WO2022160916A1 - 处理数据的方法、装置、系统及存储介质 - Google Patents

处理数据的方法、装置、系统及存储介质 Download PDF

Info

Publication number
WO2022160916A1
WO2022160916A1 PCT/CN2021/134273 CN2021134273W WO2022160916A1 WO 2022160916 A1 WO2022160916 A1 WO 2022160916A1 CN 2021134273 W CN2021134273 W CN 2021134273W WO 2022160916 A1 WO2022160916 A1 WO 2022160916A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
feature
type
data set
network element
Prior art date
Application number
PCT/CN2021/134273
Other languages
English (en)
French (fr)
Inventor
潘继雨
尹志东
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 EP21922506.7A priority Critical patent/EP4277223A4/en
Publication of WO2022160916A1 publication Critical patent/WO2022160916A1/zh
Priority to US18/360,899 priority patent/US20230370318A1/en

Links

Images

Classifications

    • 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
    • H04L41/02—Standardisation; Integration
    • H04L41/0246—Exchanging or transporting network management information using the Internet; Embedding network management web servers in network elements; Web-services-based protocols
    • 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
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
    • 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
    • H04L41/14—Network analysis or design
    • 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
    • H04L41/22—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks comprising specially adapted graphical user interfaces [GUI]
    • 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
    • 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
    • H04L43/16—Threshold monitoring

