Metering information data intelligent analysis method and system based on big data analysis

Abstract

The invention discloses a metering information data intelligent analysis method and system based on big data analysis, and belongs to the technical field of metering data analysis. Deploying distributed network probes and dividing network sections to form a dynamic monitoring range relationship; establishing a network virtual platform, and generating a reference and real-time parameter set through solid-state and dynamic mapping; parameters are analyzed according to a fixed period, and an error cost function is constructed to evaluate the metering executable degree; and evaluating association stability based on the performability, and judging whether to replace the mapping relation. According to the invention, dynamic monitoring and intelligent evaluation of metering information are realized, monitoring flexibility and evaluation accuracy are improved, and the problems of low metering efficiency, monitoring interruption caused by probe faults in traditional fixed mapping and poor adaptability in the prior art are solved. The method is suitable for network accurate metering and intelligent switching decision in scenes such as computer architecture application, and robustness and data continuity of network state monitoring can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of metering data analysis, in particular to a metering information data intelligent analysis method and system based on big data analysis. BACKGROUND

[0002] In the technical field of metering data analysis, real-time state monitoring of computer networks relies on distributed probe collection devices to collect device operating parameters (such as bandwidth utilization, response time, packet loss rate, etc.), which is the core support for ensuring stable operation of the network. The existing technology generally uses a "fixed mapping" mechanism: the probe is pre-bound to a network segment (such as a probe that monitors a specific subnet), but this mechanism has significant defects in actual application:

[0003] Single point failure diffusion risk: when a probe is down due to hardware failure or network interruption, the network segment bound to the probe will immediately fall into a monitoring blind area. For example, a probe in a financial transaction system is responsible for monitoring the core transaction subnet, and after it is down, real-time bandwidth and transaction response data of the subnet will be lost, which may cause transaction delays or system misjudgments;

[0004] Response delay and human dependence: after the probe fails, manual intervention is required to redistribute the monitoring tasks, and the average response time is more than 30 minutes. In a high-load network (such as an e-commerce platform during peak hours), long monitoring interruptions may cause abnormal traffic to be discovered in time, resulting in economic losses;

[0005] Resource allocation imbalance: fixed mapping cannot dynamically adjust the probe coverage, resulting in "uneven busy and idle" - some probes are overloaded due to covering high-load segments, while the other standby probes are idle, reducing the overall monitoring efficiency;

[0006] Loss of data continuity: during manual switching, the interruption in the collection of key parameters (such as the response speed of service agencies on a technical transaction platform) causes "time gaps" in metering data, affecting the accuracy of subsequent data analysis (such as deviations in service agency capacity assessment).

[0007] Therefore, there is an urgent need for an intelligent analysis technology that can automatically switch monitoring relationships and dynamically evaluate metering reliability when a probe fails. SUMMARY

[0008] The present application aims to provide a metering information data intelligent analysis method and system based on big data analysis to solve the problems raised in the background art. Specifically, the core of the present application is a three-layer collaborative mechanism of "dynamic mapping-quantitative evaluation-intelligent decision-making", which includes:

[0009] The dynamic mapping layer realizes the elastic adjustment of the monitoring relationship through "solid-state mapping + dynamic mapping"; the solid-state mapping is used as an initialization reference to pre-bind the probe and the network segment (such as "segment 1→probe 1"); when the probe fails, the dynamic mapping automatically calls the standby probe (such as "segment 1→standby probe 3"), so as to avoid the monitoring blind area and solve the single-point failure diffusion problem.

[0010] The quantitative evaluation layer samples the running parameters (such as response time and packet loss rate) at a fixed period (such as 1 day), calculates the deviation of the reference parameters (generated by the solid-state mapping) and the real-time parameters (generated by the dynamic mapping) through a parameter boundary error cost function, and evaluates the "measurable degree" of the dynamic monitoring relationship (the data reliability after the probe switching to avoid the overload of the probe in the high-load segment).

[0011] The intelligent decision-making layer calculates the "correlation stability" (reflecting the long-term reliability of the dynamic mapping) based on the measurable degree, automatically replaces the solid-state mapping when the stability reaches a preset threshold (such as 0.7), and ensures the continuous and reliable execution of the measurement task to avoid the delay caused by manual intervention.

