Network performance measurement method, system and device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050185632A1 (en) * 2004-02-23 2005-08-25 Microsoft Corporation System and method for link quality source routing
WO2019030552A1 (en) * 2017-08-08 2019-02-14 Telefonaktiebolaget Lm Ericsson (Publ) Scalable network path tracing

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080159287A1 (en) * 2006-12-29 2008-07-03 Lucent Technologies Inc. EFFICIENT PERFORMANCE MONITORING USING IPv6 CAPABILITIES
CN106789730B (en) * 2016-12-29 2020-02-11 杭州迪普科技股份有限公司 Method and device for processing fragment message
CN109981457B (en) * 2017-12-27 2021-09-07 华为技术有限公司 Message processing method, network node and system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050185632A1 (en) * 2004-02-23 2005-08-25 Microsoft Corporation System and method for link quality source routing
WO2019030552A1 (en) * 2017-08-08 2019-02-14 Telefonaktiebolaget Lm Ericsson (Publ) Scalable network path tracing

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
G. MIRSKY ZTE CORP. W. MENG ZTE CORPORATION B. KHASNABISH C. WANG INDIVIDUAL CONTRIBUTOR: "Active OAM for Service Function Chains in Networks; draft-ietf-sfc-multi-layer-oam-03.txt", ACTIVE OAM FOR SERVICE FUNCTION CHAINS IN NETWORKS; DRAFT-IETF-SFC-MULTI-LAYER-OAM-03.TXT; INTERNET-DRAFT: SFC WG, INTERNET ENGINEERING TASK FORCE, IETF; STANDARDWORKINGDRAFT, INTERNET SOCIETY (ISOC) 4, RUE DES FALAISES CH- 1205 GENEVA, SWITZERLAND, no. 03, 9 May 2019 (2019-05-09), Internet Society (ISOC) 4, rue des Falaises CH- 1205 Geneva, Switzerland, pages 1 - 18, XP015132880 *

Also Published As

Publication number Publication date
CN113812119A (en) 2021-12-17
CN113812119B (en) 2023-04-18

Similar Documents

Publication Publication Date Title
US10819605B2 (en) Method, apparatus, and system for implementing delay measurement
US9106548B2 (en) Network fault localization
US7902973B2 (en) Alarm reordering to handle alarm storms in large networks
JP5646090B2 (en) Traceroute_delay diagnostic command
EP3735762B1 (en) In-band telemetry with limited extra bytes
US20050232227A1 (en) Method and apparatus for characterizing an end-to-end path of a packet-based network
US20070064611A1 (en) Method for monitoring packet loss ratio
CN108370333B (en) System, method and node for performance measurement in a segmented routing network
US9225616B2 (en) Feedback-based tuning of control plane traffic by proactive user traffic observation
US20140258524A1 (en) Detection of Load Balancing Across Network Paths in a Communication Network
EP3369213B1 (en) Performance measurement in a packet-switched communication network
US20100054140A1 (en) Fault detection in a transport network
EP3188413B1 (en) Method, apparatus, and system for implementing packet loss detection
US20140369218A1 (en) Measurement on a data flow in a communication network
CN112689977B (en) Techniques for collecting and transmitting telemetry information in a communication network
JP6616230B2 (en) Network equipment
US11165671B2 (en) Performance measurement in a packet-switched communication network
JP6740371B2 (en) Performance measurement for multi-point packet flow
WO2021051419A1 (en) Network node for performance measurement
EP2957069B1 (en) Method and network entity for evaluating a link between a first network node and a second network node
US10805206B1 (en) Method for rerouting traffic in software defined networking network and switch thereof
US10855593B1 (en) Methods and apparatus for load balancing in a network switching system
JP2008219722A (en) Node, communication system and program for node
Nguyen et al. Evaluation of an end-to-end delay estimation in the case of multiple flows in SDN networks
WO2021052554A1 (en) Network node for supporting performance evaluation using a performance bitmap

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: 19946237

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: 19946237

Country of ref document: EP

Kind code of ref document: A1