Definitions

  • the present application relates to the field of communications, and in particular, to a method, apparatus, system and storage medium for processing data.
  • the network element may collect multiple data corresponding to key performance indicators (key performance indicators, KPIs) related to services and/or networks. Usually, the network element may periodically collect data to obtain multiple data corresponding to the KPI, analyze whether the multiple data corresponding to the KPI is abnormal, and determine whether there is a risk in the service or network based on the analysis result.
  • KPIs key performance indicators
  • the network environment where the network element is located is constantly changing. After the network environment changes, the data collected by the network element also changes, which makes it difficult for the network element to analyze the abnormality of the collected data and reduces the accuracy of the collected data analysis. Spend.
  • the present application provides a method, device, system and storage medium for processing data, so as to improve the accuracy of data analysis.
  • the technical solution is as follows:
  • the present application provides a method for processing data.
  • a first data set sent by a network element is received, where the first data set includes a plurality of first data acquired by the network element.
  • At least one data feature corresponding to the plurality of first data is acquired based on the first data set.
  • Sending trigger information to the network element the trigger information including the at least one data feature and/or at least one feature type, the at least one feature type being related to the at least one data feature.
  • the at least one data feature is acquired based on the first data set, the at least one data feature corresponds to the data behavior of the first data in the first data set, and the at least one feature type is based on the at least one data feature.
  • a data feature is obtained. Even if the data collected by the network element changes due to changes in the network environment, through the above method, it can be analyzed that the data behavior in the first data set has changed, and the corresponding feature type can be obtained through at least one data feature of the data in the first data set.
  • the network element acquires at least one feature type based on the trigger information, and the network situation can be accurately reflected based on the at least one feature type and the data index acquired from the second data set. Therefore, the above method reduces the difficulty of analyzing the data in the second data set, and improves the accuracy of the data analysis in the second data set.
  • the at least one feature type is acquired based on the at least one data feature and the first corresponding relationship, because the first corresponding relationship includes the at least one data feature and the at least one feature
  • the at least one feature type can be quickly acquired based on the first correspondence, simplifying the implementation complexity, so that the trigger information can be sent to the network element in time.
  • the at least one feature type is acquired based on the at least one data feature, the first correspondence, and at least one data type, and the at least one data type includes the first data The type to which the first data in the set belongs, and the first correspondence includes the at least one data feature, the at least one data type, and/or the at least one feature type.
  • the acquired at least one feature type is associated with the at least one data type, so that the accuracy of acquiring the feature type can be improved.
  • the trigger information includes the at least one data characteristic and/or at least one data type, and the at least one data type includes a type to which the first data in the first data set belongs. Since the trigger information includes the at least one data type, the network element acquires the at least one feature type based on the at least one data type, which improves the accuracy of acquiring the at least one feature type.
  • the trigger information further includes an object identifier of at least one first object, and the at least one first object is located in the network element.
  • the network element can acquire the second data set based on the object identifier of the at least one first object, and the second data set includes the second data acquired by the at least one first object. Since the data in the second dataset is obtained by the at least one first object, the at least one first object may be associated with the first dataset, which can improve the accuracy of analyzing the data in the second dataset.
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set and an object identifier of at least one second object; wherein the at least one second object includes an object for acquiring each first data in the first data set.
  • the execution body of the method includes: a data processing system, a controller or a management device.
  • the first data set includes the second data set or part of the data in the second data set, or the second data set includes the network element sending the The data obtained after the first data set is described.
  • the present application provides a method for processing data.
  • a first data set is obtained, and the first data set includes a plurality of first data.
  • At least one feature type is acquired based on the first data set, and the at least one feature type corresponds to at least one data feature corresponding to the plurality of first data.
  • Data metrics are obtained based on the at least one feature type and a second data set, the second data set including a plurality of second data.
  • the at least one data feature is acquired based on the first data set, the at least one data feature corresponds to the data behavior of the first data in the first data set, and the at least one feature type is based on the at least one data feature.
  • a data feature is obtained. Even if the acquired data changes due to changes in the network environment, through the above method, it can be analyzed that the data behavior in the first data set has changed, and the corresponding feature type is obtained through at least one data feature of the data in the first data set.
  • the at least one feature type and the data index obtained from the second data set can accurately reflect the network situation. Therefore, the above method reduces the difficulty of analyzing the data in the second data set, and improves the accuracy of analyzing the data in the second data set.
  • the first data set is sent, and the first data set is used for the receiver of the first data set to obtain the at least one data feature and/or the at least one feature type .
  • Trigger information is received, the trigger information including the at least one data feature and/or the at least one feature type.
  • the at least one feature type is acquired based on the trigger information.
  • the network element does not need to acquire the at least one data feature and/or the at least one feature type based on the first data set, so that when the computing capability and storage capability of the network element are low, it does not affect the network element's acquisition of the data. Describe at least one feature type.
  • the trigger information includes the at least one data feature, and the at least one feature type is acquired based on the at least one data feature and a first correspondence, and the first correspondence includes the at least one data feature and the at least one feature type. Since the first correspondence includes the at least one data feature and the at least one feature type, the at least one feature type can be quickly acquired based on the first correspondence, the implementation complexity is simplified, and the at least one feature type can be quickly obtained A feature type.
  • the trigger information further includes at least one data type, and the at least one data type includes a type to which the first data in the first data set belongs.
  • the at least one feature type based on the at least one data feature, the at least one data type, and the first correspondence, where the first correspondence includes the at least one data feature, the at least one data type and the at least one feature type. Since the trigger information further includes the at least one data type, the acquired at least one feature type is associated with the at least one data type, so that the accuracy of acquiring the feature type can be improved.
  • the trigger information further includes an object identifier of at least one first object.
  • the second data set is obtained based on the object identification of the at least one first object, the second data set includes a plurality of second data obtained by the at least one first object. Since the data in the second dataset is obtained by the at least one first object, the at least one first object may be associated with the first dataset, which can improve the accuracy of analyzing the data in the second dataset.
  • the at least one data feature is acquired based on the first data set.
  • the at least one feature type is acquired based on the at least one data feature and a first corresponding relationship, where the first corresponding relationship includes the at least one data feature and the at least one feature type.
  • the network element directly acquires the at least one feature type without sending the first data set to the first device, thereby saving network resources.
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set and an object identifier of at least one second object; wherein the at least one second object includes an object for obtaining the first data in the first data set.
  • an abnormal situation of the second data in the second data set is determined based on the data indicator. Sending the determined abnormal condition and/or the data indicator.
  • the first data set includes the second data set or part of the data in the second data set, or the second data set includes the network element in the acquisition of all the data. The data obtained after the first data set is described.
  • the present application provides an apparatus for processing data, for executing the method in the first aspect or any possible implementation manner of the first aspect.
  • the apparatus includes a unit for performing the method in the first aspect or any possible implementation manner of the first aspect.
  • the present application provides an apparatus for processing data, for executing the method in the second aspect or any possible implementation manner of the second aspect.
  • the apparatus includes a unit for performing the method in the second aspect or any one possible implementation manner of the second aspect.
  • the present application provides a device for processing data, the device includes a processor and a computer program, the processor is configured to execute the computer program in the memory, so that the device completes the first aspect or the first A method in any possible implementation of an aspect.
  • the present application provides a device for processing data, the device includes a processor and a computer program, the processor is configured to execute the computer program in the memory, so that the device completes the second aspect or the second A method in any possible implementation of an aspect.
  • the present application provides a computer program product, the computer program product includes a computer program, and the computer program is loaded by a computer to realize any possible implementation of the first aspect, the second aspect, and the first aspect manner or any possible method of implementing the second aspect.
  • the present application provides a computer-readable storage medium for storing a computer program, and the computer program is loaded by a processor to execute the first aspect, the second aspect, and any possible implementation manner of the first aspect. or any possible implementation of the second aspect.
  • the present application provides a chip, including a memory and a processor, the memory is used to store computer instructions, and the processor is used to call and run the computer instructions from the memory to execute the above-mentioned first aspect, second aspect, and the first aspect.
  • the present application provides a system for processing data.
  • a data acquisition unit acquires a first data set, and the first data set includes a plurality of first data.
  • the data sending unit sends the first data set.
  • the data processing unit acquires at least one data feature corresponding to the plurality of first data based on the first data set.
  • the information sending unit sends trigger information, where the trigger information includes the at least one data feature and/or at least one feature type, and the at least one feature type is acquired by the data processing unit based on the at least one data feature.
  • the type acquisition unit acquires the at least one feature type based on the trigger information.
  • the indicator acquiring unit acquires data indicators based on the at least one feature type and a second data set, the second data set including a plurality of second data.
  • the data processing unit acquires the at least one data feature based on the first data set, the at least one data feature corresponds to the data behavior of the first data in the first data set, and the at least one feature type is all
  • the data processing unit acquires based on the at least one data feature. Even if the acquired data changes due to changes in the network environment, through the above system, it can be analyzed that the data behavior in the first data set has changed, and the type acquiring unit acquires the corresponding feature through at least one data feature of the data in the first data set In this way, the data index obtained by the index obtaining unit based on the at least one feature type and the second data set can accurately reflect the network situation.
  • the above system reduces the difficulty of analyzing the data in the second data set, and improves the accuracy of analyzing the data in the second data set.
  • the data processing unit further acquires the at least one feature type based on the at least one data feature and a first correspondence, where the first correspondence includes the at least one data feature and the at least one feature type. Since the first correspondence includes the at least one data feature and the at least one feature type, the data processing unit can quickly acquire the at least one feature type based on the first correspondence, simplifying the implementation complexity, and thus can The trigger information is sent to the network element in time.
  • the data processing unit acquires the at least one feature type based on the at least one data feature, the first correspondence and at least one data type, and the at least one data type includes The type to which the first data in the first data set belongs, and the first correspondence includes the at least one data feature, the at least one data type, and the at least one feature type.
  • the acquired at least one feature type is associated with the at least one data type, so that the accuracy of acquiring the feature type can be improved.
  • the trigger information includes the at least one data feature
  • the type obtaining unit obtains the at least one feature type based on the at least one data feature and the first corresponding relationship
  • the The first correspondence includes the at least one data feature and the at least one feature type.
  • the trigger information further includes at least one data type, and the at least one data type includes a type to which the first data in the first data set belongs.
  • the type obtaining unit obtains the at least one feature type based on the at least one data feature, the at least one data type, and the first correspondence, where the first correspondence includes the at least one data feature, the the at least one data type and the at least one feature type. Since the trigger information includes the at least one data type, the type obtaining unit obtains the at least one feature type based on the at least one data type, thereby improving the accuracy of obtaining the at least one feature type.
  • the trigger information further includes an object identifier of at least one first object.
  • the second data set includes a plurality of second data acquired by the at least one first object. Since the data in the second dataset is obtained by the at least one first object, the at least one first object may be associated with the first dataset, which can improve the accuracy of analyzing the data in the second dataset.
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set and an object identifier of at least one second object; wherein the at least one second object includes an object for obtaining the first data in the first data set.
  • the indicator acquiring unit further determines an abnormal situation of the second data in the second data set based on the data indicator.
  • the data sending unit also sends the determined abnormal situation and/or the data indicator.
  • the first data set includes the second data set or part of the data in the second data set, or the second data set includes when the first data is acquired data acquired after the set.
  • FIG. 1 is a schematic diagram of a network architecture provided by an embodiment of the present application.
  • FIG. 2 is a flowchart of a method for processing data provided by an embodiment of the present application
  • FIG. 3 is a schematic structural diagram of a configuration interface provided by an embodiment of the present application.
  • FIG. 5 is a flowchart of another method for processing data provided by an embodiment of the present application.
  • FIG. 6 is a flowchart of another method for processing data provided by an embodiment of the present application.
  • FIG. 7 is a schematic structural diagram of an apparatus for processing data provided by an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of another apparatus for processing data provided by an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of a system for processing data provided by an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a device for processing data provided by an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of another apparatus for processing data provided by an embodiment of the present application.
  • an embodiment of the present application provides a network architecture 100, including:
  • a first device 101 and at least one network element 102 the first device 101 communicates with each of the at least one network element 102.
  • the first device 101 and each of the network elements access a network, and the network includes a local area network or a wide area network. That is, the first device 101 communicates with each of the network elements in the local area network, or the first device 101 may be deployed remotely or in the cloud.
  • the first device 101 establishes a network connection with the network element to implement communication with the network element.
  • the first device 101 communicates with the network element, which will not be listed one by one here.
  • the first device 101 is a network cloud engine (NCE), a data processing system, a controller or a management device, and the like.
  • the at least one network element 102 includes one or more of the following network devices: a probe, a server, a base station, a switch, a gateway, a router, an optical network unit (ONU), an optical line terminal (OLT) , wireless local area network (wireless local area network, WLAN) equipment or firewall, etc.
  • the network element 102 includes functions such as processing data and/or forwarding services.
  • the first device 101 includes functions such as analyzing, managing and/or controlling the network element 102 .
  • the process of processing the data by the network element 102 is as follows: the network element 102 acquires data, analyzes whether the data is abnormal based on at least one feature type, and the at least one feature type is abnormal.
  • a characteristic type corresponds to the data behavior of the data; it is determined whether there is a risk in the business or network related to the data based on the analysis result, the business includes the business transmitted by the network element 102, and the network includes the network.
  • the network environment where the network element 102 is located is constantly changing. After the network environment where the network element 102 is located changes, the data behavior of the data acquired by the network element 102 may change. At this time, the at least one feature The type may no longer correspond to the changed data behavior. If the network element 102 continues to analyze the acquired data based on the at least one feature type, the accuracy of analyzing the data will be reduced.
  • the data processed by the router is the number of routes, and at least one feature type in the router includes an average value.
  • the router obtains a plurality of first route numbers, and the data behavior of the plurality of first route numbers corresponds to an average value.
  • An average value of the number of the multiple first routes is acquired, and whether the number of the multiple first routes is abnormal is analyzed based on the acquired average value.
  • the router obtains multiple second route numbers, and the multiple second route numbers no longer correspond to the average value, but correspond to the variance. At this time, the router continues to obtain the multiple second route numbers.
  • the average value of the number of the second routes is analyzed based on the obtained average value to determine whether the number of the second routes is abnormal, and an error occurs in the analysis, which reduces the accuracy of analyzing the number of the plurality of second routes.
  • the first device 101 controls or manages the network element 102 .
  • the process of control or management is:
  • the first device 101 acquires a first data set sent by the network element 102 , where the first data set includes a plurality of first data acquired by the network element 102 .
  • the first device 101 acquires at least one data feature corresponding to the plurality of first data based on the first data set; and sends trigger information to the network element 102, where the trigger information includes the at least one data feature and/or at least one A feature type, the at least one feature type is obtained based on the at least one data feature.
  • the network element 102 receives the trigger information, acquires the at least one feature type based on the trigger information, and analyzes second data in a second data set based on the at least one feature type, where the second data set includes the network element 102 to obtain a plurality of second data.
  • the first data set includes the second data set or part of the second data in the second data set, or the second data set is data obtained by the network element 102 after sending the first data set. Since the at least one feature type is obtained based on the first data acquired by the network element 102, the at least one feature type corresponds to the data behavior of the first data in the first data set, so based on the at least one feature When analyzing the second data in the second data set, the accuracy of the analysis can be improved. In this way, the first device 101 controls or manages the network element 102. Of course, there are other ways for the first device 101 to control or manage the network element 102, which are not listed here.
  • the first device 101 obtains the first data set, and obtains the feature based on the number of the plurality of second routes in the first data set
  • the type includes variance
  • trigger information is sent to the router, and the trigger information includes variance.
  • the router obtains the variance of the second data set, and analyzes whether the number of routes in the second data set is abnormal based on the obtained variance.
  • the second data set includes the number of routes obtained by the router after the network environment changes.
  • the data in the data set includes data corresponding to at least one KPI.
  • the at least one KPI includes one or more of the number of routes, the number of lost packets, the number of routing entries, the number of bit errors, and the delay.
  • the data behavior of the first data in the first data set includes data change behavior.
  • the first data in the first data set may be periodically changing data, shock changing data, or trend changing data.
  • the at least one data feature is used to reflect the data behavior of the first data in the first data set.
  • the at least one data feature includes: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set, and/or an object identifier of at least one object.
  • the at least one object includes an object for acquiring each first data in the first data set.
  • the data waveform features include one or more of periodic type, oscillation type, horizontal type, sudden change type and trend type.
  • the at least one feature type includes one or more of mean, variance, median, maximum, minimum, and the like.
  • the first device 101 stores the first correspondence, and/or, for each network element in the at least one network element 102, the network element 102 stores the first correspondence.
  • the first correspondence includes data features and feature types.
  • Each record in the first correspondence may be configured by a technician, and/or learned by the first device 101 or the network element 102 by itself.
  • the record may be configured by a technician in advance, and/or the record is generated by the first device 101, and /or, the records are sent by other devices, and the other devices include network elements and/or knowledge bases (expert knowledge bases and/or experience knowledge bases) and the like.
  • the first device acquires at least one data feature and at least one feature type
  • the first device 101 further queries whether a target record is saved in the first correspondence, where the target record includes the at least one data feature and the at least one feature type.
  • the target record is saved in the first corresponding relationship.
  • the records stored in the first pair of relationships in the network element 102 are also obtained according to the above process, which will not be described in detail here.
  • the record includes one or more data features and one or more feature types.
  • the data characteristic included in the first record is "periodic type", and the characteristic type is "average value”.
  • the second record in Table 1 the second record includes two data features and two feature types, the two data features are “periodic and oscillatory”, and the two feature types are "variance and medium” value”.
  • serial number Data features Feature type 1 Periodic average value 2 Periodic and Oscillating variance and median ... ... ...
  • the records may further include one or more data types, and the data types may be KPIs, business data, and/or control information, and the like.
  • the data characteristic included in the first record is "periodic”
  • the data type is "number of routes”
  • the characteristic type is "average value”.
  • the second record in Table 2 the second record includes two data features, three data types and two feature types, the two data features are “periodic and oscillating", the three data types are "number of lost packets, number of routes and delay", and the two feature types are "variance and median”.
  • an embodiment of the present application provides a method 200 for processing data.
  • the method 200 can be applied to the network architecture shown in FIG. 1 .
  • Device 101 including:
  • Step 201 The first device receives a first message, where the first message includes a second data set, and the second data set includes data acquired by a network element.
  • step 201 the first device periodically receives the first message sent by the network element, and the second data set in the first message includes data acquired by the network element in one period.
  • the network element periodically sends the second data set, wherein the network element determines the first period, acquires a plurality of data in the first period to obtain the second data set, and the second data set is included in the first period and send a first message including the second data set to the first device.
  • the network element also continues to determine the first period, and continues to acquire a plurality of data within the determined first period, so that the network element periodically sends the second data set.
  • the manner in which the network element obtains data includes a manner of collecting data and/or a manner of receiving data, and the like.
  • the network element collects KPIs using the method of collecting data, that is, the second data set includes KPIs collected by the network element; and/or, the network element uses the receiving data method to receive data in the network, that is, the second data set includes the data received by the network element.
  • the data received by the network element may be service data and/or control information.
  • the second data set includes at least one first data sequence, for each first data sequence, the first data sequence corresponds to at least one object and data type, the at least one object being located in a network element.
  • the data in the first data sequence is data belonging to the data type acquired by the at least one object. That is, the data in the second data set is data obtained by at least one object in the network element, and the data in the second data set belongs to at least one data type.
  • the at least one object includes an interface, a collector and/or a single board on the network element.
  • the data types include KPIs, business data and/or control information, and the like.
  • the data in the second data set belong to the same data type, or belong to different data types. Taking KPI as an example, that is, the data in the second data set belongs to the same KPI, or, the data in the second data set belongs to multiple KPIs.
  • the KPI includes one or more of the number of routes, the number of lost packets, the number of routing entries, the number of bit errors, and the delay.
  • the data in the second data set belongs to the same KPI
  • the KPI is the number of routes
  • the data in the second data set includes the number of routes acquired by the network element in a first period.
  • the data in the second dataset belongs to multiple KPIs
  • the multiple KPIs include the number of routes, the number of lost packets, and the number of routing entries
  • the data in the second dataset includes routes obtained by the network element in the first cycle. number, number of lost packets, and number of routing table entries.
  • the first period length of the first period is configured by the network element itself based on requirements, or the first period length is configured on the network element by the first device.
  • the first device before step 201 is executed, the first device generates task information, the task information includes the first cycle length, sends the task information to the network element, and the network element receives the first cycle length.
  • a task message sent by a device In this way, after receiving the task information, the network element periodically acquires the second data set based on the length of the first period.
  • the task information further includes at least one data type, so that the network element acquires data of each data type in the at least one data type in the first cycle to obtain the second data set.
  • the operations for the first device to generate task information are:
  • the first device acquires the first period length, generates task information including the first period length, and sends the task information to the network element.
  • the first device further acquires the at least one data type, and the task information thus generated further includes the at least one data type.
  • the first device further acquires a network element range, where the network element range includes a network element identifier of at least one network element. In this way, after generating the task information, the first device sends the task information to each network element based on the network element identifier of each network element included in the network element range, so as to trigger the task information within the network element range.
  • Each network element obtains data.
  • the first cycle length is configured by a technician on the first device.
  • the first device displays a configuration interface
  • the technician inputs the first cycle length in the configuration interface
  • the first device obtains the first cycle length from the configuration interface.
  • the technician also inputs at least one data type on the configuration interface, and the first device further obtains the at least one data type from the configuration interface.
  • the technician also inputs a network element range on the configuration interface, where the network element range includes a network element identifier of at least one network element.
  • the first device further acquires the network element range from the configuration interface.
  • the first device displays the configuration interface
  • the technician enters the first cycle length in the configuration interface as 5 minutes
  • the input data type is the number of routes
  • the input network element range includes the network element identifier of network element 1" ID-NE1" and the NE ID of NE 2 "ID-NE2”, and then click the OK button in the configuration interface to make the OK button generate a trigger command.
  • the length of the first period obtained from the configuration interface is 5 minutes
  • the type of data obtained is the number of routes
  • the range of network elements obtained includes the network element identifier of network element 1.
  • ID-NE1 and the network element identifier "ID-NE2" of the network element 2.
  • Generate task information the task information includes a first cycle length of 5 minutes and a data type of "number of routes", based on the network element identifier "ID-NE1" of the network element 1 and the network element identifier "ID-NE1" of the network element 2.
  • NE2 respectively send the task information to network element 1 and network element 2 to trigger both network element 1 and network element 2 to obtain the number of routes, and obtain a second data set including the number of routes every 5 minutes.
  • the first message further includes attribute information corresponding to each first data sequence in the second data set.
  • the attribute information corresponding to the first data sequence includes the acquisition time corresponding to each data in the first data sequence, the data type corresponding to the first data sequence and/or the The object identifier of the object corresponding to the first data sequence.
  • the above-mentioned first message also includes network element information of the network element, where the network element information includes one or more of the network element identifier, network element name, network element address, and network element type of the network element.
  • the network element sends the first message to the first device based on a network transmission protocol, where the format of the first message is a message format defined by the network transmission protocol.
  • the network transfer protocol includes a network configuration protocol (network configuration protocol, Netconf) protocol, a secure file transfer protocol (secret file transfer protocol, SFTP) file transfer protocol or telemetry (telemetry) protocol and the like.
  • Netconf network configuration protocol
  • SFTP secure file transfer protocol
  • telemetry telemetry
  • a data modeling language (yet another next generation, YANG) model is used to encapsulate the second data set and/or attribute information of the second data set in the first message.
  • Step 202 The first device saves the second data set in the first message.
  • the first device acquires the network element information of the network element, acquires a record including the network element information based on the network element information and the second correspondence, and saves the second data set to the acquired record middle.
  • Each record in the second correspondence relationship includes network element information of one network element and a first data set corresponding to the one network element, and the first data set includes the data obtained by the one network element.
  • the first data set may include at least one second data sequence.
  • the second data sequence corresponds to an object and a data type in the one network element, and the second data sequence includes the data type acquired by the one object and belonging to the data type. data.
  • the record may also include attribute information corresponding to each second data sequence.
  • the operation of the first device to save the second data set in the acquired record is: in the case that the first message further includes attribute information corresponding to each first data sequence in the second data set, the first device based on each For attribute information corresponding to a data sequence, each first data sequence is stored in the acquired record.
  • the first device further performs preprocessing on the data in the second data set, where the preprocessing includes denoising processing, complementary value processing, and/or stitching processing, and the like.
  • the so-called denoising processing refers to that the first device removes the noise of the data included in the second data set.
  • the noise may be generated during the transmission of the second data set.
  • the so-called complementary value processing means that the first device detects whether there is data loss in the second data set, and if there is data loss, supplements the missing data in the second data set.
  • the data loss may be generated during the transmission of the second data set.
  • the first device performs complementary value processing on the second data set.
  • the process of the complementary value processing is as follows:
  • the first device For each first data sequence in the second data set, the first device obtains the acquisition time corresponding to each data in the first data sequence from the attribute information corresponding to the first data sequence, and based on the acquisition time corresponding to each data, Determine the acquisition time corresponding to the lost data; based on the determined acquisition time, select the first data and the second data from the first data sequence, the first data is the most recent data acquired before the lost data, and the second data is in the The most recent data acquired after the missing data; based on the first data and the second data, the missing data is supplemented in the first data sequence.
  • the first device calculates a mean between the first data and the second data, and uses the calculated mean as the missing data.
  • the first device calculates the variance between the first data and the second data, and treats the calculated variance as missing data.