[0012] To solve the above technical problems, the present application provides the following technical solutions:

[0013] The measurement information data intelligent analysis system based on big data analysis comprises a monitoring node deployment module, a virtual platform building module, a parameter analysis evaluation module and a correlation stability evaluation module.

[0014] The monitoring node deployment module is used to deploy distributed network probes in the computer network, divide the network segments and form the dynamic monitoring range relationship.

[0015] The virtual platform building module is used to build a network virtual platform, perform solid-state mapping and dynamic mapping on the network segments and the distributed measurement nodes, and generate a reference running parameter set and a real-time measurement parameter set.

[0016] The parameter analysis evaluation module is used to divide the sampling time nodes at a fixed period, evaluate the measurable degree of the dynamic monitoring range relationship after classifying the running parameters.

[0017] The correlation stability evaluation module evaluates the correlation stability of the dynamic monitoring range relationship based on the measurable degree, and judges whether to replace the solid-state mapping relationship.

[0018] Further, the monitoring node deployment module comprises a probe deployment unit and a segment division unit.

[0019] The probe deployment unit is used to deploy distributed network probes in the computer network, form the distributed measurement nodes to monitor the running state of the network equipment.

[0020] The segment division unit divides network segments based on network probe coverage, so that the distributed metering nodes form a dynamic monitoring range relationship with the network segments.

[0021] Further, the virtual platform building module includes a mapping relationship construction unit and a parameter set generation unit.

[0022] The mapping relationship construction unit is used to perform solid-state mapping, i.e., initialization mapping, and dynamic mapping, i.e., mapping based on real-time connection state adjustment, of the network segments and the distributed metering nodes.

[0023] The parameter set generation unit generates a reference operation parameter set based on the solid-state mapping relationship and a real-time metering parameter set based on the dynamic mapping relationship.

[0024] Further, the parameter analysis and evaluation module includes a sampling node division unit and an executable degree calculation unit.

[0025] The sampling node division unit is used to divide sampling time nodes at a fixed period and classify operation parameters.

[0026] The executable degree calculation unit is used to construct a parameter boundary error cost function, quantify the parameter boundary error cost, and evaluate the metering executable degree of the dynamic monitoring range relationship.

[0027] Further, the correlation stability evaluation module includes a stability calculation unit and a mapping replacement judgment unit.

[0028] The stability calculation unit evaluates the correlation stability of the dynamic monitoring range relationship based on the metering executable degree.

[0029] The mapping replacement judgment unit is used to compare the correlation stability with a preset threshold to determine whether to replace the solid-state mapping relationship to perform metering.

[0030] The method for intelligent analysis of metering information data based on big data analysis includes the following steps:

[0031] Step S1: Deploy distributed network probes in a computer network, divide network segments, and form a dynamic monitoring range relationship between the distributed metering nodes and the network segments.

[0032] Step S2: Build a network virtual platform for mapping the network segments and the distributed metering nodes, which includes solid-state mapping and dynamic mapping to generate a reference operation parameter set and a real-time metering parameter set.

[0033] Step S3: Divide sampling time nodes at a fixed period, classify operation parameters, and evaluate the metering executable degree of the dynamic monitoring range relationship.

[0034] Step S4: based on the executable degree of measurement, the correlation stability of the dynamic monitoring range relationship is evaluated to determine whether to replace the solid-state mapping relationship to execute the measurement.

[0035] Further, the specific implementation process of step S1 includes:

[0036] Distributed network probes are deployed in the computer network to form distributed measurement nodes, wherein one network probe corresponds to one distributed measurement node, and the distributed measurement node is used to monitor the running state of network devices in the coverage range of the network probe;

[0037] The network section is divided by the network probe coverage corresponding to the distributed measurement node, so that the distributed measurement node and the network section form a dynamic monitoring range relationship, and one distributed measurement node corresponds to at least one network section, and a plurality of network devices run in one network section. The dynamic monitoring range relationship is characterized by the real-time connection state of the network probe in the running process of the computer network.