  • the first device performs weighted calculation on the first data and the second data, and uses the calculated value as the missing data.
  • the first device may also acquire the lost data in other ways, which will not be listed one by one here.
  • the so-called splicing process means that the first device splices the second data sequence included in the acquired record and the first data sequence in the second data set into a data sequence, and updates the second data sequence included in the acquired record to the spliced data sequence. , so as to save the second dataset into the acquired record.
  • the first message may include attribute information corresponding to each first data sequence in the second data set.
  • the first device obtains the object identifier and data type of the object from attribute information corresponding to the first data sequence. Based on the object identification of the object and the data type, a second data sequence corresponding to the object and the data type is selected from the acquired records. The first data sequence and the second data sequence are spliced into a data sequence, and the second data sequence is updated to a spliced data sequence in the acquired record.
  • Step 203 The first device acquires a first data set, where the first data set includes at least one second data set sent by the network element.
  • the first data set includes the second data set sent by the network element received by the first device in the second period, and the second period length of the second period is greater than or equal to the first period length.
  • step 203 the first device determines a second period, and acquires the first data set corresponding to the network element from the second correspondence at the end of the second period.
  • the first device also obtains from the second correspondence relationship at least one data type corresponding to the first data set and/or an object identifier of at least one object, the at least one data type includes the type to which the data in the first data set belongs, and the at least one object is An object that gets the data in the first dataset.
  • the first device after acquiring the first data set corresponding to the network element from the second correspondence, the first device further deletes the record including the network element information of the network element from the second correspondence. In this way, when the first device receives the data set sent by the network element in the next second period, based on the network element information of the network element and the received data set, it recreates the network element information including the network element information and the received data set in the second correspondence
  • the record of the data set ensures that the data included in each data set in the second corresponding relationship is the data acquired in a second period.
  • the second cycle length is configured by the first device, and the first device configures the second cycle length based on requirements. or,
  • the second cycle length is configured by the technician in the first device.
  • the first device displays the configuration interface as shown in FIG. 3, and the technician inputs the second cycle length in the configuration interface.
  • the second cycle length is 24 hours, and the first device obtains the second cycle length from the configuration interface. .
  • Step 204 The first device acquires, based on the first data set, at least one data feature corresponding to the data in the first data set.
  • the first data set includes at least one second data sequence, and each second data sequence in the at least one second data sequence is processed to obtain data features corresponding to each second data sequence, respectively.
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the data in the first data set, a data content feature in the first data set, and an object identifier of at least one object.
  • the at least one object includes an object for acquiring each data in the first data set.
  • a point corresponding to each data is determined in a coordinate system, and the horizontal axis of the coordinate system is time, The vertical axis is the data value. Connect the points corresponding to each data in the coordinate system to obtain the data waveform corresponding to the second data sequence, process the data waveform using a waveform processing model, and output the data waveform characteristics corresponding to the second data sequence .
  • the data waveform characteristics include one or more of periodic type, oscillation type, horizontal type, sudden change type and trend type.
  • each training sample includes at least one data waveform and the at least one data waveform Corresponding data waveform characteristics.
  • An intelligent algorithm is trained using the plurality of training samples to obtain a waveform processing model.
  • Intelligent algorithms include convolutional neural networks, random forests, logistic regression, or support vector machines (SVMs).
  • the data content features include service features and/or user features and the like.
  • the service characteristics include a service identifier, the number of times of access to a service source and/or the number of service flows, and the like, and the user characteristics include the number of times a user requests a service, and the like.
  • the first data set may include service data, etc.
  • the service data includes a service flow received by a network element, etc.
  • the service flow includes a service identifier, quintuple information, a domain name of the service, and/or a uniform resource locator (uniform resource locator).
  • resource locator, URL uniform resource locator
  • the domain name, source address and/or URL of the service flow are used to identify the network source of the service flow, and the network source may also be referred to as a service source.
  • the number of access times to service sources corresponding to different service flows is obtained by performing statistics on source addresses, domain names or URLs in different service flows in the first data set.
  • the quintuple information of the service flow can be used to identify a service flow, and the number of service flows is obtained by counting different quintuple information in the first data set.
  • the destination address of the service flow may be the address of the user requesting the service flow. For any destination address in the first data set, the user request corresponding to the destination address is obtained by collecting statistics on the service flow including the destination address. business times.
  • the first device obtains the object identifier of the object corresponding to each second data sequence in the at least one second data sequence from the second correspondence, that is, obtains the at least one object object ID.
  • the first device may also acquire at least one feature type based on the at least one data feature and the first corresponding relationship.
  • the first device determines at least one record from the first correspondence, the data feature included in the at least one record is the same as the at least one data feature, and acquires the feature type included in the at least one record.
  • the record may be configured by a technician, and/or the record is generated by the first device, and/or the record is the first
  • the device receives data sent by other devices, and other devices include network elements and/or knowledge bases.
  • the first device when the first device acquires the at least one data feature and the at least one feature type, the first device further queries whether a target record is stored in the first correspondence, where the target record includes the at least one data feature. For the data feature and the record of the at least one feature type, if the target record is not saved, the target record is saved in the first corresponding relationship.
  • the first device further acquires at least one data type, and acquires at least one characteristic type based on the at least one data characteristic, the first correspondence, and the at least one data type, and the at least one data type includes that the data in the first data set belongs to type.
  • the first device determines from the first correspondence relationship at least one record, the data feature included in the at least one record is the same as the at least one data feature, and the data type included in the at least one record is the same as the at least one data feature. If one data type is the same, obtain the feature type included in the at least one record.
  • Step 205 The first device sends trigger information to the network element, where the trigger information includes the at least one data feature and/or the at least one feature type, so that the network element processes the data acquired by the network element based on the trigger information.
  • the trigger information further includes the object identifier of the at least one object obtained above, and/or the at least one data type obtained above.
  • the first device sends trigger information to the network element through a network transmission protocol
  • the format of the trigger information is a message format defined by the network transmission protocol
  • the network transmission protocol includes Netconf, transmission control protocol (transmission control protocol, TCP) or User datagram protocol (user datagram protocol, UDP), etc.
  • the network element After receiving the trigger information, the network element processes the data acquired by the network element.
  • the process of the network element processing data is as follows:
  • At least one feature type is acquired based on the trigger information, a data indicator is acquired based on the at least one feature type and the second data set, and an abnormal condition of the data in the second data set is determined based on the data indicator.
  • the at least one feature type includes one or more of an average value, a variance, a median value, a maximum value and a minimum value, etc., so the network element obtains data indicators based on the at least one feature type and the second data set, including all The mean value of the data in the second data set, the variance of the data in the second data set, the median value of the second data set, the maximum value in the second data set, and the minimum value in the second data set one or more of the values, etc.
  • the network element includes a third correspondence
  • each record in the third correspondence includes at least one feature type and a data processing algorithm
  • the data processing algorithm is used to obtain based on the at least one feature type and the second data set A data indicator, and based on the data indicator, an abnormal situation of the data in the second data set is determined.
  • the data processing algorithm further includes at least one threshold, the data processing algorithm compares the data indicator with the at least one threshold, and determines anomalies in the data in the second data set based on a result of the comparison .
  • the network element obtains at least one data processing algorithm based on the at least one feature type and the third correspondence, and determines an abnormal situation of the data indicator by the at least one data processing algorithm based on the at least one feature type and the second data set,
  • the data indicator is compared with a threshold in the at least one data processing algorithm, and an abnormality of the data in the second data set is determined based on a result of the comparison.
  • the third correspondence includes the correspondence between the mean value and the data processing algorithm including the mean value threshold.
  • the network element receives trigger information, and at least one feature type in the trigger information includes an average value.
  • the data processing algorithm is obtained.
  • a data indicator of the second data set is obtained, where the data indicator is an average value of the data in the second data set.
  • the data indicator is compared with the average threshold in the data processing algorithm, and if the comparison result is that the data indicator is greater than the average threshold, in this case, it can be determined that the data indicator is abnormal and the data indicator is abnormal.
  • the data in the second data set is abnormal; if the comparison result is that the data index is less than or equal to the average threshold, in this case, it can be determined that the data index is normal and the data in the second data set is normal.
  • the above-mentioned second data set may be acquired by the network element before receiving the trigger information.
  • the first data set may include the second data set or part of the data in the second data set.
  • the above-mentioned second data set may also be obtained by the network element after receiving the trigger information.
  • the first data set does not include the second data set, and the second data set is obtained when the first device obtains the first data.
  • the first device receives the second data set sent by the network element and saves the second data set.
  • the first data set is processed to obtain at least one data feature, and trigger information is sent to the network element, where the trigger information includes the at least one data feature and at least one feature type, and the at least one feature type is based on the At least one data feature is acquired.
  • trigger information is received at the network element, the at least one feature type is acquired based on the trigger information, a data indicator is determined based on the at least one feature type and the second data set, and the second data is analyzed based on the data indicator set of exceptions.
  • the network element can send the second data set to the first device whenever it acquires the second data set, the network element does not need to store the second data set for a long time, and the first device processes the first data set, so that the network element does not need to store the second data set for a long time. There is no need to process the first data set to obtain data features, so the embodiments of the present application can be applied to network elements with low computing performance and/or storage performance. Since the at least one feature type is obtained based on the first data set, the at least one feature type corresponds to the data behavior in the first data set.
  • the embodiment of the present application reduces the difficulty of analyzing the data in the second data set, so that the abnormal situation of the second data set can be accurately analyzed based on the data index, and the accuracy of the data analysis is improved.
  • an embodiment of the present application provides a method 400 for processing data.
  • the method 400 may be applied to the network architecture 100 shown in FIG. 1 . $102, including:
  • Step 401 The network element acquires a plurality of second data to obtain a second data set, where the second data set includes the plurality of second data.
  • the network element determines a first period, acquires a plurality of second data in the first period to obtain a second data set, and the second data set includes the plurality of second data acquired in the first period.
  • the manner in which the network element acquires data includes acquiring data in real time, acquiring data periodically, and/or acquiring data randomly, and the like.
  • the second data set includes at least one first data sequence, each first data sequence including data arranged by acquisition time.
  • the data in each first data sequence may be data acquired by at least one object in the network element, and/or data of the same data type.
  • the network element continues to determine the first period, and continues to acquire a plurality of data within the determined first period, so that the network element periodically acquires the second data set.
  • the network element includes at least one object, and for each of the at least one object, the object acquires data in the first cycle.
  • the network element obtains the data obtained by each object to obtain the second data set. That is, the data included in the second data set may be data obtained by the same object, or the data included in the second data set may be data obtained by multiple objects.
  • the second data set includes a first data sequence corresponding to each object, and the first data sequence corresponding to the object includes data acquired by the object.
  • the at least one object includes an interface, a collector and/or a single board on the network element.
  • the data in the second data set belongs to the same data type, or the data in the second data set belongs to multiple data types.
  • the above data types are KPIs, business data and/or control information and the like.
  • KPI as an example, that is, the data in the second data set belongs to the same KPI, or, the data in the second data set belongs to multiple KPIs.
  • the KPI includes one or more of the number of routes, the number of lost packets, the number of routing entries, the number of bit errors, and the delay.
  • the object obtains data of at least one data type
  • the object in the second data set corresponds to at least one first data sequence
  • each first data sequence corresponds to One data type
  • the data included in each first data sequence belong to the same data type.
  • the second data set includes at least one first data sequence, each first data sequence corresponding to an object and data type.
  • step 401 after obtaining the second data set, the network element further analyzes the abnormal situation of the data in the second data set, that is, analyzes whether the data in the second data set is abnormal.
  • the process of analyzing the second data set in detail by the network element will be described in detail in the subsequent content, and will not be described in detail here.
  • the first period length of the first period may be configured by the network element itself based on requirements, or the first period length may be configured on the network element by the first device.
  • the network element receives task information sent by the first device, where the task information includes the first cycle length. After receiving the task information, the network element periodically acquires the second data set based on the first period length.
  • the task information may further include at least one data type, so that the network element acquires data of each data type in the at least one data type in the first cycle to obtain the second data set.
  • Step 402 The network element sends a first message to the first device, where the first message includes the second data set.
  • the first message further includes attribute information corresponding to each first data sequence in the second data set.
  • the attribute information corresponding to the first data sequence includes the acquisition time corresponding to each data in the first data sequence, the data type corresponding to the first data sequence and/or the The object identifier of the object corresponding to the first data sequence.
  • the first message further includes network element information of the network element, where the network element information includes one or more of the network element identifier, network element name, network element address, and network element type of the network element. .
  • the network element sends the first message to the first device based on a network transmission protocol, where the format of the first message is a message format defined by the network transmission protocol.
  • the network transfer protocol includes Netconf protocol, SFTP file transfer protocol, telemetry protocol, and the like.
  • the YANG model is used to encapsulate the second data set and/or attribute information of the second data set in the first message.
  • the network element repeats the operations of steps 401 to 402 above, that is, the network element periodically sends the second data set to the first device.
  • the first device saves the received second data set, and acquires the first data set when receiving the second data set for a period of time, where the first data set includes at least one second data set.
  • the first device acquires at least one data feature and/or at least one feature type based on the first data set; and sends trigger information to the network element, where the trigger information includes the at least one data feature and/or the at least one feature type.
  • the trigger information further includes at least one feature type and/or at least one object identifier of the object.
  • the at least one feature type includes a data type to which the data in the first data set belongs, and the at least one object includes an object for acquiring data in the first data set.
  • the first device acquires at least one data feature and/or at least one feature type based on the first data set and sends the trigger information for a detailed implementation process, refer to the content of the method 200 shown in FIG. 2 , and will not be described in detail here.
  • Step 403 The network element receives trigger information sent by the first device, where the trigger information includes at least one data type and/or at least one feature type.
  • the network element receives the trigger information sent by the first device through a network transmission protocol, where the format of the trigger information is a message format defined by the network transmission protocol, and the network transmission protocol includes Netconf, TCP or UDP, etc. .
  • Step 404 The network element acquires the at least one feature type based on the trigger information, and acquires a data indicator based on the at least one feature type and the second data set.
  • the network element reads the at least one feature type from the trigger information.
  • the network element acquires the at least one feature type based on the at least one data feature and the first corresponding relationship.
  • the network element acquires the at least one characteristic type based on the at least one data characteristic, the at least one data type and the first correspondence.
  • the at least one feature type includes one or more of mean, variance, median, maximum value, and minimum value, etc., so based on the at least one feature type and the data indicators obtained from the second data set, including the second data One or more of the mean of the data in the set, the variance of the data in the second data set, the median of the second data set, the maximum value in the second data set, the minimum value in the second data set, and the like.
  • the network element After acquiring the data indicator, the network element determines an abnormal situation of the data in the second data set based on the data indicator.
  • the network element includes a third correspondence
  • each record in the third correspondence includes at least one feature type and a data processing algorithm
  • the data processing algorithm is used to obtain based on the at least one feature type and the second data set A data indicator, and based on the data indicator, an abnormal situation of the data in the second data set is determined.
  • the data processing algorithm further includes at least one threshold, the data processing algorithm compares the data indicator with the at least one threshold, and determines anomalies in the data in the second data set based on a result of the comparison .
  • step 404 the operation of the network element to obtain the data indicator is as follows:
  • the network element obtains at least one data processing algorithm based on the at least one feature type and the third correspondence, and determines an abnormal situation of the data indicator by the at least one data processing algorithm based on the at least one feature type and the second data set,
  • the data indicator is compared with a threshold in the at least one data processing algorithm, and an abnormality of the data in the second data set is determined based on a result of the comparison.
  • the third correspondence includes the correspondence between the mean value and the data processing algorithm including the mean value threshold.
  • the network element receives trigger information, and at least one feature type in the trigger information includes an average value. Based on the average value and the third correspondence, the data processing algorithm is obtained. Based on the at least one feature type and the second data set, a data index of the second data set is obtained, where the data index is an average value of the data in the second data set.
  • Comparing the data indicator with the average threshold if the comparison result is that the data indicator is greater than the average threshold, in this case, it can be determined that the data indicator is abnormal and the data in the second data set is abnormal ; If the comparison result is that the data index is less than or equal to the average value threshold, in this case, it is determined that the data index is normal and the data in the second data set is normal.
  • the record includes at least one feature type, at least one data type and a data processing algorithm, and the record indicates that the data processing algorithm is used for pairing belonging to the attribute type based on the at least one feature type Data of the at least one data type is processed.
  • the operation of the network element acquisition data processing algorithm is: when the trigger information includes at least one data type, the network element acquires data processing based on the at least one data type, the at least one feature type and the third correspondence algorithm.
  • the network element also sends the determined abnormal condition and/or data indicator to the first device.
  • the network element When determining that the data in the second data set is abnormal, the network element sends the abnormal situation to the first device. Wherein, the network element may send the abnormal situation to the first device in the manner of event reporting.
  • the above-mentioned second data set may be acquired by the network element before receiving the trigger information.
  • the first data set may include the second data set or part of the data in the second data set.
  • the above-mentioned second data set may also be obtained by the network element after receiving the trigger information.
  • the first data set does not include the second data set, and the second data set is obtained when the first device obtains the first data.
  • the network element when the trigger information further includes the object identifier of the at least one object, the network element further acquires a second data set based on the object identifier of the at least one object, and the second data set includes the at least one object acquired of multiple data.
  • the network element may not need to send the second data set to the first device, so that the first device can acquire at least one data feature and/or at least one feature type.
  • the network element after acquiring the second data set, the network element locally saves the second data set.
  • a first data set is acquired, the first data set includes at least one second data set acquired within the period of time. The network element processes the data in the first data set to obtain the at least one data feature, and obtains the at least one feature type based on the at least one data feature.
  • the network element includes a first correspondence, and the network element acquires the at least one feature type based on the at least one data feature and the first correspondence.
  • the network element acquires at least one data type, the at least one data type includes the data type to which the data in the first data set belongs, and acquires the at least one data feature based on the at least one data feature, the at least one data type and the first correspondence At least one feature type.
  • the network element acquires multiple pieces of data, obtains a second data set including the multiple pieces of data, and sends the second data set to the first device.
  • the network element periodically sends the second data set to the first device, so that the first device processes the first data set to obtain at least one data feature and/or at least one feature type, and the first data set includes the data sent by the network element. at least one second dataset.
  • the network element receives trigger information sent by the first device, where the trigger information includes the at least one data feature and/or at least one feature type.
  • the network element acquires the at least one feature type based on the trigger information, determines a data indicator based on the at least one feature type and the second data set, and analyzes an abnormal situation of the second data set based on the data indicator. Since the network element can send the second data set to the first device whenever it acquires the second data set, the network element does not need to store the second data set for a long time, and the first device processes the first data set, so that the network element does not need to store the second data set for a long time. There is no need to process the first data set to obtain data features, so the embodiments of the present application can be applied to network elements with low computing performance and/or storage performance.
  • the at least one feature type is acquired based on the first data set
  • the at least one feature type corresponds to the data behavior in the first data set.
  • the embodiment of the present application can analyze that the data behavior in the first data set has changed, and obtain the corresponding at least one data feature through at least one data feature of the data in the first data set.
  • Feature type The data index obtained by the network element based on the at least one feature type and the second data set cannot accurately reflect the network situation, so that the abnormal situation of the second data set can be accurately analyzed based on the data index, and the accuracy of the data analysis is improved. .
  • the network element does not need to send the second data set to the first device, that is, the network element obtains the first data set by itself, processes the first data set to obtain at least one data feature, and obtains at least one data feature based on the at least one data feature type, which can save network resources.
  • an embodiment of the present application provides a method 500 for processing data, and the method 500 is applied to the network architecture 100 shown in FIG. 1 .
  • the first device receives the data obtained by the network element to obtain a first data set, the first data set includes a plurality of data obtained by the network element, and obtains at least one data feature or at least one feature type based on the first data set.
  • the first device sends trigger information to the network element, where the trigger information includes at least one data feature or at least one feature type, and the network element processes the second data set based on the trigger information.
  • the method 500 includes:
  • Step 501 The network element acquires a plurality of data to obtain a second data set, and the second data set includes the plurality of data.
  • step 401 in the method 400 shown in FIG. 4 please refer to the relevant content in step 401 in the method 400 shown in FIG. 4 , which will not be described in detail here.
  • Step 502 The network element sends a first message to the first device, where the first message includes the second data set.
  • Step 503 The first device receives the first message, and saves the second data set in the first message.
  • Step 504 The first device acquires a first data set, where the first data set includes at least one second data set sent by the network element.
  • Step 505 The first device acquires, based on the first data set, at least one data feature corresponding to the data in the first data set.
  • Step 506 The first device sends trigger information to the network element, where the trigger information includes the at least one data feature and/or at least one feature type, and the at least one feature type is obtained based on the at least one data feature.
  • Step 507 The network element receives the trigger information, obtains at least one feature type based on the trigger information, and obtains a data indicator based on the at least one feature type and the second data set.
  • step 404 in the method 400 shown in FIG. 4 For a detailed implementation process for the first device to acquire at least one feature type and a data indicator, please refer to the relevant content in step 404 in the method 400 shown in FIG. 4 , which will not be described in detail here.
  • the first device receives the second data set sent by the network element and saves the second data set.
  • the first data set is processed to obtain at least one data feature, and trigger information is sent to the network element, where the trigger information includes the at least one data feature and at least one feature type.
  • the network element acquires at least one feature type based on the trigger information, determines a data indicator based on the at least one feature type and the second data set, and analyzes an abnormal situation of the second data set based on the data indicator. Since the at least one feature type is acquired based on the first data set, the at least one feature type corresponds to the data behavior in the first data set.
  • the embodiments of the present application reduces the difficulty of analyzing the data in the second data set, thereby accurately analyzing the abnormal situation of the second data set based on the data indicators, and improving the accuracy of the data analysis.
  • the embodiment of the present application provides a specific example to describe the method 500 in detail.
  • the network element is a router as an example, and it is assumed that the router includes The first veneer and the second veneer, the first veneer is the first object of the router, and is used to obtain the number of routes, and the second veneer is the second object of the router, and is also used to obtain the number of routes.
  • the examples include:
  • Step 601 The router acquires the number of multiple routes acquired by the first board in a first cycle, and acquires the number of multiple routes acquired by the second board in the first cycle, to obtain a second data set.
  • the second data set includes two first data sequences, which are first data sequence 1 and first data sequence 2, respectively.
  • the first data sequence 1 corresponds to the first board, and the first data sequence 1 includes multiple route numbers obtained by the first board, and the multiple route numbers include 1, 3, 4, and 5.
  • the first data sequence 2 corresponds to the second single board, and the first data sequence 2 includes multiple route numbers obtained by the second single board, and the multiple route numbers include 2, 3, 4, and 5.
  • Step 602 The router sends a first message to the first device, where the first message includes a second data set, and the second data set includes a first data sequence 1 and a first data sequence 2.
  • the first message further includes attribute information 1 corresponding to the first data sequence 1 and attribute information 2 corresponding to the first data sequence 2 .
  • the attribute information 1 includes the data type corresponding to the first data sequence 1 and/or the object identifier of the object, the data type is "number of routes", and the object identifier of the object is the identifier "ID-ob1" of the first board.
  • the attribute information 2 includes the data type corresponding to the first data sequence 2 and/or the object identifier of the object, the data type is also "number of routes", and the object identifier of the object is the identifier "ID-ob2" of the second board .
  • Step 603 The first device receives the first message, and saves the first data sequence 1 and the first data sequence 2 included in the second data set into the second correspondence.
  • the first device locally stores the second correspondence.
  • Each record in the second correspondence includes network element information of the network element and a first data set, the first data set includes data sent by the network element, and the network element information includes the identification and name of the network element , address and/or type.
  • the first data set may include at least one second data sequence.
  • the second data sequence corresponds to an object and a data type in the network element, and the second data sequence includes data belonging to the data type acquired by the one object .
  • the record may also include attribute information corresponding to each second data sequence.
  • the first record in the second correspondence includes the identifier of the router, the first data set, and attribute information.
  • the identifier of the router is "ID-NE1”
  • the first data set includes the second data sequence 1 and the second data sequence 2
  • the second data sequence 1 includes the number of routes 1, 3, and 4 obtained by the first board in the router.
  • the second data sequence 2 includes the number of routes 2, 3, 4 and 5 that have been acquired by the second board in the router.
  • the attribute information includes attribute information 1 of the second data sequence 1 and attribute information 2 of the second data sequence 2; the attribute information 1 includes the object identifier and data type, and the object identifier is the identifier "ID-ob1" of the first board, and the data type is "number of routes"; attribute information 2 also includes the object identifier and data type, the object identifier is the identifier "ID-ob2" of the second board, and the data type is "number of routes”.
  • the second data set in the first message includes a first data sequence 1 (including data 1, 3, 4, and 5) and a second data sequence 2 ( Including data 2, 3, 4, and 5), the first message also includes attribute information 1 of the first data sequence 1 and attribute information 2 of the first data sequence 2.
  • the attribute information 1 includes the object identifier and data type of the object corresponding to the first data sequence 1, the object identifier is the identifier "ID-ob1" of the first board, and the data type is "number of routes”.
  • the attribute information 2 includes the object identifier and data type of the object corresponding to the first data sequence 2, the object identifier is the identifier "ID-ob2" of the second board, and the data type is "number of routes”.
  • the first device acquires the identifier "ID-NE1" of the router, and based on the "ID-NE1", acquires the first record including "ID-NE1" from the second correspondence shown in Table 3.
  • the first device obtains the second data sequence from the first record based on the identifier "ID-ob1" and "route number" of the first board 1 (including data 1, 3, 4, 5).
  • the first data sequence 1 and the second data sequence 1 are spliced into one data sequence, and the spliced data sequence includes data 1, 3, 4, 5, 1, 3, 4 and 5. Referring to Table 4 below, the second data sequence 1 in the first record is updated to the spliced data sequence, so as to save the first data sequence 1 into the second correspondence.