[0038] Further, the specific implementation process of step S2 includes:

[0039] The solid-state mapping is an initialization mapping, including:

[0040] The network section and the distributed measurement node are initialized respectively, and the network section set {U i |i∈[1,I]} and the distributed measurement node set {V j |j∈[1,J]} are generated in sequence respectively, wherein U i represents the i-th network section, V j represents the j-th distributed measurement node, and I and J represent the total number of network sections and distributed measurement nodes respectively.

[0041] Based on the coverage range of the monitoring network probe, a solid-state mapping relationship U i →V j is formed.

[0042] The dynamic mapping is dynamically mapped based on the real-time connection state of the network probe in the running process of the computer network, including:

[0043] If the network probe corresponding to the j-th distributed measurement node has a connection state of down, the network probe corresponding to the j'-th distributed measurement node is dynamically adjusted to cover the network section U i , and a dynamic monitoring range relationship U i →V j' is formed.

[0044] The operation parameters of the network section formed by the solid mapping relationship simulation form a reference operation parameter set, and the operation parameters of the network section formed by the dynamic monitoring range relationship in real time form a real-time measurement parameter set.

[0045] Further, the specific implementation process of the step S3 includes:

[0046] The time range of a day is divided into a plurality of sampling time nodes with a fixed cycle period of a day; the operation parameters are classified, and based on the classification result, the reference operation parameter set of the network section generated by the rth operation parameter changing with the sampling time node t is recorded as [U i →V j ] r (t), and the real-time measurement parameter set of the network section generated by the rth operation parameter changing with the sampling time node t is recorded as [U i →V j' ] r (t).

[0047] The xth operation parameter in the reference operation parameter set [U i →V j ] r (t) is recorded as The yth operation parameter in the real-time measurement parameter set [U i →V j' ] r (t) is recorded as

[0048] A parameter boundary error cost function is constructed to quantify the parameter boundary error cost E r (x, y):

[0049]

[0050] Based on the parameter boundary error cost function, the dynamic monitoring range relationship U i →V j' is evaluated within a fixed cycle period, and the measurement executability of the rth operation parameter is h is the serial number of the fixed cycle period.

[0051] Further, the specific implementation process of the step S4 includes:

[0052] Within the coverage range of the network probe corresponding to the j'th distributed measurement node, the correlation stability of the dynamic monitoring range relationship is dynamically evaluated based on the replacement of the dynamic monitoring range relationship In the formula, R represents the total number of operation parameters, and H represents the serial number of the currently sampled fixed cycle period.

[0053] A preset correlation stability threshold is set, and if the correlation stability is greater than or equal to the correlation stability threshold, it is determined that the dynamic monitoring range relationship U i → V j' The solid-state mapping relationship U i → V j The j'th distributed metering node is dynamically adjusted to perform metering of the i'th network segment U i .

[0054] Compared with the prior art, the beneficial effects achieved by the present application are: in the metering information data intelligent analysis method and system based on big data analysis provided by the present application, distributed network probes are deployed and network segments are divided to form a dynamic monitoring range relationship; a network virtual platform is built, and a reference and a real-time parameter set are generated through solid-state and dynamic mapping; parameters are analyzed at a fixed period, an error cost function is constructed to evaluate the executable degree of metering; the correlation stability is evaluated based on the executable degree to determine whether to replace the mapping relationship. The present application realizes dynamic monitoring and intelligent evaluation of metering information, improves monitoring flexibility and evaluation accuracy, solves the problems of low metering efficiency, monitoring interruption caused by traditional fixed mapping due to probe failure and poor adaptability in the prior art, and is suitable for network accurate metering and intelligent switching decision in computer architecture application and other scenes, and can improve the robustness and data continuity of network state monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0055] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application.

[0056] Figure 1 is a step schematic diagram of the metering information data intelligent analysis method based on big data analysis of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0058] In the first embodiment of the present application: a metering information data intelligent analysis system based on big data analysis is provided, which comprises: a monitoring node deployment module, a virtual platform building module, a parameter analysis evaluation module and a correlation stability evaluation module;

[0059] The monitoring node deployment module is used to deploy distributed network probes in a computer network, divide network segments and form a dynamic monitoring range relationship;

[0060] The monitoring node deployment module comprises a probe deployment unit and a section division unit.