  • the first device obtains the second data sequence from the first record based on the identifier "ID-ob2" and the "number of routes" of the second board 2 (including data 2, 3, 4, 5).
  • the first data sequence 2 and the second data sequence 2 are spliced into one data sequence, and the spliced data sequence includes data 2 , 3 , 4 , 5 , 2 , 3 , 4 and 5 .
  • the second data sequence 2 in the first record is updated to the spliced data sequence, so as to save the first data sequence 2 into the second correspondence.
  • Step 604 The first device obtains a first data set from the second correspondence, where the first data set includes at least one second data set sent by the router in the second period.
  • the first device obtains the first data set corresponding to the router from the first record in the second correspondence shown in Table 4.
  • the first data set includes the second data sequence 1 and the second data sequence 2.
  • the second data set Sequence 1 includes data 1, 3, 4, 5, 1, 3, 4, and 5, and second data sequence 2 includes data 2, 3, 4, 5, 2, 3, 4, and 5.
  • the first device also obtains an object identifier and at least one data type of at least one object from the first record, where the object identifier of the at least one object includes the identifier "ID-ob1" of the first board and the identifier of the second board "ID-ob2", the at least one data type includes the number of routes.
  • Step 605 The first device acquires, based on the first data set, at least one data feature corresponding to the data in the first data set.
  • the first data set includes a second data series 1 and a second data series 2, the second data series 1 includes data 1, 3, 4, 5, 1, 3, 4 and 5, and the second data series 2 includes data 2, 3 , 4, 5, 2, 3, 4 and 5.
  • the second data sequence 1 including data 1, 3, 4, 5, 1, 3, 4 and 5 is processed to obtain a data waveform characteristic of a periodic type.
  • the second data sequence 2 including data 2, 3, 4, 5, 2, 3, 4 and 5 is processed to obtain a data waveform characteristic of a periodic type. Accordingly, the at least one data feature obtained by processing the first data set includes a periodic type.
  • the first device further acquires at least one feature type based on the periodic type, the data type "number of routes" to which the first data set belongs, and the first correspondence shown in Table 2, the at least one feature type.
  • Feature types include mean.
  • Step 606 The first device sends trigger information to the router, where the trigger information includes the data characteristic "period type” and/or the characteristic type "average value”.
  • the trigger information further includes the identifier "ID-ob1" of the first board, the identifier "ID-ob2" of the second board, and/or the "number of routes”.
  • Step 607 The router receives the trigger information, obtains at least one feature type based on the trigger information, and obtains a data indicator based on the at least one feature type and the second data set.
  • the router stores a third corresponding relationship, and the third corresponding relationship includes the corresponding relationship between the average value and the data processing algorithm.
  • the data math algorithm includes an average threshold.
  • the router receives trigger information, and obtains at least one feature type based on the trigger information, where the at least one feature type includes an average value. Based on the average value and the third correspondence, the data processing algorithm is obtained. A data indicator is obtained based on the at least one feature type and the second data set, where the data indicator is the average number of routes in the second data set. Compare the average number of routes and the average threshold, if the comparison result is that the average number of routes is greater than the average threshold, it is determined that the average number of routes is abnormal and the number of routes in the second data set is abnormal; if the comparison result is that the average number of routes is average. If the value is less than or equal to the average value threshold, it is determined that the average number of routes is normal and the number of routes in the second data set is normal.
  • the first device receives the second data set sent by the router and saves the second data set.
  • the first device When receiving the second data set within a period of time, that is, when obtaining the first data set, the first device The data set is processed to obtain at least one data feature, and trigger information is sent to the router, where the trigger information includes the at least one data feature and at least one feature type.
  • the at least one feature type is acquired based on the trigger information, a data indicator is determined based on the at least one feature type and the second data set, and an abnormal situation of the second data set is analyzed based on the data indicator. Since the at least one feature type is acquired based on the first data set, the at least one feature type corresponds to the data behavior in the first data set.
  • the embodiments of the present application reduces the difficulty of analyzing the data in the second data set, thereby accurately analyzing the abnormal situation of the second data set based on the data indicators, and improving the accuracy of the data analysis.
  • an embodiment of the present application provides an apparatus 700 for processing data.
  • the apparatus 700 may be deployed on the first device provided in any of the foregoing embodiments, for example, deployed on the first device in the network architecture 100 shown in FIG. 1 .
  • the apparatus 700 includes:
  • a receiving unit 701 configured to receive a first data set sent by a network element, where the first data set includes a plurality of first data acquired by the network element;
  • a processing unit 702 configured to acquire at least one data feature corresponding to the plurality of first data based on the first data set;
  • a sending unit 703, configured to send trigger information to the network element, where the trigger information includes the at least one data feature and/or at least one feature type, and the at least one feature type is related to the at least one data feature.
  • the network element receives the trigger information, acquires the at least one feature type based on the trigger information, acquires a data indicator based on the at least one feature type and a second data set, and the second data The set includes a plurality of second data acquired by the network element.
  • steps 201-203 of the method 200 shown in FIG. 2 steps 503-504 of the method 500 shown in FIG. 5, and shown in FIG. 6.
  • steps 603-604 of the method 600 will not be described in detail here.
  • step 204 of the method 200 shown in FIG. 2 For the detailed implementation process of acquiring the at least one data feature by the processing unit 702, see step 204 of the method 200 shown in FIG. 2 , step 505 of the method 500 shown in FIG. 5 , and step 605 of the method 600 shown in FIG. 6 . The content will not be described in detail here.
  • processing unit 702 is further configured to:
  • the at least one feature type is acquired based on the at least one data feature and a first corresponding relationship, where the first corresponding relationship includes the at least one data feature and the at least one feature type.
  • the processing unit 702 is configured to:
  • the at least one feature type based on the at least one data feature, the first correspondence, and at least one data type, where the at least one data type includes the type to which the first data in the first data set belongs, and the The first correspondence includes the at least one data feature, the at least one data type, and/or the at least one feature type.
  • the processing unit 702 obtains the detailed implementation process of the at least one feature type, see step 204 of the method 200 shown in FIG. 2 , step 505 of the method 500 shown in FIG. 5 , and method 600 shown in FIG. 6 .
  • the relevant content in step 605 of will not be described in detail here.
  • the trigger information includes the at least one data characteristic and/or at least one data type, and the at least one data type includes a type to which the first data in the first data set belongs.
  • the network element acquires the at least one feature type based on the at least one data feature and/or the at least one data type.
  • the trigger information further includes an object identifier of at least one first object, and the at least one first object is located in the network element.
  • the network element acquires the second data set based on the object identifier of the at least one first object, and the second data set includes second data acquired by the at least one first object.
  • step 205 of the method 200 shown in FIG. 2 please refer to step 403 of the method 400 shown in FIG. 4 , step 506 of the method 500 shown in FIG. 5 , and the method shown in FIG. 6 .
  • the related content in step 606 of 600 will not be described in detail here.
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set, and at least one first data set.
  • the apparatus 700 includes a data processing system, a controller, or a management device.
  • the first data set includes the second data set or part of the data in the second data set, or the second data set includes the network element sending the first data data acquired after the set.
  • the first data set received by the receiving unit includes a plurality of first data acquired by the network element.
  • the processing unit acquires at least one data feature corresponding to the plurality of first data based on the first data set, and the at least one data feature corresponds to the data behavior of the first data in the first data set.
  • the sending unit sends trigger information to the network element, where the trigger information includes the at least one data feature and/or at least one feature type.
  • the network element acquires the at least one feature type based on the trigger information, wherein, since the at least one data feature corresponds to the data behavior of the first data in the first data set, the at least one feature type is obtained based on the at least one data feature.
  • the network element Even if the data collected by the network element changes due to changes in the network environment, through the embodiment of the present application, it can be analyzed that the data behavior in the first data set has changed, and at least one corresponding data feature of the data in the first data set can be obtained by obtaining at least one Feature type.
  • the network element can accurately reflect the network situation based on the at least one feature type and the data index obtained from the second data set. Therefore, the network element reduces the difficulty of analyzing the data in the second data set through the embodiment of the present application, which can improve the accuracy of the acquired data indicators and the accuracy of analyzing the data in the second data set.
  • an embodiment of the present application provides an apparatus 800 for processing data, and the apparatus 800 may be deployed on the network element provided in any of the foregoing embodiments, for example, the network element 102 deployed in the network architecture 100 shown in FIG. 1 . , or the network element in the method 400 shown in FIG. 4 , or the network element in the method 500 shown in FIG. 5 , or the router in the method 600 shown in FIG. 6 .
  • the apparatus 800 includes:
  • a processing unit 801 configured to acquire a first data set, where the first data set includes a plurality of acquired first data
  • the processing unit 801 is further configured to acquire at least one feature type based on the first data set, where the at least one feature type corresponds to at least one data feature corresponding to the plurality of first data;
  • the processing unit 801 is further configured to acquire data indicators based on the at least one feature type and a second data set, where the second data set includes a plurality of acquired second data.
  • the processing unit 801 acquires the at least one feature type and acquires the detailed implementation process of the data indicator, see step 404 of the method 400 shown in FIG. 4 , step 507 of the method 500 shown in FIG. 5 , and FIG. 6 The relevant content in step 607 of the illustrated method 600 will not be described in detail here.
  • the apparatus 800 further includes: a sending unit 802 and a receiving unit 803;
  • the sending unit 802 is configured to send the first data set, where the first data set is used for the receiver of the first data set to acquire the at least one data feature and/or the at least one feature type;
  • the receiving unit 803 is configured to receive trigger information, where the trigger information includes the at least one data feature and/or the at least one feature type;
  • the processing unit 801 is configured to acquire the at least one feature type based on the trigger information.
  • step 402 of the method 400 shown in FIG. 4 for the detailed implementation process of sending the first data set by the sending unit 802, see step 402 of the method 400 shown in FIG. 4 , step 502 of the method 500 shown in FIG. 5 , and method 600 shown in FIG. 6 .
  • the relevant content in step 602 of will not be described in detail here.
  • step 403 of the method 400 shown in FIG. 4 for the detailed implementation process of receiving the trigger information by the receiving unit 803, refer to step 403 of the method 400 shown in FIG. 4 , step 507 of the method 500 shown in FIG. 5 , and steps of the method 600 shown in FIG. 6 .
  • the relevant content in 607 will not be described in detail here.
  • the trigger information includes the at least one data feature
  • the processing unit 801 is configured to:
  • the at least one feature type is acquired based on the at least one data feature and a first corresponding relationship, where the first corresponding relationship includes the at least one data feature and the at least one feature type.
  • the trigger information further includes at least one data type, and the at least one data type includes a type to which the first data in the first data set belongs;
  • the processing unit 801 is used for:
  • the at least one feature type based on the at least one data feature, the at least one data type, and the first correspondence, where the first correspondence includes the at least one data feature, the at least one data type and the at least one feature type.
  • the processing unit 801 obtains the detailed implementation process of the at least one feature type based on the at least one data type, see step 404 of the method 400 shown in FIG. 4 and step 507 of the method 500 shown in FIG. 5 , And the related content in step 607 of the method 600 shown in FIG. 6 will not be described in detail here.
  • the trigger information further includes an object identifier of at least one first object
  • the processing unit 801 is further configured to:
  • the second data set is obtained based on the object identification of the at least one first object, the second data set includes a plurality of second data obtained by the at least one first object.
  • step 205 of the method 200 shown in FIG. 2 please refer to step 403 of the method 400 shown in FIG. 4 , step 506 of the method 500 shown in FIG. 5 , and the method shown in FIG. 6 .
  • the related content in step 606 of 600 will not be described in detail here.
  • the processing unit 801 is configured to:
  • the at least one feature type is acquired based on the at least one data feature and a first corresponding relationship, where the first corresponding relationship includes the at least one data feature and the at least one feature type.
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set, and at least one first data set.
  • processing unit 801 is further configured to:
  • step 404 of the method 400 shown in FIG. 4 step 507 of the method 500 shown in FIG. 5
  • step 607 of the method 600 shown in FIG. 6 step 607 of the method 600 shown in FIG. 6 .
  • the related content will not be described in detail here.
  • the first data set includes the second data set or part of the data in the second data set, or the second data set includes the network element acquiring the first data data acquired after the set.
  • the processing unit acquires a first data set, where the first data set includes a plurality of acquired first data. At least one feature type is acquired based on the first data set, and the at least one feature type corresponds to at least one data feature corresponding to the plurality of data. Since the at least one data feature corresponds to the data behavior of the first data in the first data set, and the at least one feature type corresponds to the at least one data feature, the at least one feature type is based on the at least one feature type. A data feature is obtained.
  • the processing unit can analyze that the data behavior in the first data set has changed, and obtain the corresponding feature type through at least one data feature of the data in the first data set, and the processing unit The data indicators obtained based on the at least one feature type and the second data set can accurately reflect the network situation. Therefore, the processing unit reduces the difficulty of analyzing the data in the second data set through the embodiment of the present application, and improves the accuracy of analyzing the data in the second data set.
  • the present application provides a system 900 for processing data
  • the system 900 includes:
  • a data acquisition unit 901 configured to acquire a first data set, where the first data set includes a plurality of first data
  • a data sending unit 902 configured to send the first data set
  • a data processing unit 903 configured to acquire at least one data feature corresponding to the plurality of first data based on the first data set;
  • An information sending unit 904 configured to send trigger information, where the trigger information includes the at least one data feature and/or at least one feature type, the at least one feature type is the data processing unit 903 based on the at least one data feature obtained;
  • a type acquiring unit 905, configured to acquire the at least one feature type based on the trigger information
  • An indicator acquiring unit 906, configured to acquire data indicators based on the at least one feature type and a second data set, where the second data set includes a plurality of second data.
  • the plurality of second data included in the second data set are acquired by the data acquisition unit 901 or acquired by the indicator acquisition unit 906 .
  • some or all of the data acquisition unit 901 , data transmission unit 902 , data processing unit 903 , information transmission unit 904 , type acquisition unit 905 and indicator acquisition unit 906 are deployed on the first device and / or on the network element.
  • the data acquisition unit 901, the data transmission unit 902, the type acquisition unit 905 and the indicator acquisition unit 906 are deployed on the network element, and the data processing unit 903 and the information transmission unit 904 are deployed on the first device.
  • the data acquisition unit 901 , the data transmission unit 902 , the data processing unit 903 , the information transmission unit 904 , the type acquisition unit 905 and/or the indicator acquisition unit 906 are deployed on the network element.
  • the data acquisition unit 901 , the data transmission unit 902 , the data processing unit 903 , the information transmission unit 904 , the type acquisition unit 905 and/or the indicator acquisition unit 906 are deployed on the first device.
  • the first device may be the first device provided in any of the foregoing embodiments, for example, may be the first device 101 in the network architecture 100 shown in FIG. 1 , or the first device in the method 200 shown in FIG. 2 . device, or the first device in the method 500 shown in FIG. 5 , or the first device in the method 600 shown in FIG. 6 , or the apparatus 700 shown in FIG. 7 .
  • the network element may be the network element provided in any of the foregoing embodiments, for example, may be the network element 102 in the network architecture 100 shown in FIG. 1 , or the network element in the method 400 shown in FIG. 4 , or The network element in the method 500 shown in FIG. 5 or the router in the method 600 shown in FIG. 6 , or the apparatus 800 shown in FIG. 8 .
  • step 401 of the method 400 shown in FIG. Step 501 or step 504, and the related content in step 601 or step 604 of the method 600 shown in FIG. 6 will not be described in detail here.
  • step 204 of the method 200 shown in FIG. 2 step 404 of the method 400 shown in FIG. Step 505 or step 507, and the related content in step 605 or step 607 of the method 600 shown in FIG. 6 will not be described in detail here.
  • the type obtaining unit 905 obtains the detailed implementation process of the at least one feature type, see step 204 of the method 200 shown in FIG. 2, step 404 of the method 400 shown in FIG. Step 505 or step 507, and the related content in step 605 or step 607 of the method 600 shown in FIG. 6 will not be described in detail here.
  • step 404 of the method 400 shown in FIG. 4 for the detailed implementation process of acquiring data indicators by the indicator acquiring unit 906, refer to step 404 of the method 400 shown in FIG. 4 , step 507 of the method 500 shown in FIG. 5 , and step 607 of the method 600 shown in FIG. 6 .
  • step 404 of the method 400 shown in FIG. 4 for the detailed implementation process of acquiring data indicators by the indicator acquiring unit 906, refer to step 404 of the method 400 shown in FIG. 4 , step 507 of the method 500 shown in FIG. 5 , and step 607 of the method 600 shown in FIG. 6 .
  • the related content will not be described in detail here.
  • the data processing unit 903 is further configured to acquire the at least one feature type based on the at least one data feature and a first correspondence, where the first correspondence includes the at least one data feature and the at least one feature type.
  • the data processing unit 903 is configured to obtain the at least one feature type based on the at least one data feature, the first correspondence and at least one data type, where the at least one data type includes the the type to which the first data in the first data set belongs, and the first correspondence includes the at least one data feature, the at least one data type, and the at least one feature type.
  • the trigger information includes the at least one data characteristic
  • the type obtaining unit 905 is configured to obtain the at least one feature type based on the at least one data feature and a first corresponding relationship, where the first corresponding relationship includes the at least one data feature and the at least one feature type .
  • the trigger information further includes at least one data type, and the at least one data type includes a type to which the first data in the first data set belongs,
  • the type obtaining unit 905 is configured to obtain the at least one feature type based on the at least one data feature, the at least one data type, and the first correspondence, where the first correspondence includes the at least one A data feature, the at least one data type, and the at least one feature type.
  • step 205 of the method 200 shown in FIG. 2 please refer to step 403 of the method 400 shown in FIG. 4 , step 506 of the method 500 shown in FIG. 5 , and the method shown in FIG. 6 .
  • the related content in step 606 of 600 will not be described in detail here.
  • the trigger information further includes an object identifier of at least one first object
  • the second data set includes a plurality of second data acquired by the at least one first object
  • the at least one data feature includes one or more of the following information: a data waveform feature corresponding to the first data in the first data set, a data content feature in the first data set, and at least one second The object identifier of the object; wherein the at least one second object includes an object for obtaining the first data in the first data set.
  • the indicator obtaining unit 906 is further configured to determine an abnormal situation of the second data in the second data set based on the data indicator;
  • the data sending unit 902 is further configured to send the determined abnormal situation and/or the data indicator.
  • the detailed implementation process of determining the abnormal situation by the indicator obtaining unit 906 may refer to step 404 of the method 400 shown in FIG. 4 , step 507 of the method 500 shown in FIG. 5 , and step 607 of the method 600 shown in FIG. 6 .
  • the relevant content in will not be described in detail here.
  • the first dataset includes the second dataset or a portion of the data in the second dataset, or the second dataset includes data acquired after acquiring the first dataset .
  • the data acquisition unit acquires a first data set, where the first data set includes a plurality of acquired first data.
  • the data processing unit acquires at least one data feature corresponding to the plurality of first data based on the first data set, so that the at least one data feature corresponds to the data behavior of the first data in the first data set, and At least one feature type is acquired by the data processing unit based on the at least one data feature.
  • the type acquiring unit acquires the corresponding feature through at least one data feature of the data in the first data set type, so that the data indicator acquired by the indicator acquiring unit based on the at least one feature type and the second data set can accurately reflect the network situation.
  • the index obtaining unit reduces the difficulty of analyzing the data in the second data set through the above system, and improves the accuracy of analyzing the data in the second data set.
  • various modules in the system 900 for processing data may be deployed in the same physical device; in other embodiments, various modules in the system 900 for processing data may be deployed in multiple different physical devices .
  • Each module in the system 900 for processing data may be a hardware module or a module combining software and hardware.
  • an embodiment of the present application provides a schematic diagram of a device 1000 for processing data.
  • the device 1000 may be the first device provided in any of the foregoing embodiments, for example, may be the first device 101 in the network architecture shown in FIG. 1 , or the first device in the method 200 shown in FIG.
  • the first device in the method 500 shown in FIG. 6 or the first device in the method 600 shown in FIG. 6 is shown.
  • the device 1000 includes at least one processor 1001 , internal connections 1002 , memory 1003 and at least one network interface 1004 .
  • the device 1000 is a hardware-structured device.
  • the processing unit 702 in the apparatus 700 shown in FIG. 7 can be implemented by calling the code in the memory 1003 by the at least one processor 1001, and the receiving unit 701 and the receiving unit 701 in the apparatus 700 shown in FIG.
  • the sending unit 703 may be implemented through the at least one network interface 1004 . or,
  • memory 1003 is used to store program modules and data.
  • the program modules include a receiving module 10031 , a processing module 10032 and a sending module 10033 .
  • each module in the memory 1003 in FIG. 10 corresponds to each module shown in FIG. 7
  • the processor 1001 can execute the computer-readable instructions in each module in the memory 1003 by executing the computer-readable instructions in FIG. 7 . operations that can be performed by each of the modules shown.
  • the device 1000 may also be used to implement the function of the first device in any of the foregoing embodiments.
  • the above-mentioned processor 1001 is, for example, a general-purpose central processing unit (Central Processing Unit, CPU), a digital signal processor (Digital Signal Processor, DSP), a network processor (Network Processor, NP), a graphics processor (Graphics Processing Unit, GPU) , a neural network processor (Neural-network Processing Units, NPU), a data processing unit (Data Processing Unit, DPU), a microprocessor or one or more integrated circuits for implementing the solution of the present application.
  • the processor 701 includes an application-specific integrated circuit (ASIC), a programmable logic device (Programmable Logic Device, PLD) or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.
  • ASIC application-specific integrated circuit
  • PLD programmable Logic Device
  • the PLD is, for example, a Complex Programmable Logic Device (CPLD), a Field-Programmable Gate Array (FPGA), a Generic Array Logic (GAL) or any combination thereof. It can implement or execute various logical blocks, modules and circuits described in connection with the disclosure of the embodiments of the present application.
  • the processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.
  • the internal connection 1002 described above may include a path to transfer information between the above described components.
  • the internal connection 1002 can be a single board or a bus or the like.
  • the bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus or the like.
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into address bus, data bus, control bus and so on. For ease of presentation, only one thick line is used in FIG. 10, but it does not mean that there is only one bus or one type of bus.
  • the above-mentioned at least one network interface 1004 uses any device such as a transceiver for communicating with other devices or a communication network, and the communication network can be an Ethernet, a wireless access network, or a wireless local area network (Wireless Local Area Networks, WLAN) and the like.
  • the network interface 1004 may include a wired communication interface and may also include a wireless communication interface.
  • the network interface 1004 may be an Ethernet interface, a Fast Ethernet (FE) interface, a Gigabit Ethernet (GE) interface, an Asynchronous Transfer Mode (ATM) interface, a wireless local area network (WLAN) interface, a cellular A network communication interface or a combination thereof.
  • the Ethernet interface can be an optical interface, an electrical interface or a combination thereof.
  • the network interface 1004 may be used for the device 1000 to communicate with other devices.
  • the above-mentioned memory 1003 can be a read-only memory (read-only memory, ROM) or other types of static storage devices that can store static information and instructions, a random access memory (random access memory, RAM) or other types of storage devices that can store information and instructions.
  • ROM read-only memory
  • RAM random access memory
  • Types of dynamic storage devices which can also be electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), or other optical disk storage, optical disks storage (including compact discs, laser discs, compact discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by Any other medium accessed by the computer, but not limited to this.
  • the memory can exist independently and be connected to the processor through a bus.
  • the memory 1003 may also be integrated with the processor 1001 .
  • the processor 1001 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 10 . Each of these CPUs can be a single-core processor or a multi-core processor.
  • a processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • the device 1000 may include multiple processors, such as the processor 1001 and the processor 1007 in FIG. 10 .
  • processors can be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor.
  • a processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • the device 1000 may further include an output device and an input device.
  • the output device communicates with the processor 1001 and can display information in a variety of ways.
  • the output device may be a liquid crystal display (Liquid Crystal Display, LCD), a light emitting diode (Light Emitting Diode, LED) display device, a cathode ray tube (Cathode Ray Tube, CRT) display device, or a projector, and the like.
  • the input device communicates with the processor 1001 and can receive user input in a variety of ways.
  • the input device may be a mouse, a keyboard, a touch screen device, or a sensor device, or the like.
  • the device 1000 in this embodiment of the present application may correspond to the above-mentioned multiple embodiments, for example, the first device in the multiple embodiments corresponding to FIG. 1 , FIG. 2 , FIG. 5 , and FIG. 6 , so
  • the processor 1001 in the device 1000 reads the instructions in the memory 1003, so that the device 1000 shown in FIG. 10 can perform all or part of the operations of the first device in the above-mentioned multiple embodiments.
  • an embodiment of the present application provides a schematic diagram of a device 1100 for processing data.
  • the device 1100 may be a network element provided in any of the foregoing embodiments, for example, may be the network element 102 in the network architecture shown in FIG. 1 , or the network element in the method 400 shown in FIG. 4 , or the method 500 shown in FIG. 5 .
  • the device 1100 includes at least one processor 1101 , internal connections 1102 , memory 1103 and at least one network interface 1104 .
  • the device 1100 is a hardware-structured device.
  • the processing unit 801 in the apparatus 800 shown in FIG. 8 can be implemented by calling the code in the memory 1103 by the at least one processor 1101, and the sending unit 802 and the sending unit 802 in the apparatus 800 shown in FIG.
  • the receiving unit 803 may be implemented through the at least one network interface 1104 . or,
  • memory 1103 is used to store program modules and data.
  • the program modules include a processing module 11031 , a sending module 11032 and a receiving module 11033 .
  • each module in the memory 1103 in FIG. 11 corresponds to each module shown in FIG. 8
  • the processor 1101 can execute the computer-readable instructions in each module in the memory 1103 by executing the computer-readable instructions in FIG. 8 . operations that can be performed by each of the modules shown.
  • the device 1100 may also be used to implement the functions of the network element in any of the foregoing embodiments.
  • the above-mentioned processor 1101 is, for example, a general-purpose central processing unit (Central Processing Unit, CPU), a digital signal processor (Digital Signal Processor, DSP), a network processor (Network Processor, NP), a graphics processor (Graphics Processing Unit, GPU) , a neural network processor (Neural-network Processing Units, NPU), a data processing unit (Data Processing Unit, DPU), a microprocessor or one or more integrated circuits for implementing the solution of the present application.
  • the processor 701 includes an application-specific integrated circuit (ASIC), a programmable logic device (Programmable Logic Device, PLD) or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.
  • ASIC application-specific integrated circuit
  • PLD programmable Logic Device
  • the PLD is, for example, a Complex Programmable Logic Device (CPLD), a Field-Programmable Gate Array (FPGA), a Generic Array Logic (GAL) or any combination thereof. It can implement or execute various logical blocks, modules and circuits described in connection with the disclosure of the embodiments of the present application.
  • the processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.
  • the aforementioned internal connection 1102 may include a path for transferring information between the aforementioned components.
  • the internal connection 1102 may be a single board or a bus or the like.
  • the bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus or the like.
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into address bus, data bus, control bus and so on. For ease of presentation, only one thick line is used in FIG. 11, but it does not mean that there is only one bus or one type of bus.
  • the above-mentioned at least one network interface 1104 uses any device such as a transceiver for communicating with other devices or a communication network, and the communication network can be an Ethernet, a wireless access network, or a wireless local area network (Wireless Local Area Networks, WLAN) and the like.
  • the network interface 1104 may include a wired communication interface and may also include a wireless communication interface.
  • the network interface 1104 may be an Ethernet interface, a Fast Ethernet (FE) interface, a Gigabit Ethernet (GE) interface, an Asynchronous Transfer Mode (ATM) interface, a wireless local area network (WLAN) interface, a cellular A network communication interface or a combination thereof.
  • the Ethernet interface can be an optical interface, an electrical interface or a combination thereof.
  • the network interface 1104 may be used for the device 1100 to communicate with other devices.
  • the above-mentioned memory 1103 can be a read-only memory (read-only memory, ROM) or other types of static storage devices that can store static information and instructions, a random access memory (random access memory, RAM) or other types of storage devices that can store information and instructions.
  • ROM read-only memory
  • RAM random access memory
  • Types of dynamic storage devices which can also be electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), or other optical storage, CD-ROM storage (including compact discs, laser discs, compact discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by Any other medium accessed by the computer, but not limited to this.
  • the memory can exist independently and be connected to the processor through a bus.
  • the memory 1103 may also be integrated with the processor 1101 .
  • the processor 1101 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 11 . Each of these CPUs can be a single-core processor or a multi-core processor.
  • a processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • the device 1100 may include multiple processors, such as the processor 1101 and the processor 1107 in FIG. 11 .
  • processors can be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor.
  • a processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • the device 1100 may further include an output device and an input device.
  • the output device communicates with the processor 1101 and can display information in a variety of ways.
  • the output device may be a liquid crystal display (Liquid Crystal Display, LCD), a light emitting diode (Light Emitting Diode, LED) display device, a cathode ray tube (Cathode Ray Tube, CRT) display device, or a projector, and the like.
  • the input device communicates with the processor 1101 and can receive user input in a variety of ways.
  • the input device may be a mouse, a keyboard, a touch screen device, or a sensor device, or the like.
  • the device 1100 in this embodiment of the present application may correspond to the above-mentioned multiple embodiments, for example, the network element in the multiple embodiments corresponding to FIG. 1 , FIG. 4 , FIG. 5 , and FIG. 6 .
  • the processor 1101 in the device 1100 reads the instructions in the memory 1103, so that the device 1100 shown in FIG. 11 can perform all or part of the operations of the network elements in the above multiple embodiments.