[0061] The probe deployment unit is configured to deploy distributed network probes in the computer network to form distributed metering nodes for monitoring the running state of the network equipment.

[0062] The section division unit is configured to divide network sections based on the coverage range of the network probes, so that the distributed metering nodes form a dynamic monitoring range relationship with the network sections.

[0063] The virtual platform building module is configured to build a network virtual platform, perform solid-state mapping and dynamic mapping on the network sections and the distributed metering nodes, and generate a benchmark running parameter set and a real-time metering parameter set.

[0064] The virtual platform building module comprises a mapping relationship construction unit and a parameter set generation unit.

[0065] The mapping relationship construction unit is configured to perform solid-state mapping, i.e., initialization mapping, and dynamic mapping, i.e., mapping based on real-time connection state adjustment, on the network sections and the distributed metering nodes.

[0066] The parameter set generation unit is configured to generate the benchmark running parameter set based on the solid-state mapping relationship and generate the real-time metering parameter set based on the dynamic mapping relationship.

[0067] The parameter analysis and evaluation module is configured to divide sampling time nodes at a fixed period, evaluate the metering executability of the dynamic monitoring range relationship after classifying the running parameters, and perform the following operations.

[0068] The parameter analysis and evaluation module comprises a sampling node division unit and an executability calculation unit.

[0069] The sampling node division unit is configured to divide sampling time nodes at a fixed period and classify the running parameters.

[0070] The executability calculation unit is configured to construct a parameter boundary error cost function, quantify the parameter boundary error cost, and evaluate the metering executability of the dynamic monitoring range relationship.

[0071] The correlation stability evaluation module is configured to evaluate the correlation stability of the dynamic monitoring range relationship based on the metering executability and determine whether to replace the solid-state mapping relationship.

[0072] The correlation stability evaluation module comprises a stability calculation unit and a mapping replacement judgment unit.

[0073] The stability calculation unit is configured to evaluate the correlation stability of the dynamic monitoring range relationship based on the metering executability.

[0074] The mapping replacement judging unit is configured to compare the correlation stability with a preset threshold value, and judge whether to replace the solid-state mapping relationship to perform metering.

[0075] Referring to Figure 1 In the second embodiment, a metering information data intelligent analysis method based on big data analysis is provided, which is applicable to the first embodiment. The method comprises the following steps:

[0076] Step S1: deploying a distributed network probe in a computer network, dividing network segments, and forming a dynamic monitoring range relationship between the distributed metering nodes and the network segments.

[0077] For example, a distributed network probe is deployed in a computer network to form a distributed metering node, wherein one network probe corresponds to one distributed metering node, and the distributed metering node is used to monitor the running state of network equipment within the coverage range of the network probe.

[0078] The network segments are divided by the coverage range of the network probe corresponding to the distributed metering node, so that a dynamic monitoring range relationship is formed between the distributed metering node and the network segment. One distributed metering node corresponds to at least one network segment, and a plurality of network equipment runs in one network segment. The dynamic monitoring range relationship is characterized by the real-time connection state of the network probe in the running process of the computer network.

[0079] For example, taking a certain technology transaction platform as an example, the platform contains 5 transaction subnets (each subnet contains 20 servers), and needs to monitor the response time (≤50ms), packet loss rate (≤0.1%) and bandwidth utilization rate (≤80%) of the servers in real time to ensure transaction continuity.

[0080] Ten distributed network probes (numbered V1-V10) are deployed, and each two probes correspond to one transaction subnet (5 network segments U1-U5 in total), forming a dynamic monitoring relationship: U1→V1, U2→V3, U3→V5, U4→V7, U5→V9 (one main monitoring, one standby).

[0081] When V1 (the main probe of U1) is down due to network interruption, the system detects that the connection state is abnormal in real time, automatically switches the monitoring task of U1 to the standby probe V2, and completes the dynamic monitoring relationship update (U1→V2) within 5 seconds to avoid the monitoring blind area.