Landscapes

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

Abstract

本申请公开一种处理数据的方法、装置、系统及存储介质，属于通信领域。所述方法包括：接收网元发送的第一数据集，所述第一数据集包括所述网元获取的多个第一数据；基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征；向所述网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型与所述至少一个数据特征有关。本申请能够提高对数据进行分析的准确度。

Description

处理数据的方法、装置、系统及存储介质
本申请要求于2021年1月28日提交的申请号为202110119289.9、发明名称为“处理数据的方法、装置、系统及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信领域，特别涉及一种处理数据的方法、装置、系统及存储介质。
背景技术
网元可以采集与业务和/或网络相关的关键性能指标(key performance indicator，KPI)对应的多个数据。通常网元可以周期性地进行采集，以得到该KPI对应的多个数据，分析该KPI对应的多个数据是否异常，基于分析的结果确定业务或网络是否存在风险。
网元所在的网络环境是不断变化的，在网络环境变化后，网元采集的数据也随之在变化，这样给网元分析采集的数据的异常带来难度，降低对采集的数据分析的准确度。
发明内容
本申请提供了一种处理数据的方法、装置、系统及存储介质，以提高对数据分析的准确度。所述技术方案如下：
第一方面，本申请提供了一种处理数据的方法，在所述方法中：接收网元发送的第一数据集，所述第一数据集包括所述网元获取的多个第一数据。基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征。向所述网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型与所述至少一个数据特征有关。
由于基于所述第一数据集获取所述至少一个数据特征，使得所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应，而至少一个特征类型是基于所述至少一个数据特征获取的。即使因为网络环境变化导致网元采集的数据变化，通过上述方法，可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的特征类型，这样网元在接收所述触发信息时，基于所述触发信息获取至少一个特征类型，基于所述至少一个特征类型和第二数据集获取的数据指标就可以准确的反映网络情况。所以通过上述方法降低了分析第二数据集中的数据的难度，提高对第二数据集中的数据分析的准确度。
在一种可能的实现方式中，基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，由于所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型，基于第一对应关系能够快速地获取所述至少一个特征类型，简化实现复杂度，从而能够及时向网元发送所述触发信息。
在另一种可能的实现方式中，基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和/或所述至少一个特征类型。这样使得获取的所述至少一个特征类型与所述至少一个数据类型相关联，从而能够提高获取特征类型的准确度。
在另一种可能的实现方式中，所述触发信息包括所述至少一个数据特征和/或至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型。由于所述触发信息包括所述至少一个数据类型，这样使得网元基于所述至少一个数据类型获取所述至少一个特征类型，提高获取所述至少一个特征类型的准确度。
在另一种可能的实现方式中，所述触发信息还包括至少一个第一对象的对象标识，所述至少一个第一对象位于所述网元中。这样网元可以基于所述至少一个第一对象的对象标识获取第二数据集，且第二数据集包括所述至少一个第一对象获取的第二数据。由于第二数据集中数据是所述至少一个第一对象获取的，所述至少一个第一对象与所述第一数据集可能关联，这样可以提高对第二数据集中的数据进行分析的准确度。
在另一种可能的实现方式中，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；其中，所述至少一个第二对象包括获取所述第一数据集中的各第一数据的对象。
在另一种可能的实现方式中，所述方法的执行主体包括：数据处理系统、控制器或管理设备。
在另一种可能的实现方式中，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括所述网元在发送所述第一数据集之后获取的数据。
第二方面，本申请提供了一种处理数据的方法，在所述方法中，获取第一数据集，所述第一数据集包括多个第一数据。基于所述第一数据集获取至少一个特征类型，所述至少一个特征类型与所述多个第一数据对应的至少一个数据特征相对应。基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括多个第二数据。
由于基于所述第一数据集获取所述至少一个数据特征，使得所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应，而至少一个特征类型是基于所述至少一个数据特征获取的。即使因为网络环境变化导致获取的数据变化，通过上述方法，可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的特征类型，基于所述至少一个特征类型和第二数据集获取的数据指标就可以准确地反映网络情况。所以通过上述方法降低了分析第二数据集中的数据的难度，提高对第二数据集中的 数据进行分析的准确度。
在一种可能的实现方式中，发送所述第一数据集，所述第一数据集用于所述第一数据集的接收方获取所述至少一个数据特征和/或所述至少一个特征类型。接收触发信息，所述触发信息包括所述至少一个数据特征和/或所述至少一个特征类型。基于所述触发信息获取所述至少一个特征类型。这样网元可以不用基于第一数据集获取所述至少一个数据特征和/或所述至少一个特征类型，这样在网元的计算能力和存储能力较低的情况下，也不影响网元得到所述至少一个特征类型。
在另一种可能的实现方式中，所述触发信息包括所述至少一个数据特征，基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。由于所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型，基于第一对应关系能够快速地获取所述至少一个特征类型，简化实现复杂度，从而能够快速获取所述至少一个特征类型。
在另一种可能的实现方式中，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型。基于所述至少一个数据特征、所述至少一个数据类型和所述第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。由于所述触发信息还包括所述至少一个数据类型，这样使得获取的所述至少一个特征类型与所述至少一个数据类型相关联，从而能够提高获取特征类型的准确度。
在另一种可能的实现方式中，所述触发信息还包括至少一个第一对象的对象标识。基于所述至少一个第一对象的对象标识获取所述第二数据集，所述第二数据集包括所述至少一个第一对象获取的多个第二数据。由于第二数据集中数据是所述至少一个第一对象获取的，所述至少一个第一对象与所述第一数据集可能关联，这样可以提高对第二数据集中的数据进行分析的准确度。
在另一种可能的实现方式中，基于所述第一数据集获取所述至少一个数据特征。基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。这样由网元直接获取所述至少一个特征类型，不用向第一设备发送第一数据集，从而可以节省网络资源。
在另一种可能的实现方式中，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
在另一种可能的实现方式中，基于所述数据指标确定所述第二数据集中的第二数据的异 常情况。发送所述确定的异常情况和/或所述数据指标。
在另一种可能的实现方式中，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括所述网元在获取所述第一数据集之后获取的数据。
第三方面，本申请提供了一种处理数据的装置，用于执行第一方面或第一方面的任意一种可能的实现方式中的方法。具体地，所述装置包括用于执行第一方面或第一方面的任意一种可能的实现方式中的方法的单元。
第四方面，本申请提供了一种处理数据的装置，用于执行第二方面或第二方面的任意一种可能的实现方式中的方法。具体地，所述装置包括用于执行第二方面或第二方面的任意一种可能的实现方式中的方法的单元。
第五方面，本申请提供了一种处理数据的设备，所述设备包括处理器及计算机程序，所述处理器用于执行所述存储器中的计算机程序，使得所述设备完成第一方面或第一方面的任意可能的实现方式中的方法。
第六方面，本申请提供了一种处理数据的设备，所述设备包括处理器及计算机程序，所述处理器用于执行所述存储器中的计算机程序，使得所述设备完成第二方面或第二方面的任意可能的实现方式中的方法。
第七方面，本申请提供了一种计算机程序产品，所述计算机程序产品包括计算机程序，并且所述计算程序通过计算机进行加载来实现上述第一方面、第二方面、第一方面任意可能的实现方式或第二方面任意可能的实现方式的方法。
第八方面，本申请提供了一种计算机可读存储介质，用于存储计算机程序，所述计算机程序通过处理器进行加载来执行上述第一方面、第二方面、第一方面任意可能的实现方式或第二方面任意可能的实现方式的方法。
第九方面，本申请提供了一种芯片，包括存储器和处理器，存储器用于存储计算机指令，处理器用于从存储器中调用并运行该计算机指令，以执行上述第一方面、第二方面、第一方面任意可能的实现方式或第二方面任意可能的实现方式的方法。
第十方面，本申请提供了一种处理数据的系统，在所述系统中，数据获取单元获取第一数据集，所述第一数据集包括多个第一数据。数据发送单元发送所述第一数据集。数据处理单元基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征。信息发送单元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型是所述数据处理单元基于所述至少一个数据特征获取的。类型获取单元基于所述触发信息获取所述至少一个特征类型。指标获取单元基于所述至少一个特征类型和第二数据 集获取数据指标，所述第二数据集包括多个第二数据。
由于数据处理单元基于所述第一数据集获取所述至少一个数据特征，使得所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应，而至少一个特征类型是所述数据处理单元基于所述至少一个数据特征获取的。即使因为网络环境变化导致获取的数据变化，通过上述系统，可以分析出第一数据集中的数据行为发生了变化，所述类型获取单元通过第一数据集中数据的至少一个数据特征获取相对应的特征类型，这样指标获取单元基于所述至少一个特征类型和第二数据集获取的数据指标就可以准确地反映网络情况。通过上述系统降低了分析第二数据集中的数据的难度，提高对第二数据集中的数据进行分析的准确度。
在一种可能的实现方式中，所述数据处理单元还基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。由于所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型，所述数据处理单元基于第一对应关系能够快速地获取所述至少一个特征类型，简化实现复杂度，从而能够及时向网元发送所述触发信息。
在另一种可能的实现方式中，所述数据处理单元基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。这样使得获取的所述至少一个特征类型与所述至少一个数据类型相关联，从而能够提高获取特征类型的准确度。
在另一种可能的实现方式中，所述触发信息包括所述至少一个数据特征，所述类型获取单元基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
在另一种可能的实现方式中，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型。所述类型获取单元基于所述至少一个数据特征、所述至少一个数据类型和所述第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。由于所述触发信息包括所述至少一个数据类型，这样使得类型获取单元基于所述至少一个数据类型获取所述至少一个特征类型，提高获取所述至少一个特征类型的准确度。
在另一种可能的实现方式中，所述触发信息还包括至少一个第一对象的对象标识。所述第二数据集包括所述至少一个第一对象获取的多个第二数据。由于第二数据集中数据是所述至少一个第一对象获取的，所述至少一个第一对象与所述第一数据集可能关联，这样可以提高对第二数据集中的数据进行分析的准确度。
在另一种可能的实现方式中，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个 第二对象的对象标识；其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
在另一种可能的实现方式中，所述指标获取单元还基于所述数据指标确定所述第二数据集中的第二数据的异常情况。所述数据发送单元还发送所述确定的异常情况和/或所述数据指标。
在另一种可能的实现方式中，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括在获取所述第一数据集之后获取的数据。
附图说明
图1是本申请实施例提供的一种网络架构示意图；
图2是本申请实施例提供的一种处理数据的方法流程图；
图3是本申请实施例提供的一种配置界面的结构示意图；
图4是本申请实施例提供的另一种处理数据的方法流程图；
图5是本申请实施例提供的另一种处理数据的方法流程图；
图6是本申请实施例提供的另一种处理数据的方法流程图；
图7是本申请实施例提供的一种处理数据的装置结构示意图；
图8是本申请实施例提供的另一种处理数据的装置结构示意图；
图9是本申请实施例提供的一种处理数据的系统结构示意图；
图10是本申请实施例提供的一种处理数据的设备结构示意图；
图11是本申请实施例提供的另一种处理数据的装置结构示意图。
具体实施方式
下面将结合附图对本申请实施方式作进一步地详细描述。
参见图1，本申请实施列提供了一种网络架构100，包括：
第一设备101和至少一个网元102，第一设备101与所述至少一个网元102中的每个网元通信。
在一些实施例中，第一设备101和所述每个网元接入网络中，所述网络包括局域网或广域网。也就是说，第一设备101在局域网内与所述每个网元通信，或者，第一设备101可远程部署或云端部署。
在一些实施例中，对于所述至少一个网元102中的每个网元，第一设备101与所述网元建立网络连接，以实现与所述网元通信。当然，还有其他实现第一设备101与所述网元通信的方式，在此不再一一列举。
第一设备101为网络云化引擎(network cloud engine，NCE)、数据处理系统、控制器或管理设备等。所述至少一个网元102包括如下一个或多个网络设备：探针、服务器、基站、交换机、网关、路由器、光网络单元(optical network unit，ONU)、光线路终端(optical line terminal，OLT)、无线局域网(wireless local area network，WLAN)设备或防火墙等。
其中，对于所述至少一个网元102中的每个网元，所述网元102包括处理数据和/或转发业务等功能。第一设备101包括对所述网元102进行分析、管理和/或控制等功能。
例如，以所述网元102包括的处理数据功能为例，所述网元102处理数据的过程为：所述网元102获取数据，基于至少一个特征类型分析所述数据是否异常，所述至少一个特征类型与所述数据的数据行为相对应；基于分析的结果确定所述数据相关的业务或网络是否存在风险，所述业务包括所述网元102传输的业务，所述网络包括所述网元所在的网络。
但所述网元102所在的网络环境是不断变化的，在所述网元102所在的网络环境变化后，所述网元102获取的数据的数据行为可能发生变化，此时所述至少一个特征类型与变化后的数据行为可能不再相对应，如果所述网元102继续基于所述至少一个特征类型分析获取的数据，会降低对数据进行分析的准确度。
接下来以所述网元102为路由器为例，假设路由器处理的数据为路由数目，路由器中的至少一个特征类型包括平均值。路由器获取多个第一路由数目，所述多个第一路由数目的数据行为与平均值相对应。获取所述多个第一路由数目的平均值，基于获取的平均值分析所述多个第一路由数目是否异常。
当路由器所在的网络环境发生变化后，路由器获取多个第二路由数目，所述多个第二路由数目不再与平均值相对应，而是与方差相对应，此时路由器继续获取所述多个第二路由数目的平均值，基于获取的平均值分析所述多个第二路由数目是否异常，就会分析出错，降低对所述多个第二路由数目进行分析的准确度。
在一些实施例中，为了提高对数据进行分析的准确度，第一设备101控制或管理所述网元102。控制或管理的过程为：
第一设备101获取网元102发送的第一数据集，第一数据集包括所述网元102获取的多个第一数据。第一设备101基于第一数据集获取所述多个第一数据对应的至少一个数据特征；向所述网元102发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型是基于所述至少一个数据特征获取的。所述网元102接收所述触发信息，基于所述触发信息获取所述至少一个特征类型，基于所述至少一个特征类型分析第二数据集中的第二数据，第二数据集包括所述网元102获取的多个第二数据。
其中，第一数据集包括第二数据集或第二数据集中的部分第二数据，或者，第二数据集是网元102在发送第一数据集之后获取的数据。由于所述至少一个特征类型是基于所述网元102获取的第一数据得到的，所述至少一个特征类型与第一数据集中的第一数据的数据行为相对应，因此基于所述至少一个特征类型分析第二数据集中的第二数据时，可以提高分析的准确度。如此实现第一设备101控制或管理所述网元102，当然，还有其他实现第一设备101控制或管理所述网元102的方式，在此不再一一列举。
接下来继续以上述路由器为例，假设第一数据集包括路由器获取的多个第二路由数目，第一设备101获取第一数据集，基于第一数据集中的多个第二路由数目，获取特征类型包括方差，向路由器发送触发信息，所述触发信息包括方差。这样路由器获取第二数据集的方差，基于获取的方差分析第二数据集中的路由数目是否异常，第二数据集包括在网络环境发生变化后，路由器获取的多个路由数目。
对于上述数据集(包括第一数据集和/或第二数据集)，所述数据集中的数据包括至少一个KPI对应的数据。
所述至少一个KPI包括路由数目、丢包数目、路由表项数目、误码数量和时延等中的一个或多个。
第一数据集中的第一数据的数据行为包括数据变化行为，例如，第一数据集中的第一数据可能是周期变化的数据、震荡变化的数据或趋势变化的数据等。
所述至少一个数据特征用于反映第一数据集中的第一数据的数据行为。所述至少一个数据特征包括：第一数据集中的第一数据对应的数据波形特征，第一数据集中的数据内容特征和/或至少一个对象的对象标识。其中，所述至少一个对象包括获取第一数据集中的各第一数据的对象。
所述数据波形特征包括周期型、震荡型、水平型、突变型和趋势型等中的一个或多个。
所述至少一个特征类型包括平均值、方差、中值、最大值和最小值等中的一个或多个。
在一些实施例中，第一设备101存储有第一对应关系，和/或，对于所述至少一个网元102中的每个网元，所述网元102存储有第一对应关系。第一对应关系包括数据特征和特征类型。
第一对应关系中的各记录可以是技术人员配置的，和/或，是第一设备101或所述网元102自己学习的。
在一些实施例中，对于第一设备101中的第一对应关系保存的记录，所述记录可能是事先由技术人员配置的，和/或，所述记录是由第一设备101产生的，和/或，所述记录是其他设备发送的，其他设备包括网元和/或知识库(专家知识库和/或经验知识库)等。之后，在第一设备获取到至少一个数据特征和至少一个特征类型时，第一设备101还查询第一对应关系中是否保存目标记录，目标记录是包括所述至少一个数据特征和所述至少一个特征类型的记录，如果没有保存目标记录，将目标记录保存在第一对应关系中。同理，所述网元102中的第一对关系保存的记录也按上述过程得到的，在此不再详细说明。
对于第一对应关系中的任一条记录，所述记录包括一个或多个数据特征，以及一个或多个特征类型。例如，参见下表1所示的第一对应关系，对于表1中的第一条记录，第一条记录包括的数据特征为“周期型”，特征类型为“平均值”。对于表1的第二条记录，第二条记录包括两个数据特征和两个特征类型，所述两个数据特征为“周期型和震荡型”，所述两个特征类型为“方差和中值”。
表1
序号 数据特征 特征类型
1 周期型 平均值
2 周期型和震荡型 方差和中值
…… …… ……
对于上述第一对应关系中的记录，所述记录还可能包括一个或多个数据类型，数据类型可以是KPI、业务数据和/或控制信息等。例如，参见下表2所示的第一对应关系，对于表2中的第一条记录，第一条记录包括的数据特征为“周期型”，数据类型为“路由数目”，特征类型为“平均值”。对于表2的第二条记录，第二条记录包括两个数据特征、三个数据类型和两个特征类型，所述两个数据特征为“周期型和震荡型”，所述三个数据类型为“丢包数目、路由数目和时延”，所述两个特征类型为“方差和中值”。
表2
Figure PCTCN2021134273-appb-000001
参见图2，本申请实施例提供了一种处理数据的方法200，所述方法200可以应用于图1所示的网络架构，所述方法200的执行主体为所述网络架构100中的第一设备101，包括：
步骤201：第一设备接收第一消息，第一消息包括第二数据集，第二数据集包括网元获取的数据。
在步骤201中，第一设备周期性地接收网元发送的第一消息，第一消息中的第二数据集包括网元在一个周期内获取的数据。
也就是说，网元周期性地发送第二数据集，其中，网元确定第一周期，在第一周期内获取多个数据，以得到第二数据集，第二数据集包括在第一周期内获取的多个数据，向第一设备发送包括第二数据集的第一消息。在第一周期结束时，网元还继续确定第一周期，并继续在确定的第一周期内获取多个数据，使得网元周期地发送第二数据集。
网元获取数据的方式包括采集数据方式和/或接收数据方式等。例如，网元使用采集数据方式采集KPI，即第二数据集包括网元采集的KPI；和/或，网元使用接收数据方式接收网络中的数据，即第二数据集包括网元接收的数据。网元接收的数据可能是业务数据和/或控制信息等。
第二数据集包括至少一个第一数据序列，对于每个第一数据序列，所述第一数据序列与至少一个对象和数据类型相对应，所述至少一个对象位于网元中。所述第一数据序列中的数据是所述至少一个对象获取的属于所述数据类型的数据。即第二数据集中的数据是网元中的至少一个对象获取的数据，第二数据集中的数据属于至少一个数据类型。
所述至少一个对象包括网元上的接口、采集器和/或单板。
所述数据类型包括KPI、业务数据和/或控制信息等。第二数据集中的数据属于同一个数据类型，或者，属于不同的数据类型。以KPI为例，也就是说，第二数据集中的数据属于同一个KPI，或者，第二数据集中的数据属于多个KPI。
KPI包括路由数目、丢包数目、路由表项数目、误码数量和时延等中的一个或多个。
例如，假设第二数据集中的数据属于同一个KPI，KPI为路由数目，第二数据集中的数据包括所述网元在一个第一周期内获取的路由数目。
再例如，假设第二数据集中的数据属于多个KPI，多个KPI包括路由数目、丢包数目和路由表项数目，第二数据集中的数据包括所述网元在第一周期内获取的路由数目、丢包数目和路由表项数目。
对于上述第一周期，第一周期的第一周期长度是网元自己基于需求配置的，或者，第一周期长度是第一设备在网元上配置的。
在第一周期长度是第一设备配置的情况下，在执行步骤201之前，第一设备生成任务信息，所述任务信息包括第一周期长度，向网元发送所述任务信息，网元接收第一设备发送的 任务信息。这样网元接收所述任务信息后，基于第一周期长度周期性地获取第二数据集。
在一些实施例中，所述任务信息还包括至少一个数据类型，这样网元在第一周期内获取所述至少一个数据类型中的每个数据类型的数据，以得到第二数据集。
第一设备生成任务信息的操作为：
第一设备获取第一周期长度，生成包括第一周期长度的任务信息，向网元发送所述任务信息。
在一些实施例中，第一设备还获取所述至少一个数据类型，这样生成的所述任务信息还包括所述至少一个数据类型。
在一些实施例中，第一设备还获取网元范围，所述网元范围包括至少一个网元的网元标识。这样第一设备在生成所述任务信息后，基于所述网元范围包括的每个网元的网元标识，分别向每个网元发送所述任务信息，以触发所述网元范围内的各网元获取数据。
对于上述第一周期长度，第一周期长度是技术人员在第一设备上配置的。在实现时，第一设备显示配置界面，技术人员在配置界面中输入第一周期长度，第一设备从配置界面中获取第一周期长度。
在一些实施例中，技术人员还在配置界面上输入至少一个数据类型，第一设备还从配置界面中获取所述至少一个数据类型。
在一些实施例中，技术人员还在配置界面上输入网元范围，所述网元范围包括至少一个网元的网元标识。第一设备还从配置界面中获取所述网元范围。
例如，参见图3，第一设备显示配置界面，技术人员在配置界面中输入第一周期长度为5分钟，输入的数据类型为路由数目，输入的网元范围包括网元1的网元标识“ID-NE1”和网元2的网元标识“ID-NE2”，然后点击配置界面中的确认按钮，使确认按钮产生触发命令。
这样，第一设备检测到确认按钮的触发命令时，从配置界面中获取的第一周期长度为5分钟，获取的数据类型为路由数目，以及获取的网元范围包括网元1的网元标识“ID-NE1”和网元2的网元标识“ID-NE2”。生成任务信息，所述任务信息包括的第一周期长度为5分钟以及数据类型为“路由数目”，基于网元1的网元标识“ID-NE1”和网元2的网元标识“ID-NE2”，分别向网元1和网元2发送所述任务信息，以触发网元1和网元2均获取路由数目，并且每5分钟得到包括路由数目的第二数据集。
在一些实施例中，第一消息还包括第二数据集中的每个第一数据序列对应的属性信息。对于每个第一数据序列，所述第一数据序列对应的属性信息包括所述第一数据序列中的每个数据对应的获取时间，所述第一数据序列对应的数据类型和/或所述第一数据序列对应的对象的对象标识。
上述第一消息还包括网元的网元信息，所述网元信息包括网元的网元标识、网元名称、网元地址和网元类型等中一个或多个。
在一些实施例中，网元基于网络传输协议向第一设备发送第一消息，第一消息的格式为所述网络传输协议定义的消息格式。
所述网络传输协议包括网络配置协议(network configuration protocol，Netconf)协议、安全文件传送协议(secret file transfer protocol，SFTP)文件传输协议或遥感勘测(telemetry)协议等。
在所述网络传输协议为Netconf协议的情况，使用数据建模语言(yet another next  generation，YANG)模型在第一消息中封装第二数据集和/或第二数据集的属性信息。
步骤202：第一设备保存第一消息中的第二数据集。
在步骤202中，第一设备获取网元的网元信息，基于所述网元信息和第二对应关系，获取包括所述网元信息的记录，将所述第二数据集保存到获取的记录中。第二对应关系中的每条记录包括一个网元的网元信息和所述一个网元对应的第一数据集，所述第一数据集包括所述一个网元已获取的数据。
对于所述记录中的第一数据集，第一数据集可能包括至少一个第二数据序列。对于每个第二数据序列，所述第二数据序列与所述一个网元中的一个对象和数据类型相对应，所述第二数据序列包括所述一个对象已获取的属于所述数据类型的数据。
所述记录中还可能包括每个第二数据序列对应的属性信息。这样，第一设备将第二数据集保存到获取的记录中的操作为：在第一消息还包括第二数据集中的各第一数据序列对应的属性信息的情况下，第一设备基于各第一数据序列对应的属性信息，将各第一数据序列保存到获取的记录中。
在一些实施例中，第一设备还对第二数据集中的数据进行预处理，所述预处理包括去噪处理、补值处理和/或拼接处理等。
所谓去噪处理是指第一设备去除第二数据集包括的数据的噪声。其中，噪声可能是在传输该第二数据集的过程中产生。
所谓补值处理是指第一设备检测第二数据集是否存在数据丢失，如果存在数据丢失，在第二数据集中补充丢失的数据。其中，数据丢失可能是在传输第二数据集的过程中产生的。
在第二数据集中的数据是网元周期性获取的情况下，第一设备对第二数据集执行补值处理。所述补值处理的过程为：
对于第二数据集中的每个第一数据序列，第一设备从第一数据序列对应的属性信息中获取第一数据序列中的每个数据对应的获取时间，基于每个数据对应的获取时间，确定丢失的数据对应的获取时间；基于确定的获取时间，从第一数据序列中选择第一数据和第二数据，第一数据是在丢失的数据之前最近一次获取的数据，第二数据是在丢失的数据之后最近一次获取的数据；基于第一数据和第二数据，在第一数据序列中补充丢失的数据。
在一些实施例中，第一设备计算第一数据和第二数据之间的均值，将计算的均值作为丢失的数据。第一设备计算第一数据和第二数据之间的方差，将计算的方差作为丢失的数据。或者，第一设备对第一数据和第二数据进行加权计算，将计算得到的值作为丢失的数据。例如，可以采用加权公式进行加权计算，加权公式为Data＝Data1*a+Data2*b，Data为丢失的数据，Data1为第一数据，Data2为第二数据，a和b为两个指定的加权值，*为乘法运算，a+b＝1。当然，第一设备还可以采用其他方式获取丢失的数据，在此不再一一列举。
所谓拼接处理是指第一设备将获取的记录包括的第二数据序列和第二数据集中的第一数据序列拼接成一条数据序列，将获取的记录包括的第二数据序列更新为拼接的数据序列，以实现将第二数据集保存到获取的记录中。
第一消息中可能包括第二数据集中的每个第一数据序列对应的属性信息。对于第二数据集中的每个第一数据序列，第一设备从所述第一数据序列对应的属性信息中获取对象的对象标识和数据类型。基于所述对象的对象标识和所述数据类型，从获取的记录中选择与所述对象和所述数据类型相对应的第二数据序列。将所述第一数据序列和所述第二数据序列拼接成 一个数据序列，在获取的记录中将所述第二数据序列更新为拼接的一个数据序列。
步骤203：第一设备获取第一数据集，第一数据集包括网元发送的至少一个第二数据集。
第一数据集包括第一设备在第二周期内接收网元发送的第二数据集，第二周期的第二周期长度大于或等于第一周期长度。
在步骤203，第一设备确定第二周期，在第二周期结束时从第二对应关系中获取该网元对应的第一数据集。
第一设备还从第二对应关系中获取第一数据集对应的至少一个数据类型和/或至少一个对象的对象标识，至少一个数据类型包括第一数据集中的数据属于的类型，至少一个对象是获取第一数据集中的数据的对象。
在一些实施例中，在第一设备从第二对应关系中获取到网元对应的第一数据集后，还从第二对应关系中删除包括网元的网元信息的记录。这样第一设备在下一个第二周期内接收到网元发送的数据集时，基于网元的网元信息和接收的数据集在第二对应关系中重新创建包括网元的网元信息和接收的数据集的记录，保证第二对应关系中的每个数据集包括的数据是一个第二周期内获取的数据。
第二周期长度是第一设备配置的，第一设备基于需求配置第二周期长度。或者，
第二周期长度是技术人员在第一设备中配置的。例如，第一设备显示如图3所示的配置界面，技术人员在配置界面中输入第二周期长度，例如，第二周期长度为24小时，第一设备从配置界面中获取该第二周期长度。
步骤204：第一设备基于第一数据集，获取第一数据集中的数据对应的至少一个数据特征。
第一数据集包括至少一个第二数据序列，对至少一个第二数据序列中的每个第二数据序列进行处理，分别得到每个第二数据序列对应的数据特征。
至少一个数据特征包括如下一个或多个信息：第一数据集中的数据对应的数据波形特征，第一数据集中的数据内容特征和至少一个对象的对象标识。其中，至少一个对象包括获取第一数据集中的各数据的对象。
对于数据波形特征，接下来列举了一种获取数据波形特征的示例。所述示例为：
对于每个第二数据序列，基于所述第二数据序列包括的每个数据和每个数据的获取时间，在坐标系中确定每个数据对应的点，所述坐标系的横轴为时间，纵轴为数据值。在所述坐标系中串联每个数据对应的点，得到所述第二数据序列对应的数据波形，使用波形处理模型对所述数据波形进行处理，输出所述第二数据序列对应的数据波形特征。
数据波形特征包括周期型、震荡型、水平型、突变型和趋势型等中的一个或多个。
对于所述波形处理模型，事先获取多个不同数据波形，并标记每个数据波形对应的数据波形特征，以得到多个训练样本，每个训练样本包括至少一个数据波形和所述至少一个数据波形对应的数据波形特征。使用所述多个训练样本训练智能算法，得到波形处理模型。
智能算法包括卷积神经网络、随机森林算法、逻辑回归算法或支持向量机(support vector machine，SVM)等。
对于数据内容特征，数据内容特征包括业务特征和/或用户特征等。
在一些实施例中，业务特征包括业务标识、业务源访问次数和/或业务流数目等，用户特征包括用户请求业务次数等。
在一些实施例中，第一数据集可能包括业务数据等，业务数据包括网元接收的业务流等，业务流包括业务标识、五元组信息、业务的域名和/或统一资源定位符(uniform resource locator，URL)等，业务流的域名、源地址和/或URL用于标识业务流的网络来源，该网络来源又可称为业务源。通过对第一数据集中的不同业务流中的源地址、域名或URL进行统计得到不同业务流对应的业务源访问次数。
业务流的五元组信息可用于标识一个业务流，通过对第一数据集中的不同五元组信息进行统计，得到业务流数目。业务流的目的地址可能是请求所述业务流的用户的地址，对于第一数据集中的任一个目的地址，通过对包括所述目的地址的业务流进行统计，得到所述目的地址对应的用户请求业务次数。
对于所述至少一个对象的对象标识，第一设备从第二对应关系中获取所述至少一个第二数据序列中的每个第二数据序列对应的对象的对象标识，即得到所述至少一个对象的对象标识。
在步骤204中，第一设备还可以基于至少一个数据特征和第一对应关系，获取至少一个特征类型。
在实现时，第一设备从第一对应关系中确定至少一条记录，所述至少一条记录包括的数据特征和所述至少一个数据特征相同，获取所述至少一条记录中包括的特征类型。
在一些实施例中，对于第一对应关系中保存的记录，所述记录可能是技术人员配置的，和/或，所述记录是第一设备产生的，和/或，所述记录是第一设备接收其他设备发送的，其他设备包括网元和/或知识库。
在一些实施例中，在第一设备获取到所述至少一个数据特征和所述至少一个特征类型时，第一设备还查询第一对应关系中是否保存目标记录，目标记录是包括所述至少一个数据特征和所述至少一个特征类型的记录，如果没有保存目标记录，将目标记录保存在第一对应关系中。
在一些实施例中，第一设备还获取至少一个数据类型，基于至少一个数据特征、第一对应关系和至少一个数据类型，获取至少一个特征类型，至少一个数据类型包括第一数据集中的数据属于的类型。
在实现时，第一设备从第一对应关系中确定至少一条记录，所述至少一条记录包括的数据特征和所述至少一个数据特征相同，以及所述至少一条记录包括的数据类型和所述至少一个数据类型相同，获取所述至少一条记录中包括的特征类型。
步骤205：第一设备向网元发送触发信息，触发信息包括所述至少一个数据特征和/或所述至少一个特征类型，以使网元基于触发信息处理网元获取的数据。
在一些实施例中，触发信息还包括上述获取的至少一个对象的对象标识，和/或，上述获取的至少一个数据类型。
第一设备通过网络传输协议向网元发送触发信息，所述触发信息的格式是所述网络传输协议定义的消息格式，所述网络传输协议包括Netconf、传输控制协议(transmission control protocol，TCP)或用户数据报协议(user datagram protocol，UDP)等。
网元接收所述触发信息后，处理网元获取的数据。其中，网元处理数据的过程为：
基于所述触发信息获取至少一个特征类型，基于所述至少一个特征类型和第二数据集获取数据指标，基于所述数据指标确定第二数据集中的数据的异常情况。
所述至少一个特征类型包括平均值、方差、中值、最大值和最小值等中的一个或多个，所以网元基于所述至少一个特征类型和第二数据集获取的数据指标，包括所述第二数据集中的数据的平均值、所述第二数据集中的数据的方差、所述第二数据集的中值、所述第二数据集中的最大值和所述第二数据集中的最小值等中的一个或多个。
其中，网元中包括第三对应关系，第三对应关系中的每条记录包括至少一个特征类型和数据处理算法，所述数据处理算法用于基于所述至少一个特征类型和第二数据集获取数据指标，并基于所述数据指标确定第二数据集中的数据的异常情况。
在一些实施例中，所述数据处理算法还包括至少一个阈值，所述数据处理算法将所述数据指标与所述至少一个阈值进行比较，基于比较的结果确定第二数据集中的数据的异常情况。
这样网元处理数据的过程具体为：
网元基于所述至少一个特征类型和第三对应关系，获取至少一个数据处理算法，基于所述至少一个特征类型和第二数据集，通过所述至少一个数据处理算法确定数据指标的异常情况，将所述数据指标与所述至少一个数据处理算法中的阈值进行比较，基于比较的结果确定第二数据集中的数据的异常情况。
例如，假设第三对应关系包括平均值与数据处理算法的对应关系，所述数据数理算法包括平均值阈值。网元接收触发信息，所述触发信息中的至少一个特征类型包括平均值。基于平均值和第三对应关系，获取所述数据处理算法。基于所述至少一个特征类型和第二数据集，得到所述第二数据集的数据指标，所述数据指标为第二数据集中的数据的平均值。将所述数据指标与所述数据处理算法中的平均值阈值进行比较，如果比较结果为所述数据指标大于所述平均值阈值，在这种情况下，可确定所述数据指标异常及所述第二数据集中的数据异常；如果比较结果为所述数据指标小于或等于所述平均值阈值，在这种情况下，可确定所述数据指标正常及所述第二数据集中的数据正常。
上述第二数据集可能是网元在接收所述触发信息之前获取的，在此情况下第一数据集可能包括第二数据集或第二数据集中的部分数据。或者，上述第二数据集也可以是网元在接收所述触发信息之后获取的，在此情况下第一数据集不包括第二数据集，第二数据集是在第一设备获取第一数据集之后网元获取的数据集。
在本申请实施例中，第一设备接收网元发送的第二数据集并保存第二数据集，在接收到一个第二周期内的第二数据集时，即得到第一数据集时，对第一数据集进行处理得到至少一个数据特征，向网元发送触发信息，该所述触发信息包括该所述至少一个数据特征和至少一个特征类型，该所述至少一个特征类型是基于该所述至少一个数据特征获取的。这样在网元接收触发信息，基于该所述触发信息获取该所述至少一个特征类型，基于该所述至少一个特征类型和第二数据集确定数据指标，基于该所述数据指标分析第二数据集的异常情况。由于网元每当获取到第二数据集可以向第一设备发送第二数据集，这样网元不需要长时间保存第二数据集，且第一设备对第一数据集进行处理，这样网元不用对第一数据集进行处理得到数据特征，如此本申请实施例可以适用于计算性能和/或存储性能较低的网元。又由于基于第一数据集获取到该所述至少一个特征类型，该所述至少一个特征类型与第一数据集中的数据行为相对应。即使因为网络环境变化导致网元采集的数据变化，通过本申请实施例，可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的至少一个特征类型，网元基于所述至少一个特征类型获取的数据指标反映网络情况。 所以通过本申请实施例降低了分析第二数据集中的数据的难度，从而基于该所述数据指标可以准确地分析第二数据集的异常情况，提高对数据分析的准确度。
参见图4，本申请实施例提供了一种处理数据的方法400，所述方法400可以应用于图1所示的网络架构100，所述方法400的执行主体为所述网络架构100中的网元102，包括：
步骤401：网元获取多个第二数据，得到第二数据集，第二数据集包括所述多个第二数据。
在步骤401中，网元确定第一周期，在第一周期内获取多个第二数据，以得到第二数据集，第二数据集包括在第一周期内获取的多个第二数据。
在第一周期内，网元获取数据的方式包括实时获取数据、周期性地获取数据和/或随机获取数据等。第二数据集包括至少一个第一数据序列，每个第一数据序列包括按获取时间排列的数据。每个第一数据序列中的数据可能是网元中的至少一个对象获取的数据，和/或，是同一数据类型的数据。
其中，在第一周期结束时，网元还继续确定第一周期，并继续在确定的第一周期内获取多个数据，使得网元周期地获取第二数据集。
网元包括至少一个对象，对于至少一个对象中的每个对象，所述对象在第一周期内获取数据。网元获取每个对象获取的数据，以得到第二数据集。即第二数据集包括的数据可能是同一个对象获取的数据，或者，第二数据集包括的数据可能是多个对象获取的数据。
在一些实施例中，第二数据集包括每个对象对应的第一数据序列，所述对象对应的第一数据序列包括所述对象获取的数据。
所述至少一个对象包括网元上的接口、采集器和/或单板。
在一些实施例中，第二数据集中的数据属于同一个数据类型，或者，第二数据集中的数据属于多个数据类型。
在一些实施例中，上述数据类型为KPI、业务数据和/或控制信息等。以KPI为例，也就是说，第二数据集中的数据属于同一个KPI，或者，第二数据集中的数据属于多个KPI。
KPI包括路由数目、丢包数目、路由表项数目、误码数量和时延等中的一个或多个。
在一些实施例中，对于网元中的每个对象，所述对象获取至少一种数据类型的数据，在第二数据集中所述对象对应至少一个第一数据序列，每个第一数据序列对应一个数据类型，且每个第一数据序列包括的数据属于同一数据类型。
基于上述论述，能够得出第二数据集包括至少一个第一数据序列，每个第一数据序列与一个对象和数据类型相对应。
在步骤401中，网元在得到第二数据集后，还分析第二数据集中的数据的异常情况，即分析第二数据集中的数据是否异常。其中，网元详细分析第二数据集的过程将在后续内容进行详细说明，在此先不做具体介绍。
对于上述第一周期，第一周期的第一周期长度可以是网元自己基于需求配置的，或者，第一周期长度是第一设备在该网元上配置的。
在第一周期长度是第一设备配置的情况下，在执行步骤401之前，网元接收第一设备发送的任务信息，所述任务信息包括第一周期长度。网元接收所述任务信息后，基于第一周期长度周期性地获取第二数据集。
在一些实施例中，所述该任务信息还可以包括至少一个数据类型，这样网元在第一周期内获取所述至少一个数据类型中的每个数据类型的数据，以得到第二数据集。
步骤402：网元向第一设备发送第一消息，第一消息包括第二数据集。
在一些实施例中，第一消息还包括第二数据集中的每个第一数据序列对应的属性信息。对于每个第一数据序列，所述第一数据序列对应的属性信息包括所述第一数据序列中的每个数据对应的获取时间，所述第一数据序列对应的数据类型和/或所述第一数据序列对应的对象的对象标识。
在一些实施例中，所述第一消息还包括网元的网元信息，所述网元信息包括网元的网元标识、网元名称、网元地址和网元类型等中一个或多个。
在一些实施例中，网元基于网络传输协议向第一设备发送第一消息，第一消息的格式为所述网络传输协议定义的消息格式。
所述网络传输协议包括Netconf协议、SFTP文件传输协议或telemetry协议等。
在一些实施例中，在所述网络传输协议为Netconf协议的情况，使用YANG模型在第一消息中封装第二数据集和/或第二数据集的属性信息。
其中，网元重复上述步骤401至402的操作，即网元周期性地向第一设备发送第二数据集。第一设备保存接收到的第二数据集，在接收一段时间的第二数据集时，获取第一数据集，第一数据集包括至少一个第二数据集。第一设备基于第一数据集，获取至少一个数据特征和/或至少一个特征类型；向网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或所述至少一个特征类型。
在一些实施例中，所述触发信息还包括至少一个特征类型和/或至少一个对象的对象标识。所述至少一个特征类型包括第一数据集中的数据属于的数据类型，所述至少一个对象包括获取第一数据集中的数据的对象。
其中，第一设备基于第一数据集获取至少一个数据特征和/或至少一个特征类型以及发送触发信息的详细实现过程，参见图2所示方法200中的内容，在此不再详细说明。
步骤403：网元接收第一设备发送的触发信息，所述触发信息包括至少一个数据类型和/或至少一个特征类型。
在一些实施例中，网元通过网络传输协议接收第一设备发送的触发信息，所述触发信息的格式是所述网络传输协议定义的消息格式，所述网络传输协议包括Netconf、TCP或UDP等。
步骤404：网元基于所述触发信息获取所述至少一个特征类型，基于所述至少一个特征类型和第二数据集获取数据指标。
在所述触发信息包括所述至少一个特征类型时，网元从所述触发信息中读取所述至少一个特征类型。
在所述触发信息包括所述至少一个数据特征时，网元基于所述至少一个数据特征和第一对应关系获取所述至少一个特征类型。或者，在所述触发信息包括所述至少一个数据特征和至少一个数据类型时，网元基于所述至少一个数据特征、所述至少一个数据类型和第一对应关系获取所述至少一个特征类型。
所述至少一个特征类型包括平均值、方差、中值、最大值和最小值等中的一个或多个，所以基于所述至少一个特征类型和第二数据集获取的数据指标，包括第二数据集中的数据的 平均值、第二数据集中的数据的方差、第二数据集的中值、第二数据集中的最大值和第二数据集中的最小值等中的一个或多个。
网元在获取到所述数据指标后，基于所述数据指标确定第二数据集中的数据的异常情况。
其中，网元中包括第三对应关系，第三对应关系中的每条记录包括至少一个特征类型和数据处理算法，所述数据处理算法用于基于所述至少一个特征类型和第二数据集获取数据指标，并基于所述数据指标确定第二数据集中的数据的异常情况。
在一些实施例中，所述数据处理算法还包括至少一个阈值，所述数据处理算法将所述数据指标与所述至少一个阈值进行比较，基于比较的结果确定第二数据集中的数据的异常情况。
这样在步骤404中，网元获取数据指标的操作为：
网元基于所述至少一个特征类型和第三对应关系，获取至少一个数据处理算法，基于所述至少一个特征类型和第二数据集，通过所述至少一个数据处理算法确定数据指标的异常情况，将所述数据指标与所述至少一个数据处理算法中的阈值进行比较，基于比较的结果确定第二数据集中的数据的异常情况。
例如，假设第三对应关系包括平均值与数据处理算法的对应关系，所述数据数理算法包括平均值阈值。网元接收触发信息，所述触发信息中的至少一个特征类型包括平均值。基于所述平均值和第三对应关系，获取所述数据处理算法。基于所述至少一个特征类型和第二数据集，得到所述第二数据集的数据指标，所述数据指标是第二数据集中的数据的平均值。将所述数据指标与所述平均值阈值进行比较，如果比较结果为所述数据指标大于所述平均值阈值，在这种情况下，可确定所述数据指标异常及第二数据集中的数据异常；如果比较结果为所述数据指标小于或等于所述平均值阈值，在这种情况下，确定所述数据指标正常及第二数据集中的数据正常。
对于上述第三对应关系中的每条记录，所述记录包括至少一个特征类型，至少一个数据类型和数据处理算法，所述记录表示所述数据处理算法用于基于所述至少一个特征类型对属于所述至少一个数据类型的数据进行处理。
这样网元获取数据处理算法的操作为：在所述触发信息包括至少一个数据类型的情况下，网元基于所述至少一个数据类型，所述至少一个特征类型和第三对应关系，获取数据处理算法。
在一些实施例中，网元还向第一设备发送确定的异常情况和/或数据指标。
网元在确定第二数据集中的数据的异常时，向第一设备发送所述异常情况。其中，网元可以采用事件上报的方式向第一设备发送所述异常情况。
上述第二数据集可能是网元在接收所述触发信息之前获取的，在此情况下第一数据集可能包括第二数据集或第二数据集中的部分数据。或者，上述第二数据集也可能是网元在接收所述触发信息之后获取的，在此情况下第一数据集不包括第二数据集，第二数据集是在第一设备获取第一数据集之后网元获取的数据集。
在一些实施例中，在所述触发信息还包括至少一个对象的对象标识时，网元还基于所述至少一个对象的对象标识获取第二数据集，第二数据集包括所述至少一个对象获取的多个数据。
在一些实施例中，如果网元的计算性能和/或存储性能较高，网元可不用向第一设备发送第二数据集，以让第一设备获取至少一个数据特征和/或至少一个特征类型。在此情况下，网 元在获取到第二数据集后，在本地保存第二数据集。在获取一段时间，获取第一数据集，第一数据集包括在所述一段时间内获取的至少一个第二数据集。网元对第一数据集中的数据进行处理，得到所述至少一个数据特征，基于所述至少一个数据特征获取所述至少一个特征类型。
在此情况下，网元包括第一对应关系，网元基于所述至少一个数据特征和第一对应关系获取所述至少一个特征类型。或者，网元获取至少一个数据类型，所述至少一个数据类型包括第一数据集中的数据属于的数据类型，基于所述至少一个数据特征、所述至少一个数据类型和第一对应关系获取所述至少一个特征类型。
在本申请实施例中，网元获取多个数据，得到包括所述多个数据的第二数据集，向第一设备发送第二数据集。其中，网元周期性地向第一设备发送第二数据集，使第一设备第一数据集进行处理，得到至少一个数据特征和/或至少一个特征类型，第一数据集包括网元发送的至少一个第二数据集。网元接收第一设备发送的触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型。网元基于所述触发信息获取所述至少一个特征类型，基于所述至少一个特征类型和第二数据集确定数据指标，基于所述数据指标分析第二数据集的异常情况。由于网元每当获取到第二数据集可以向第一设备发送第二数据集，这样网元不需要长时间保存第二数据集，且第一设备对第一数据集进行处理，这样网元不用对第一数据集进行处理得到数据特征，如此本申请实施例可以适用于计算性能和/或存储性能较低的网元。又由于基于第一数据集获取到所述至少一个特征类型，所述至少一个特征类型与第一数据集中的数据行为相对应。这样即使因为网络环境变化导致网元采集的数据变化，通过本申请实施例可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的至少一个特征类型。网元基于所述至少一个特征类型和第二数据集获取的数据指标不可以准确反映网络情况，从而基于所述数据指标可以准确地分析第二数据集的异常情况，提高对数据分析的准确度。另外，网元也可不用向第一设备发送第二数据集，即网元自己获取第一数据集，对第一数据集进行处理得到至少一个数据特征，基于该至少一个数据特征获取至少一个特征类型，这样能够节省网络资源。
参见图5，本申请实施例提供了一种处理数据的方法500，所述方法500应用于图1所示的网络架构100。在所述方法500中第一设备接收网元获取的数据得到第一数据集，第一数据集包括网元获取的多个数据，基于第一数据集获取至少一个数据特征或至少一个特征类型。第一设备向网元发送触发信息，该触发信息包括至少一个数据特征或至少一个特征类型，网元基于该触发信息处理第二数据集。所述方法500包括：
步骤501：网元获取多个数据，得到第二数据集，第二数据集包括所述多个数据。
网元得到第二数据集的详细实现过程，参见图4所示的方法400中的步骤401中的相关内容，在此不再详细说明。
步骤502：网元向第一设备发送第一消息，第一消息包括第二数据集。
网元发送第一消息的详细实现过程，参见图4所示的方法400中的步骤402中的相关内容，在此不再详细说明。
步骤503：第一设备接收第一消息，保存第一消息中的第二数据集。
第一设备保存第一消息中的第二数据集的详细实现过程，参见图2所示的方法200中的 步骤202中的相关内容，在此不再详细说明。
步骤504：第一设备获取第一数据集，第一数据集包括网元发送的至少一个第二数据集。
第一设备获取第一数据集的详细实现过程，参见图2所示的方法200中的步骤203中的相关内容，在此不再详细说明。
步骤505：第一设备基于第一数据集，获取第一数据集中的数据对应的至少一个数据特征。
第一设备获取至少一个数据特征的详细实现过程，参见图2所示的方法200中的步骤204中的相关内容，在此不再详细说明。
步骤506：第一设备向网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型是基于所述至少一个数据特征得到的。
第一设备发送触发信息的详细实现过程，参见图2所示的方法200中的步骤205中的相关内容，在此不再详细说明。
步骤507：网元接收所述触发信息，基于所述触发信息获取至少一个特征类型，基于所述至少一个特征类型和第二数据集获取数据指标。
第一设备获取至少一个特征类型和数据指标的详细实现过程，参见图4所示的方法400中的步骤404中的相关内容，在此不再详细说明。
在本申请实施例中，第一设备接收网元发送的第二数据集并保存第二数据集，在接收到一个第二周期内的第二数据集时，即得到第一数据集时，对第一数据集进行处理得到至少一个数据特征，向网元发送触发信息，所述触发信息包括该至少一个数据特征和至少一个特征类型。网元基于所述触发信息获取至少一个特征类型，基于至少一个特征类型和第二数据集确定数据指标，基于数据指标分析第二数据集的异常情况。由于基于第一数据集获取到至少一个特征类型，所述至少一个特征类型与第一数据集中的数据行为相对应。这样即使因为网络环境变化导致网元采集的数据变化，通过本申请实施例，可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的至少一个特征类型。网元基于所述至少一个特征类型获取的数据指标就可以准确反映网络情况。所以通过本申请实施例降低了分析第二数据集中的数据的难度，从而基于所述数据指标准确地分析第二数据集的异常情况，提高对数据分析的准确度。
对于图5所示的方法500，接下来，本申请实施例提供了一种具体示例，以对所述方法500进行详细说明，在所述示例中，以网元为路由器为例，假设路由器包括第一单板和第二单板，第一单板是所述路由器的第一对象，用于获取路由数目，第二单板是所述路由器的第二对象，也用于获取路由数目。参见图6，所述示例包括：
步骤601：路由器获取第一单板在一个第一周期内获取的多个路由数目，以及获取第二单板在所述第一周期内获取的多个路由数目，得到第二数据集。
在步骤601中，第二数据集包括两个第一数据序列，分别为第一数据序列1和第一数据序列2。第一数据序列1与第一单板相对应，第一数据序列1包括第一单板获取的多个路由数目，所述多个路由数目包括1、3、4和5。第一数据序列2与第二单板相对应，第一数据序列2包括第二单板获取的多个路由数目，所述多个路由数目包括2、3、4和5。
步骤602：路由器向第一设备发送第一消息，第一消息包括第二数据集，第二数据集包 括第一数据序列1和第一数据序列2。
在一些实施例中，第一消息还包括第一数据序列1对应的属性信息1和第一数据序列2对应的属性信息2。属性信息1包括第一数据序列1对应的数据类型和/或对象的对象标识，所述数据类型为“路由数目”，所述对象的对象标识是第一单板的标识“ID-ob1”。
属性信息2包括第一数据序列2对应的数据类型和/或对象的对象标识，所述数据类型也为“路由数目”，所述对象的对象标识是第二单板的标识“ID-ob2”。
步骤603：第一设备接收第一消息，将第二数据集包括的第一数据序列1和第一数据序列2保存到第二对应关系中。
第一设备的本地保存第二对应关系。第二对应关系中的每条记录包括网元的网元信息和第一数据集，第一数据集包括所述网元已发送的数据，所述网元信息包括所述网元的标识、名称、地址和/或类型。
对于所述记录中的第一数据集，第一数据集可能包括至少一个第二数据序列。对于每个第二数据序列，所述第二数据序列与所述网元中的一个对象和数据类型相对应，所述第二数据序列包括所述一个对象已获取的属于所述数据类型的数据。所述记录中还可能包括每个第二数据序列对应的属性信息。
例如，参见下表3所示的第二对应关系，第二对应关系中的第一条记录包括路由器的标识、第一数据集和属性信息。路由器的标识为“ID-NE1”，第一数据集包括第二数据序列1和第二数据序列2，第二数据序列1包括路由器中的第一单板已获取的路由数目1、3、4和5，第二数据序列2包括路由器中的第二单板已获取的路由数目2、3、4和5。属性信息包括第二数据序列1的属性信息1和第二数据序列2的属性信息2；属性信息1包括对象标识和数据类型，对象标识是第一单板的标识“ID-ob1”，数据类型为“路由数目”；属性信息2也包括对象标识和数据类型，对象标识是第二单板的标识“ID-ob2”，数据类型为“路由数目”。
表3
Figure PCTCN2021134273-appb-000002
在步骤603中，第一设备接收的路由器发送的第一消息，第一消息中的第二数据集包括第一数据序列1(包括数据1、3、4、5)和第二数据序列2(包括数据2、3、4、5)，第一消息还包括第一数据序列1的属性信息1和第一数据序列2的属性信息2。属性信息1包括第一数据序列1对应的对象的对象标识和数据类型，对象标识是第一单板的标识“ID-ob1”，数据类型为“路由数目”。属性信息2包括第一数据序列2对应的对象的对象标识和数据类型，对象标识是第二单板的标识“ID-ob2”，数据类型为“路由数目”。
第一设备获取路由器的标识“ID-NE1”，基于“ID-NE1”，从如表3所示的第二对应关系中获取包括“ID-NE1”的第一条记录。对于第一数据序列1(包括数据1、3、4、5)，第一设备基于第一单板的标识“ID-ob1”和“路由数目”，从第一条记录中获取第二数据序列1(包括数据1、3、4、5)。将第一数据序列1和第二数据序列1拼接成一个数据序列，拼接的数据 序列包括数据1、3、4、5、1、3、4和5。参见下表4，将第一条记录中第二数据序列1更新为拼接的数据序列，以实现将第一数据序列1保存到第二对应关系中。
对于第一数据序列2(包括数据2、3、4、5)，第一设备基于第二单板的标识“ID-ob2”和“路由数目”，从第一条记录中获取第二数据序列2(包括数据2、3、4、5)。将第一数据序列2和第二数据序列2拼接成一个数据序列，拼接的数据序列包括数据2、3、4、5、2、3、4和5。参见下表4，将第一条记录中第二数据序列2更新为拼接的数据序列，以实现将第一数据序列2保存到第二对应关系中。
表4
Figure PCTCN2021134273-appb-000003
步骤604：第一设备从第二对应关系中获取第一数据集，第一数据集包括在第二周期内路由器发送的至少一个第二数据集。
第一设备从如表4所示的第二对应关系中的第一条记录中获取路由器对应的第一数据集，第一数据集包括第二数据序列1和第二数据序列2，第二数据序列1包括数据1、3、4、5、1、3、4和5，第二数据序列2包括数据2、3、4、5、2、3、4和5。
第一设备还从第一条记录中获取至少一个对象的对象标识和至少一个数据类型，所述至少一个对象的对象标识包括第一单板的标识“ID-ob1”和第二单板的标识“ID-ob2”，所述至少一个数据类型包括路由数目。
步骤605：第一设备基于第一数据集，获取第一数据集中的数据对应的至少一个数据特征。
第一数据集包括第二数据序列1和第二数据序列2，第二数据序列1包括数据1、3、4、5、1、3、4和5，第二数据序列2包括数据2、3、4、5、2、3、4和5。对第二数据序列1包括数据1、3、4、5、1、3、4和5进行处理，得到数据波形特征为周期型。对第二数据序列2包括数据2、3、4、5、2、3、4和5进行处理，得到数据波形特征为周期型。因此，对第一数据集进行处理得到的至少一个数据特征包括周期型。
在一些实施例中，第一设备还基于所述周期型、第一数据集属于的数据类型“路由数目”和如表2所示的第一对应关系，获取至少一个特征类型，所述至少一个特征类型包括平均值。
步骤606：第一设备向路由器发送触发信息，所述触发信息包括的数据特征“周期型”和/或特征类型“平均值”。
在一些实施例中，所述触发信息还包括第一单板的标识“ID-ob1”、第二单板的标识“ID-ob2”和/或“路由数目”。
步骤607：路由器接收所述触发信息，基于所述触发信息获取至少一个特征类型，基于所述至少一个特征类型和第二数据集获取数据指标。
路由器保存有第三对应关系，第三对应关系包括平均值和数据处理算法的对应关系。所述数据数理算法包括平均值阈值。
在步骤607中，路由器接收触发信息，基于所述触发信息获取至少一个特征类型，所述至少一个特征类型包括平均值。基于平均值和第三对应关系，获取所述数据处理算法。基于所述至少一个特征类型和第二数据集，得到的数据指标，所述数据指标为所述第二数据集中的路由数目平均值。比较路由数目平均值与所述平均值阈值，如果比较结果为路由数目平均值大于所述平均值阈值，确定路由数目平均值异常及第二数据集中的路由数目异常；如果比较结果为路由数目平均值小于或等于所述平均值阈值，确定路由数目平均值正常及第二数据集中的路由数目正常。
在本申请实施例中，第一设备接收路由器发送的第二数据集并保存第二数据集，在接收到一段时间内的第二数据集时，即得到第一数据集时，对第一数据集进行处理得到至少一个数据特征，向路由器发送触发信息，所述触发信息包括该至少一个数据特征和至少一个特征类型。所述基于所述触发信息获取至少一个特征类型，基于至少一个特征类型和第二数据集确定数据指标，基于数据指标分析第二数据集的异常情况。由于基于第一数据集获取到至少一个特征类型，所述至少一个特征类型与第一数据集中的数据行为相对应。这样即使因为网络环境变化导致路由器采集的数据变化，通过本申请实施例，可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的至少一个特征类型。路由器基于所述至少一个特征类型获取的数据指标就可以准确的反映网络情况。所以通过本申请实施例降低了分析第二数据集中的数据的难度，从而基于所述数据指标准确地分析第二数据集的异常情况，提高对数据分析的准确度。
参见图7，本申请实施例提供了一种处理数据的装置700，所述装置700可以部署在上述任意实施例提供的第一设备上，例如部署在图1所示网络架构100中的第一设备101上，或图2所示方法200中的第一设备，或图5所示方法500中的第一设备或图6所示方法600中的第一设备上。所述装置700包括：
接收单元701，用于接收网元发送的第一数据集，所述第一数据集包括所述网元获取的多个第一数据；
处理单元702，用于基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征；
发送单元703，用于向所述网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型与所述至少一个数据特征有关。
在一些实施例中，所述网元接收所述触发信息，基于所述触发信息获取所述至少一个特征类型，基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括所述网元获取的多个第二数据。
在一些实施例中，接收单元701接收第一数据集的详细实现过程，参见上述图2所示方法200的步骤201-203，图5所示方法500的步骤503-504，以及图6所示方法600的步骤603-604中的相关内容，在此不再详细说明。
处理单元702获取所述至少一个数据特征的详细实现过程，参见上述图2所示方法200的步骤204，图5所示方法500的步骤505，以及图6所示方法600的步骤605中的相关内容，在此不再详细说明。
在一些实施例中，所述处理单元702，还用于：
基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
在一些实施例中，所述处理单元702，用于：
基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和/或所述至少一个特征类型。
在一些实施例中，处理单元702获取所述至少一个特征类型的详细实现过程，参见上述图2所示方法200的步骤204，图5所示方法500的步骤505，以及图6所示方法600的步骤605中的相关内容，在此不再详细说明。
在一些实施例中，所述触发信息包括所述至少一个数据特征和/或至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型。
在第一实施例中，所述网元基于所述至少一个数据特征和/或所述至少一个数据类型获取所述至少一个特征类型。
在一些实施例中，所述触发信息还包括至少一个第一对象的对象标识，所述至少一个第一对象位于所述网元中。
在一些实施例中，所述网元基于所述至少一个第一对象的对象标识获取所述第二数据集，所述第二数据集包括所述至少一个第一对象获取的第二数据。
在一些实施例，关于触发信息的详细说明，参见上述图2所示方法200的步骤205，图4所示方法400的步骤403，图5所示方法500的步骤506，以及图6所示方法600的步骤606中的相关内容，在此不再详细说明。
在一些实施例中，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；其中，所述至少一个第二对象包括获取所述第一数据集中的各第一数据的对象。
在一些实施例中，所述装置700包括：数据处理系统、控制器或管理设备。
在一些实施例中，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括所述网元在发送所述第一数据集之后获取的数据。
在本申请实施例中，所述接收单元接收的第一数据集包括所述网元获取的多个第一数据。所述处理单元基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征，且所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应。发送单元向所述网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型。如此，所述网元基于所述触发信息获取所述至少一个特征类型，其中，由于所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应，而至少一个特征类型是基于所述至少一个数据特征获取的。即使因为网络环境变化导致网元采集的数据变化，通过本申请实施例，可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的至少一个特征类型。网元基于所述至少一个特征类型和第二数据集获取的数据指标就可以准确的反映网络情况。所以网元通过本申请实施例降低了分析第二数据集中的数据的难度，可以提高获取的数据指标的准确性，提高对第二数据集中的数据进行分析的准确度。
参见图8，本申请实施例提供了一种处理数据的装置800，所述装置800可以部署在上述任意实施例提供的网元上，例如部署在图1所示网络架构100中的网元102上，或图4所示方法400中的网元，或图5所示方法500中的网元或图6所示方法600中的路由器上。所述装置800包括：
处理单元801，用于获取第一数据集，所述第一数据集包括获取的多个第一数据；
所述处理单元801，还用于基于所述第一数据集获取至少一个特征类型，所述至少一个特征类型与所述多个第一数据对应的至少一个数据特征相对应；
所述处理单元801，还用于基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括获取的多个第二数据。
在一些实施例中，处理单元801获取所述至少一个特征类型以及获取数据指标的详细实现过程，参见上述图4所示方法400的步骤404，图5所示方法500的步骤507，以及图6所示方法600的步骤607中的相关内容，在此不再详细说明。
在一些实施例中，所述装置800还包括：发送单元802和接收单元803；
所述发送单元802，用于发送所述第一数据集，所述第一数据集用于所述第一数据集的接收方获取所述至少一个数据特征和/或所述至少一个特征类型；
所述接收单元803，用于接收触发信息，所述触发信息包括所述至少一个数据特征和/或所述至少一个特征类型；
所述处理单元801，用于基于所述触发信息获取所述至少一个特征类型。
在一些实施例中，发送单元802发送所述第一数据集的详细实现过程，参见上述图4所示方法400的步骤402，图5所示方法500的步骤502，以及图6所示方法600的步骤602中的相关内容，在此不再详细说明。
在一些实施例中，接收单元803接收所述触发信息的详细实现过程，参见上述图4所示方法400的步骤403，图5所示方法500的步骤507，以及图6所示方法600的步骤607中的相关内容，在此不再详细说明。
在一些实施例中，所述触发信息包括所述至少一个数据特征，所述处理单元801，用于：
基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
在一些实施例中，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型；
所述处理单元801，用于：
基于所述至少一个数据特征、所述至少一个数据类型和所述第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
在一些实施例中，处理单元801基于所述至少一个数据类型获取所述至少一个特征类型的详细实现过程，参见上述图4所示方法400的步骤404，图5所示方法500的步骤507，以及图6所示方法600的步骤607中的相关内容，在此不再详细说明。
在一些实施例中，所述触发信息还包括至少一个第一对象的对象标识；
所述处理单元801，还用于：
基于所述至少一个第一对象的对象标识获取所述第二数据集，所述第二数据集包括所述至少一个第一对象获取的多个第二数据。
在一些实施例，关于触发信息的详细说明，参见上述图2所示方法200的步骤205，图4所示方法400的步骤403，图5所示方法500的步骤506，以及图6所示方法600的步骤606中的相关内容，在此不再详细说明。
在一些实施例中，所述处理单元801，用于：
基于所述第一数据集获取所述至少一个数据特征；
基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
在一些实施例中，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
在一些实施例中，所述处理单元801，还用于：
基于所述数据指标确定所述第二数据集中的第二数据的异常情况；
发送所述确定的异常情况和/或所述数据指标。
在一些实施例中，处理单元801确定异常情况的详细实现过程，参见上述图4所示方法400的步骤404，图5所示方法500的步骤507，以及图6所示方法600的步骤607中的相关内容，在此不再详细说明。
在一些实施例中，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括所述网元在获取所述第一数据集之后获取的数据。
在本申请实施例中，处理单元获取第一数据集，所述第一数据集包括获取的多个第一数据。基于所述第一数据集获取至少一个特征类型，所述至少一个特征类型与所述多个数据对应的至少一个数据特征相对应。由于所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应，而所述至少一个特征类型与所述至少一个数据特征相对应，至少一个特征类型是基于所述至少一个数据特征获取的。即使因为网络环境变化导致所述装置采集的数据变化，处理单元可以分析出第一数据集中的数据行为发生了变化，通过第一数据集中数据的至少一个数据特征获取相对应的特征类型，处理单元基于所述至少一个特征类型和第二数据集获取的数据指标就可以准确的反映网络情况。所以处理单元通过本申请实施例降低了分析第二数据集中的数据的难度，提高对第二数据集中的数据分析的准确度。
参见图9，本申请提供了一种处理数据的系统900，所述系统900包括：
数据获取单元901，用于获取第一数据集，所述第一数据集包括多个第一数据；
数据发送单元902，用于发送所述第一数据集；
数据处理单元903，用于基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征；
信息发送单元904，用于发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型是所述数据处理单元903基于所述至少一个数据特征获取的；
类型获取单元905，用于基于所述触发信息获取所述至少一个特征类型；
指标获取单元906，用于基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括多个第二数据。
所述第二数据集包括的多个第二数据是所述数据获取单元901获取的或者是所述指标获取单元906获取的。
在一些实施例中，所述数据获取单元901、数据发送单元902、数据处理单元903、信息发送单元904、类型获取单元905和指标获取单元906中的部分或全部的单元部署在第一设备和/或网元上。例如，所述数据获取单元901、数据发送单元902、类型获取单元905和指标获取单元906部署在网元上，所述数据处理单元903和信息发送单元904部署在第一设备上。或者，所述数据获取单元901、数据发送单元902、数据处理单元903、信息发送单元904、类型获取单元905和/或指标获取单元906部署在网元上。或者，所述数据获取单元901、数据发送单元902、数据处理单元903、信息发送单元904、类型获取单元905和/或指标获取单元906部署在第一设备上。
在一些实施例，所述第一设备可以是上述任意实施例提供的第一设备，例如可以是图1所示网络架构100中的第一设备101，或图2所示方法200中的第一设备，或图5所示方法500中的第一设备，或图6所示方法600中的第一设备，或图7所示装置700。
在一些实施例，所述网元可以是上述任意实施例提供的网元，例如可以是图1所示网络架构100中的网元102上，或图4所示方法400中的网元，或图5所示方法500中的网元或图6所示方法600中的路由器，或图8所示装置800。
在一些实施例，数据获取单元901获取第一数据集的详细实现过程，参见上述图2所示方法200的步骤201-203，图4所示方法400的步骤401，图5所示方法500的步骤501或步骤504，以及图6所示方法600的步骤601或步骤604中的相关内容，在此不再详细说明。
在一些实施例，数据处理单元903获取所述至少一个数据特征的详细实现过程，参见上述图2所示方法200的步骤204，图4所示方法400的步骤404，图5所示方法500的步骤505或步骤507，以及图6所示方法600的步骤605或步骤607中的相关内容，在此不再详细说明。
在一些实施例，类型获取单元905获取所述至少一个特征类型的详细实现过程，参见上述图2所示方法200的步骤204，图4所示方法400的步骤404，图5所示方法500的步骤505或步骤507，以及图6所示方法600的步骤605或步骤607中的相关内容，在此不再详细说明。
在一些实施例，指标获取单元906获取数据指标的详细实现过程，参见上述图4所示方法400的步骤404，图5所示方法500的步骤507，以及图6所示方法600的步骤607中的相关内容，在此不再详细说明。
在一些实施例，所述数据处理单元903，还用于基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
在一些实施例，所述数据处理单元903，用于基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
在一些实施例，所述触发信息包括所述至少一个数据特征，
所述类型获取单元905，用于基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
在一些实施例，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，
所述类型获取单元905，用于基于所述至少一个数据特征、所述至少一个数据类型和所述第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
在一些实施例，关于触发信息的详细说明，参见上述图2所示方法200的步骤205，图4所示方法400的步骤403，图5所示方法500的步骤506，以及图6所示方法600的步骤606中的相关内容，在此不再详细说明。
在一些实施例，所述触发信息还包括至少一个第一对象的对象标识，所述第二数据集包括所述至少一个第一对象获取的多个第二数据。
在一些实施例，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
在一些实施例，所述指标获取单元906，还用于基于所述数据指标确定所述第二数据集中的第二数据的异常情况；
所述数据发送单元902，还用于发送所述确定的异常情况和/或所述数据指标。
在一些实施例中，指标获取单元906确定异常情况的详细实现过程，参见上述图4所示方法400的步骤404，图5所示方法500的步骤507，以及图6所示方法600的步骤607中的相关内容，在此不再详细说明。
在一些实施例，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括在获取所述第一数据集之后获取的数据。
在本申请实施例中，数据获取单元获取第一数据集，所述第一数据集包括获取的多个第一数据。数据处理单元基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征，使得所述至少一个数据特征与所述第一数据集中的第一数据的数据行为相对应，而至少一个特征类型是所述数据处理单元基于所述至少一个数据特征获取的。即使因为网络环境变化导致获取的数据变化，通过上述系统，可以分析出第一数据集中的数据行为发生了变化，所述类型获取单元通过第一数据集中数据的至少一个数据特征获取相对应的特征类型，这样指标获取单元基于所述至少一个特征类型和第二数据集获取的数据指标就可以准确的反映网络情况。指标获取单元通过上述系统降低了分析第二数据集中的数据的难度，提高对第二数据集中的数据进行分析的准确度。
在一些实施例中，处理数据的系统900中的各个模块可以部署在同一个物理设备中；在另一些实施例中，处理数据的系统900中的各个模块可以部署在多台不同的物理设备中。处理数据的系统900中的各个模块可以是硬件模块或者软件和硬件相结合的模块。
参见图10，本申请实施例提供了一种处理数据的设备1000示意图。所述设备1000可以是上述任意实施例提供的第一设备，例如，可以是图1所示网络架构中的第一设备101，或 图2所示方法200中的第一设备，或图5所示方法500中的第一设备或图6所示方法600中的第一设备。所述设备1000包括至少一个处理器1001，内部连接1002，存储器1003以及至少一个网络接口1004。
所述设备1000是一种硬件结构的装置。
在一些实施例中，可以用于实现图7所述的装置700中的功能模块。例如，本领域技术人员可以想到图7所示的装置700中的处理单元702可以通过该至少一个处理器1001调用存储器1003中的代码来实现，图7所示的装置700中的接收单元701和发送单元703可以通过该至少一个网络接口1004来实现。或者，
在一些实施例中，存储器1003用于存放程序模块和数据。所述程序模块包括接收模块10031、处理模块10032和发送模块10033。在一些实施例中，图10中存储器1003中的各个模块分别和图7所示的各个模块相对应，处理器1001通过执行存储器1003中的各个模块中的计算机可读指令，能够执行图7所示的各个模块所能够执行的操作。
所述设备1000还可以用于实现上述任一实施例中第一设备的功能。
上述处理器1001例如是通用中央处理器(Central Processing Unit，CPU)、数字信号处理器(Digital Signal Processor，DSP)、网络处理器(Network Processer，NP)、图形处理器(Graphics Processing Unit，GPU)、神经网络处理器(Neural-network Processing Units，NPU)、数据处理单元(Data Processing Unit，DPU)、微处理器或者一个或多个用于实现本申请方案的集成电路。例如，处理器701包括专用集成电路(Application-specific Integrated Circuit，ASIC)，可编程逻辑器件(Programmable Logic Device，PLD)或者其他可编程逻辑器件、晶体管逻辑器件、硬件部件或者其任意组合。PLD例如是复杂可编程逻辑器件(Complex Programmable Logic Device，CPLD)、现场可编程逻辑门阵列(Field-programmable Gate Array，FPGA)、通用阵列逻辑(Generic Array Logic，GAL)或其任意组合。其可以实现或执行结合本申请实施例公开内容所描述的各种逻辑方框、模块和电路。所述处理器也可以是实现计算功能的组合，例如包括一个或多个微处理器组合，DSP和微处理器的组合等等。
上述内部连接1002可包括一通路，在上述组件之间传送信息。内部连接1002可以为单板或总线等。总线可以是外设部件互连标准(Peripheral Component Interconnect，PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture，EISA)总线等。总线可以分为地址总线、数据总线、控制总线等。为便于表示，图10中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
上述至少一个网络接口1004使用任何收发器一类的装置，用于与其它设备或通信网络通信，通信网络可以为以太网、无线接入网或无线局域网(Wireless Local Area Networks，WLAN)等。网络接口1004可以包括有线通信接口，还可以包括无线通信接口。具体的，网络接口1004可以为以太接口、快速以太(Fast Ethernet，FE)接口、千兆以太(Gigabit Ethernet，GE)接口，异步传输模式(Asynchronous Transfer Mode，ATM)接口，无线局域网WLAN接口，蜂窝网络通信接口或其组合。以太网接口可以是光接口，电接口或其组合。在本申请实施例中，网络接口1004可以用于所述设备1000与其他设备进行通信。
上述存储器1003可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically  erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过总线与处理器相连接。存储器1003也可以和处理器1001集成在一起。
在具体实现中，作为一种实施例，处理器1001可以包括一个或多个CPU，例如图10中的CPU0和CPU1。这些CPU中的每一个可以是一个单核处理器，也可以是一个多核处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，所述设备1000可以包括多个处理器，例如图10中的处理器1001和处理器1007。这些处理器中的每一个可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，所述设备1000还可以包括输出设备和输入设备。输出设备和处理器1001通信，可以以多种方式来显示信息。例如，输出设备可以是液晶显示器(Liquid Crystal Display，LCD)、发光二级管(Light Emitting Diode，LED)显示设备、阴极射线管(Cathode Ray Tube，CRT)显示设备或投影仪等。输入设备和处理器1001通信，可以以多种方式接收用户的输入。例如，输入设备可以是鼠标、键盘、触摸屏设备或传感设备等。
在具体实施例中，本申请实施例的所述设备1000可对应于上述多个实施例，例如与图1、图2、图5和图6对应的多个实施例中的第一设备，所述设备1000中的处理器1001读取存储器1003中的指令，使图10所示的设备1000能够执行上述多个实施例中第一设备的全部或部分操作。
参见图11，本申请实施例提供了一种处理数据的设备1100示意图。所述设备1100可以是上述任意实施例提供的网元，例如，可以是图1所示网络架构中的网元102，或图4所示方法400中的网元，或图5所示方法500中的网元或图6所示方法600中的网元。所述设备1100包括至少一个处理器1101，内部连接1102，存储器1103以及至少一个网络接口1104。
所述设备1100是一种硬件结构的装置。
在一些实施例中，可以用于实现图8所述的装置800中的功能模块。例如，本领域技术人员可以想到图8所示的装置800中的处理单元801可以通过该至少一个处理器1101调用存储器1103中的代码来实现，图8所示的装置800中的发送单元802和接收单元803可以通过该至少一个网络接口1104来实现。或者，
在一些实施例中，存储器1103用于存放程序模块和数据。所述程序模块包括处理模块11031、发送模块11032和接收模块11033。在一些实施例中，图11中存储器1103中的各个模块分别和图8所示的各个模块相对应，处理器1101通过执行存储器1103中的各个模块中的计算机可读指令，能够执行图8所示的各个模块所能够执行的操作。
所述设备1100还可以用于实现上述任一实施例中网元的功能。
上述处理器1101例如是通用中央处理器(Central Processing Unit，CPU)、数字信号处理 器(Digital Signal Processor，DSP)、网络处理器(Network Processer，NP)、图形处理器(Graphics Processing Unit，GPU)、神经网络处理器(Neural-network Processing Units，NPU)、数据处理单元(Data Processing Unit，DPU)、微处理器或者一个或多个用于实现本申请方案的集成电路。例如，处理器701包括专用集成电路(Application-specific Integrated Circuit，ASIC)，可编程逻辑器件(Programmable Logic Device，PLD)或者其他可编程逻辑器件、晶体管逻辑器件、硬件部件或者其任意组合。PLD例如是复杂可编程逻辑器件(Complex Programmable Logic Device，CPLD)、现场可编程逻辑门阵列(Field-programmable Gate Array，FPGA)、通用阵列逻辑(Generic Array Logic，GAL)或其任意组合。其可以实现或执行结合本申请实施例公开内容所描述的各种逻辑方框、模块和电路。所述处理器也可以是实现计算功能的组合，例如包括一个或多个微处理器组合，DSP和微处理器的组合等等。
上述内部连接1102可包括一通路，在上述组件之间传送信息。内部连接1102可以为单板或总线等。总线可以是外设部件互连标准(Peripheral Component Interconnect，PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture，EISA)总线等。总线可以分为地址总线、数据总线、控制总线等。为便于表示，图11中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
上述至少一个网络接口1104使用任何收发器一类的装置，用于与其它设备或通信网络通信，通信网络可以为以太网、无线接入网或无线局域网(Wireless Local Area Networks，WLAN)等。网络接口1104可以包括有线通信接口，还可以包括无线通信接口。具体的，网络接口1104可以为以太接口、快速以太(Fast Ethernet，FE)接口、千兆以太(Gigabit Ethernet，GE)接口，异步传输模式(Asynchronous Transfer Mode，ATM)接口，无线局域网WLAN接口，蜂窝网络通信接口或其组合。以太网接口可以是光接口，电接口或其组合。在本申请实施例中，网络接口1104可以用于所述设备1100与其他设备进行通信。
上述存储器1103可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过总线与处理器相连接。存储器1103也可以和处理器1101集成在一起。
在具体实现中，作为一种实施例，处理器1101可以包括一个或多个CPU，例如图11中的CPU0和CPU1。这些CPU中的每一个可以是一个单核处理器，也可以是一个多核处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，所述设备1100可以包括多个处理器，例如图11中的处理器1101和处理器1107。这些处理器中的每一个可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，所述设备1100还可以包括输出设备和输入设备。输出设备和处理器1101通信，可以以多种方式来显示信息。例如，输出设备可以是液晶显示器(Liquid Crystal Display，LCD)、发光二级管(Light Emitting Diode，LED)显示设备、阴极射线管(Cathode Ray Tube，CRT)显示设备或投影仪等。输入设备和处理器1101通信，可以以多种方式接收用户的输入。例如，输入设备可以是鼠标、键盘、触摸屏设备或传感设备等。
在具体实施例中，本申请实施例的所述设备1100可对应于上述多个实施例，例如与图1、图4、图5和图6对应的多个实施例中的网元，所述设备1100中的处理器1101读取存储器1103中的指令，使图11所示的设备1100能够执行上述多个实施例中网元的全部或部分操作。
本申请的说明书和权利要求书及上述附图中的操作顺序，不限于描述中特定的顺序或先后次序。应该理解这样使用的数据在适当情况下同时进行或可以改变顺序，以便描述的实施例能够以除了在附图中的图示或描述的内容以外的顺序实施。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
以上所述仅为本申请的可选实施例，并不用以限制本申请，凡在本申请的原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (46)