[0082] Step S2: building a network virtual platform for mapping the network segments and the distributed metering nodes, wherein the mapping includes solid-state mapping and dynamic mapping, to generate a set of reference running parameters and a set of real-time metering parameters.

[0083] The solid-state mapping is an initialization mapping, including:

[0084] The network segments and the distributed measurement nodes are initialized respectively, and the network segment set {U i | i ∈ [1, I]} and the distributed measurement node set {V j | j ∈ [1, J]} are generated in sequence respectively, wherein U i represents the i-th network segment, V j represents the j-th distributed measurement node, and I and J represent the total number of network segments and distributed measurement nodes respectively;

[0085] Based on the coverage of the monitoring network probe, the solid-state mapping relationship U i → V j is formed.

[0086] The dynamic mapping is dynamically mapped based on the real-time connection state of the network probe in the running process of the computer network, including:

[0087] If the network probe corresponding to the j-th distributed measurement node has a connection state of being down, the network probe corresponding to the j'-th distributed measurement node is dynamically adjusted to cover the network segment U i , and the dynamic monitoring range relationship U i → V j' is formed.

[0088] The running parameters of the network segment simulated by the solid-state mapping relationship form a set of baseline running parameters, and the running parameters of the network segment formed in real time by the dynamic monitoring range relationship form a set of real-time measurement parameters.

[0089] For example, the solid-state mapping initializes the binding relationship of U1-U5 and V1-V10, and generates a set of baseline running parameters (for example, the baseline response time of U1 is 40 ms, and the packet loss rate is 0.05%); after the dynamic mapping, V1 is down, V2 collects the real-time parameters of U1, and generates a set of real-time measurement parameters (for example, the real-time response time of U1 is 42 ms, and the packet loss rate is 0.06%), thereby ensuring that the solid-state mapping provides a baseline reference, and the dynamic mapping ensures that data collection is not interrupted when the probe fails.

[0090] Step S3: dividing the sampling time nodes with a fixed period, after classifying the running parameters, evaluating the measurement executability of the dynamic monitoring range relationship;

[0091] For example, the fixed cycle period is a day, and the time range of a day is divided into a plurality of sampling time nodes; the running parameters are classified, and based on the classification result, the set of baseline running parameters of the network segment generated by the r-th type of running parameters changing with the sampling time node t is recorded as [U i → V j ].r (t), the real-time metering parameter set of the network segment generated by the rth operating parameter changing with the sampling time node t is recorded as [U i →V j' ] r (t);

[0092] The xth operating parameter in the reference operating parameter set [U i →V j ] r (t) is recorded as The yth operating parameter in the real-time metering parameter set [U i →V j' ] r (t) is recorded as

[0093] A parameter boundary error cost function is constructed to quantify the parameter boundary error cost E r (x, y):

[0094]

[0095] Based on the parameter boundary error cost function, the dynamic monitoring range relationship U i →V j' is evaluated within a fixed cycle period, and the metering executable degree of the rth operating parameter is h is the sequence number of the fixed cycle period;

[0096] For example, the fixed period is 1 day, 4 sampling nodes (8:00, 12:00, 16:00, 20:00) are divided, and the classification parameters are “response time (r=1)” and “packet loss rate (r=2)”; the error calculation takes the r=1 parameter of U1 as an example, the reference value is 40 ms (x=1), the real-time value is 42 ms (y=1), the parameter boundary error cost E 1 (1, 1) = (40-42)2 = 4; the maximum error E_max = 9 and the minimum error E_min = 1 throughout the day; the metering executable degree C1[U1→V2|1] = (9-1) / (9+1) = 0.8 (i.e. 80%, indicating that the dynamic mapping has high metering reliability).

[0097] Step S4: based on the metering executable degree, the correlation stability of the dynamic monitoring range relationship is evaluated, and it is judged whether to replace the solid mapping relationship to perform metering;

[0098] For example, within the coverage range of the network probe corresponding to the j'th distributed metering node, based on the replacement of the dynamic monitoring range relationship, the correlation stability of the dynamic monitoring range relationship is dynamically evaluated In the formula, R represents the total number of operating parameters, and H represents the fixed cycle period number of the current sampling;

[0099] A preset correlation stability threshold is set, and if the correlation stability is greater than or equal to the correlation stability threshold, it is determined that the dynamic monitoring range relationship U i →V j' The solid-state mapping relationship U can be replaced i →V j The j'th distributed metering node is adjusted dynamically to perform metering of the ith network section U i .