  1. 一种处理数据的方法，其特征在于，所述方法包括：
    接收网元发送的第一数据集，所述第一数据集包括所述网元获取的多个第一数据；
    基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征；
    向所述网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型与所述至少一个数据特征有关。
  2. 如权利要求1所述的方法，其特征在于，所述方法还包括：
    基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  3. 如权利要求2所述的方法，其特征在于，所述基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，包括：
    基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和/或所述至少一个特征类型。
  4. 如权利要求1至3任一项所述的方法，其特征在于，所述触发信息包括所述至少一个数据特征和/或至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型。
  5. 如权利要求1至4任一项所述的方法，其特征在于，所述触发信息还包括至少一个第一对象的对象标识，所述至少一个第一对象位于所述网元中。
  6. 如权利要求1至5任一项所述的方法，其特征在于，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；
    其中，所述至少一个第二对象包括获取所述第一数据集中的各第一数据的对象。
  7. 如权利要求1至6任一项所述的方法，其特征在于，所述方法的执行主体包括：数据处理系统、控制器或管理设备。
  8. 如权利要求1至7任一项所述的方法，其特征在于，
    所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，
    所述第二数据集包括所述网元在发送所述第一数据集之后获取的数据。
  9. 一种处理数据的方法，其特征在于，所述方法包括：
    获取第一数据集，所述第一数据集包括多个第一数据；
    基于所述第一数据集获取至少一个特征类型，所述至少一个特征类型与所述多个第一数据对应的至少一个数据特征相对应；
    基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括多个第二数据。
  10. 如权利要求9所述的方法，其特征在于，所述基于所述第一数据集获取至少一个特征类型，包括：
    发送所述第一数据集，所述第一数据集用于所述第一数据集的接收方获取所述至少一个数据特征和/或所述至少一个特征类型；
    接收触发信息，所述触发信息包括所述至少一个数据特征和/或所述至少一个特征类型；
    基于所述触发信息获取所述至少一个特征类型。
  11. 如权利要求10所述的方法，其特征在于，所述触发信息包括所述至少一个数据特征，所述基于所述触发信息获取所述至少一个特征类型，包括：
    基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  12. 如权利要求11所述的方法，其特征在于，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型；
    所述基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，包括：
    基于所述至少一个数据特征、所述至少一个数据类型和所述第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
  13. 如权利要求10至12任一项所述的方法，其特征在于，所述触发信息还包括至少一个第一对象的对象标识；
    所述基于所述至少一个特征类型和第二数据集获取数据指标之前，还包括：
    基于所述至少一个第一对象的对象标识获取所述第二数据集，所述第二数据集包括所述至少一个第一对象获取的多个第二数据。
  14. 如权利要求9所述的方法，其特征在于，所述基于所述第一数据集获取至少一个特征类型，包括：
    基于所述第一数据集获取所述至少一个数据特征；
    基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  15. 如权利要求9至14任一项所述的方法，其特征在于，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中 的数据内容特征和至少一个第二对象的对象标识；
    其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
  16. 如权利要求9至15任一项所述的方法，其特征在于，所述方法还包括：
    基于所述数据指标确定所述第二数据集中的第二数据的异常情况；
    发送所述确定的异常情况和/或所述数据指标。
  17. 如权利要求9至16任一项所述的方法，其特征在于，
    所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，
    所述第二数据集包括所述网元在获取所述第一数据集之后获取的数据。
  18. 一种处理数据的系统，其特征在于，所述系统包括：
    数据获取单元，用于获取第一数据集，所述第一数据集包括多个第一数据；
    数据发送单元，用于发送所述第一数据集；
    数据处理单元，用于基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征；
    信息发送单元，用于发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型是所述数据处理单元基于所述至少一个数据特征获取的；
    类型获取单元，用于基于所述触发信息获取所述至少一个特征类型；
    指标获取单元，用于基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括多个第二数据。
  19. 如权利要求18所述的系统，其特征在于，所述数据处理单元，还用于基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  20. 如权利要求19所述的系统，其特征在于，所述数据处理单元，用于基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
  21. 如权利要求18所述的系统，其特征在于，所述触发信息包括所述至少一个数据特征，
    所述类型获取单元，用于基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  22. 如权利要求21所述的系统，其特征在于，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，
    所述类型获取单元，用于基于所述至少一个数据特征、所述至少一个数据类型和所述第 一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
  23. 如权利要求18至22任一项所述的系统，其特征在于，所述触发信息还包括至少一个第一对象的对象标识，所述第二数据集包括所述至少一个第一对象获取的多个第二数据。
  24. 如权利要求18至23任一项所述的系统，其特征在于，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；
    其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
  25. 如权利要求18至24任一项所述的系统，其特征在于，
    所述指标获取单元，还用于基于所述数据指标确定所述第二数据集中的第二数据的异常情况；
    所述数据发送单元，还用于发送所述确定的异常情况和/或所述数据指标。
  26. 如权利要求18至25任一项所述的系统，其特征在于，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括在获取所述第一数据集之后获取的数据。
  27. 一种处理数据的装置，其特征在于，所述装置包括：
    接收单元，用于接收网元发送的第一数据集，所述第一数据集包括所述网元获取的多个第一数据；
    处理单元，用于基于所述第一数据集获取所述多个第一数据对应的至少一个数据特征；
    发送单元，用于向所述网元发送触发信息，所述触发信息包括所述至少一个数据特征和/或至少一个特征类型，所述至少一个特征类型与所述至少一个数据特征有关。
  28. 如权利要求27所述的装置，其特征在于，所述处理单元，还用于：
    基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  29. 如权利要求28所述的装置，其特征在于，所述处理单元，用于：
    基于所述至少一个数据特征、所述第一对应关系和至少一个数据类型，获取所述至少一个特征类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和/或所述至少一个特征类型。
  30. 如权利要求27至29任一项所述的装置，其特征在于，所述触发信息包括所述至少一个数据特征和/或至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数 据属于的类型。
  31. 如权利要求27至30任一项所述的装置，其特征在于，所述触发信息还包括至少一个第一对象的对象标识，所述至少一个第一对象位于所述网元中。
  32. 如权利要求27至31任一项所述的装置，其特征在于，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；
    其中，所述至少一个第二对象包括获取所述第一数据集中的各第一数据的对象。
  33. 如权利要求27至32所述的装置，其特征在于，所述装置包括：数据处理系统、控制器或管理设备。
  34. 如权利要求27至33任一项所述的装置，其特征在于，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括所述网元在发送所述第一数据集之后获取的数据。
  35. 一种处理数据的装置，其特征在于，所述装置包括：
    处理单元，用于获取第一数据集，所述第一数据集包括多个第一数据；
    所述处理单元，还用于基于所述第一数据集获取至少一个特征类型，所述至少一个特征类型与所述多个数据对应的至少一个数据特征相对应；
    所述处理单元，还用于基于所述至少一个特征类型和第二数据集获取数据指标，所述第二数据集包括多个第二数据。
  36. 如权利要求35所述的装置，其特征在于，所述装置还包括：发送单元和接收单元；
    所述发送单元，用于发送所述第一数据集，所述第一数据集用于获取所述至少一个数据特征和/或所述至少一个特征类型；
    所述接收单元，用于接收触发信息，所述触发信息包括所述至少一个数据特征和/或所述至少一个特征类型；
    所述处理单元，用于基于所述触发信息获取所述至少一个特征类型。
  37. 如权利要求36所述的装置，其特征在于，所述触发信息包括所述至少一个数据特征，所述处理单元，用于：
    基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  38. 如权利要求37所述的装置，其特征在于，所述触发信息还包括至少一个数据类型，所述至少一个数据类型包括所述第一数据集中的第一数据属于的类型；
    所述处理单元，用于：
    基于所述至少一个数据特征、所述至少一个数据类型和所述第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征、所述至少一个数据类型和所述至少一个特征类型。
  39. 如权利要求36至38任一项所述的装置，其特征在于，所述触发信息还包括至少一个第一对象的对象标识；
    所述处理单元，还用于：
    基于所述至少一个第一对象的对象标识获取所述第二数据集，所述第二数据集包括所述至少一个第一对象获取的多个第二数据。
  40. 如权利要求35所述的装置，其特征在于，所述处理单元，用于：
    基于所述第一数据集获取所述至少一个数据特征；
    基于所述至少一个数据特征和第一对应关系，获取所述至少一个特征类型，所述第一对应关系包括所述至少一个数据特征和所述至少一个特征类型。
  41. 如权利要求35至40任一项所述的装置，其特征在于，所述至少一个数据特征包括如下一个或多个信息：所述第一数据集中的第一数据对应的数据波形特征，所述第一数据集中的数据内容特征和至少一个第二对象的对象标识；
    其中，所述至少一个第二对象包括获取所述第一数据集中的第一数据的对象。
  42. 如权利要求35至41任一项所述的装置，其特征在于，所述处理单元，还用于：
    基于所述数据指标确定所述第二数据集中的第二数据的异常情况；
    发送所述确定的异常情况和/或所述数据指标。
  43. 如权利要求35至42任一项所述的装置，其特征在于，所述第一数据集包括所述第二数据集或所述第二数据集中的部分数据，或者，所述第二数据集包括所述网元在获取所述第一数据集之后获取的数据。
  44. 一种处理数据的设备，其特征在于，包括处理器及计算机程序，所述处理器执行所述计算机程序时，使得所述设备实现如权利要求1-17任一项所述的方法。
  45. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被计算机执行时，实现如权利要求1-17任一项所述的方法。
  46. 一种计算机程序产品，其特征在于，包括计算机程序，所述计算机程序被计算机执行时，实现如权利要求1-17任一项所述的方法。
PCT/CN2021/134273 2021-01-28 2021-11-30 处理数据的方法、装置、系统及存储介质 WO2022160916A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21922506.7A EP4277223A4 (en) 2021-01-28 2021-11-30 DATA PROCESSING METHOD, APPARATUS AND SYSTEM AND STORAGE MEDIUM
US18/360,899 US20230370318A1 (en) 2021-01-28 2023-07-28 Data Processing Method, Apparatus, and System, and Storage Medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110119289.9 2021-01-28
CN202110119289.9A CN114826867B (zh) 2021-01-28 2021-01-28 处理数据的方法、装置、系统及存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/360,899 Continuation US20230370318A1 (en) 2021-01-28 2023-07-28 Data Processing Method, Apparatus, and System, and Storage Medium