[0100] For example, stability calculation, H=3 (sampling for 3 consecutive days), R=2 (2 types of parameters), the 3-day executable degrees of U1→V2 are 0.8, 0.75 and 0.85 respectively, the correlation stability F=(0.8+0.75+0.85) / (3*2)=0.75; threshold judgment, the preset correlation stability threshold=0.6, since 0.75>0.6, it is determined that the original solid-state mapping U1→V1 can be replaced by U1→V2, and subsequent metering tasks are performed by V2.

[0101] It should be noted that the relational terms herein such as first and second and the like are used only to differentiate one entity or action from another, and do not necessarily require or imply that any such actual relationship or order exists between or among the entities or actions. Also, the terms "comprising", "including", or any other variant thereof are intended to cover a non-exclusive inclusion, such that processes, methods, articles, or apparatuses that comprise a list of elements are not limited to those elements, but can include other elements not expressly listed or inherent to such processes, methods, articles, or apparatuses.

[0102] Finally, it should be noted that the above only describes the preferred embodiments of the present application, and is not intended to limit the present application, although the foregoing embodiments of the present application are described in detail, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for intelligent analysis of metrology information data based on big data analysis, characterized in that, The method comprises the following steps: Step S1: deploying distributed network probes in a computer network, dividing network segments, and forming a dynamic monitoring range relationship between the distributed measurement nodes and the network segments; Step S2: building a network virtual platform for mapping the network segments and the distributed measurement nodes, the mapping comprising solid-state mapping and dynamic mapping to generate a set of baseline operating parameters and a set of real-time measurement parameters; Step S3: dividing sampling time nodes at a fixed cycle, and after classifying the operating parameters, evaluating the measurement executability of the dynamic monitoring range relationship; Step S4: evaluating the correlation stability of the dynamic monitoring range relationship based on the measurement executability, and determining whether to replace the solid-state mapping relationship to perform measurement.

2. The big data analytics based metrology information data intelligent analysis method according to claim 1, characterized by, The specific implementation process of step S1 comprises: Deploying distributed network probes in a computer network to form distributed measurement nodes, wherein one network probe corresponds to one distributed measurement node, and the distributed measurement node is used to monitor the operating state of network devices within the coverage of the network probe; Dividing network segments through the coverage of the network probe corresponding to the distributed measurement node to form a dynamic monitoring range relationship between the distributed measurement node and the network segment, wherein one distributed measurement node corresponds to at least one network segment, and a plurality of network devices operate within one network segment, and the dynamic monitoring range relationship is characterized by the real-time connection state of the network probe during the operation of the computer network. 3.The method of claim 2, wherein, The specific implementation process of step S2 comprises: The solid-state mapping is an initialization mapping, comprising: Initialize network section and distributed metering node respectively, and generate network section set {U i | i ∈ [1, I]} and distributed metering node set {V j | j ∈ [1, J]} in turn respectively, wherein, U i represents the i th network section, V j represents the j th distributed metering node, and I and J represent the total number of network sections and distributed metering nodes respectively; Based on monitoring the coverage of the network probe, a solid mapping relationship U is formed i → V j ; The dynamic mapping is a dynamic mapping based on the real-time connection state of the network probe during the operation of the computer network, comprising: If the network probe corresponding to the j-th distributed metering node experiences a connection failure, then the j-th node will be dynamically adjusted. ' The network probes corresponding to each distributed metering node cover network segment U. i And constitute a dynamic monitoring range relationship U i →V j' ; The operating parameters of the network segment simulated by the solid-state mapping relationship form a set of baseline operating parameters, and the operating parameters of the network segment formed in real time by the dynamic monitoring range relationship form a set of real-time measurement parameters. 4.The method of claim 3, wherein, The specific implementation process of step S3 comprises: divide the time range of a day into a plurality of sampling time nodes with a day as a fixed cycle period; classify the operation parameters, and based on the classification result, record a set of reference operation parameters of the network section generated by the rth type of operation parameters changing with the sampling time node t as [U i →V j ] r (t), record a set of real-time metering parameters of the network section generated by the rth type of operation parameters changing with the sampling time node t as [U i →V j' ] r (t). a set of reference operating parameters [U i → V j ] r the xth operating parameter in (t) is denoted a set of real-time metering parameters [U i → V j' ] r the yth operating parameter in (t) is denoted Constructing parameter boundary error cost function, quantifying parameter boundary error cost E r (x, y): Based on the parameter boundary error cost function, the dynamic monitoring range relationship U is evaluated within a fixed cycle period i → V j' The measurement executable degree for the rth type of operating parameter h is the sequence number of the fixed cycle period. 5.The method of claim 4, wherein, The specific implementation process of step S4 comprises: In the j ' In the coverage range of the network probe corresponding to the jth distributed metering node, the correlation stability of the dynamic monitoring range relationship is dynamically evaluated based on the replacement of the dynamic monitoring range relationship In the formula, R represents the total number of operating parameters, and H represents the current sampling fixed cycle period number. A preset correlation stability threshold is set, and if the correlation stability is greater than or equal to the correlation stability threshold, it is determined that the dynamic monitoring range relationship U i → V j' An alternative solid-state mapping relationship U i → V j , and a j ' th distributed metering node adjusted dynamically is used to perform metering of the i i th network section U .

6. The system for intelligent analysis of metrology information data based on big data analysis according to any one of claims 1 to 5, wherein The system comprises a monitoring node deployment module, a virtual platform building module, a parameter analysis and evaluation module, and a correlation stability evaluation module; The monitoring node deployment module is used to deploy distributed network probes in a computer network, divide network segments, and form a dynamic monitoring range relationship; The virtual platform building module is used to build a network virtual platform, perform solid-state mapping and dynamic mapping on the network segments and the distributed measurement nodes, and generate a set of baseline operating parameters and a set of real-time measurement parameters; The parameter analysis and evaluation module is used to divide sampling time nodes at a fixed cycle, and after classifying the operating parameters, evaluate the measurement executability of the dynamic monitoring range relationship; The correlation stability evaluation module evaluates the correlation stability of the dynamic monitoring range relationship based on the measurement executability, and determines whether to replace the solid-state mapping relationship.

7. The big data analytics based metrology information data intelligent analysis system of claim 6, wherein, The monitoring node deployment module comprises a probe deployment unit and a segment division unit; The probe deployment unit is used to deploy distributed network probes in a computer network to form distributed measurement nodes for monitoring the operating state of network devices; The section division unit divides network sections based on network probe coverage, so that the distributed metering nodes and the network sections form a dynamic monitoring range relationship.

8. The big data analytics based metrology information data intelligent analysis system of claim 6, wherein, The virtual platform building module includes a mapping relationship construction unit and a parameter set generation unit. The mapping relationship construction unit is configured to perform solid-state mapping, i.e., initialization mapping, of the network sections and the distributed metering nodes, and dynamic mapping, i.e., mapping based on real-time connection state adjustment. The parameter set generation unit generates a reference operation parameter set based on the solid-state mapping relationship and generates a real-time metering parameter set based on the dynamic mapping relationship. 9.The big data analytics based metrology information data intelligent analytics system of claim 6, wherein, The parameter analysis and evaluation module includes a sampling node division unit and an executable degree calculation unit. The sampling node division unit is configured to divide sampling time nodes at a fixed period and classify operation parameters. The executable degree calculation unit is configured to construct a parameter boundary error cost function, quantify the parameter boundary error cost, and evaluate the metering executable degree of the dynamic monitoring range relationship. 10.The big data analytics based metrology information data intelligent analytics system of claim 6, wherein, The correlation stability evaluation module includes a stability calculation unit and a mapping replacement judgment unit. The stability calculation unit evaluates the correlation stability of the dynamic monitoring range relationship based on the metering executable degree. The mapping replacement judgment unit is configured to compare the correlation stability with a preset threshold value and judge whether to replace the solid-state mapping relationship to perform metering.

Images