Publications (1)

Publication Number Publication Date
WO2022160916A1 true WO2022160916A1 (zh) 2022-08-04

Family

ID=82525795

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/134273 WO2022160916A1 (zh) 2021-01-28 2021-11-30 处理数据的方法、装置、系统及存储介质

Country Status (4)

Country Link
US (1) US20230370318A1 (zh)
EP (1) EP4277223A4 (zh)
CN (1) CN114826867B (zh)
WO (1) WO2022160916A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109600790A (zh) * 2017-09-30 2019-04-09 华为技术有限公司 获取特征参数的方法和装置
CN109600243A (zh) * 2017-09-30 2019-04-09 华为技术有限公司 数据分析方法和装置
CN110019511A (zh) * 2017-09-30 2019-07-16 华为技术有限公司 通信方法及其通信设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110300008B (zh) * 2018-03-22 2021-03-23 北京华为数字技术有限公司 一种确定网络设备的状态的方法及装置
CN111901135B (zh) * 2019-05-05 2021-10-15 华为技术有限公司 一种数据分析方法及装置
CN110502398B (zh) * 2019-08-21 2022-03-29 吉林吉大通信设计院股份有限公司 一种基于人工智能的交换机故障预测系统及方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109600790A (zh) * 2017-09-30 2019-04-09 华为技术有限公司 获取特征参数的方法和装置
CN109600243A (zh) * 2017-09-30 2019-04-09 华为技术有限公司 数据分析方法和装置
CN110019511A (zh) * 2017-09-30 2019-07-16 华为技术有限公司 通信方法及其通信设备
US20200228413A1 (en) * 2017-09-30 2020-07-16 Huawei Technologies Co., Ltd. Data Analytics Method And Apparatus

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4277223A4

Also Published As

Publication number Publication date
CN114826867B (zh) 2023-11-17
CN114826867A (zh) 2022-07-29
EP4277223A4 (en) 2024-06-12
US20230370318A1 (en) 2023-11-16
EP4277223A1 (en) 2023-11-15

Similar Documents

Publication Publication Date Title
US10594582B2 (en) Introspection driven monitoring of multi-container applications
US11627149B2 (en) Security monitoring of network connections using metrics data
US10523540B2 (en) Display method of exchanging messages among users in a group
US10389596B2 (en) Discovering application topologies
US20200287927A1 (en) Anomaly detection based on changes in an entity relationship graph
US11095690B2 (en) Threat identification-based collection of forensic data from endpoint devices
US11799728B2 (en) Multistage device clustering
US10644962B2 (en) Continuous monitoring for performance evaluation of service interfaces
US8730819B2 (en) Flexible network measurement
US10257678B2 (en) Scalable data discovery in an internet of things (IoT) system
CN108833202B (zh) 故障链路检测方法、装置和计算机可读存储介质
US10693898B2 (en) Systems data validation
US10091358B1 (en) Graphical user interface for call center analysis
US11172065B1 (en) Monitoring framework
US20200329011A1 (en) Firewall rules intelligence
US20150134591A1 (en) Method, apparatus, and computer program product for data quality analysis
US11704313B1 (en) Parallel branch operation using intermediary nodes
US20180091529A1 (en) Correlating forensic data collected from endpoint devices with other non-forensic data
US11093476B1 (en) HTTP events with custom fields
CN109672558B (zh) 一种面向第三方服务资源的聚合与优化匹配方法，设备及存储介质
CN110659109A (zh) 一种openstack集群虚拟机监控系统及方法
CN117176802B (zh) 一种业务请求的全链路监控方法、装置、电子设备及介质
CN113360353A (zh) 一种测试服务器和云平台
WO2022160916A1 (zh) 处理数据的方法、装置、系统及存储介质
US11843515B2 (en) Peer risk benchmarking using generative adversarial 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: 21922506

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021922506

Country of ref document: EP

Effective date: 20230809

NENP Non-entry into the national phase

Ref country code: DE