Intelligent modal supervision system and method based on network function virtualization

Abstract

The invention discloses an intelligent modal supervision system and method based on network function virtualization, and relates to the technical field of network supervision, and the supervision method comprises the following steps: analyzing resource use conditions recorded in a plurality of resource configuration records, and obtaining a demand peak period of a certain device; analyzing the resource use condition of each resource configuration record in the resource configuration log of each device, and predicting the resource demand condition of each device; analyzing the resource demand condition of each device, and generating a resource allocation sequence; according to the allocation sequence of the devices, obtaining the predicted waiting time of the resources allocated to the devices; and when a resource configuration record of a certain device is generated in real time, obtaining a resource demand condition predicted by the certain device in an adjacent resource configuration record, analyzing an actual resource demand condition of the certain device, and judging whether the resource allocation condition of the certain device is abnormal or not for reminding.

Description

Technical Field

[0001] The present invention relates to the technical field of network supervision, and in particular to an intelligent modal supervision system and method based on network function virtualization. Background Art

[0002] Network function virtualization is a technology that decouples network functions from dedicated hardware and transfers them to the software level to provide more flexible, scalable and efficient network services. Through network function virtualization, data analysis, artificial intelligence and other technologies can be used to monitor and analyze various network traffic and data in real time.

[0003] When multiple devices require limited resources at the same time, virtualized network functions can help allocate resources, but resource allocation relies heavily on accurately predicting the resource requirements of each device. In actual situations, the resource requirements of each device will be affected by various factors, resulting in deviations in the prediction of resource requirements. In severe cases, it may even affect the operating performance of the entire system. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent modality supervision system and method based on network function virtualization to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent modality supervision method based on network function virtualization, the supervision method comprising the following steps:

[0006] Step S100: constructing an intelligent modal supervision system, storing the resource usage process of any device in each unit cycle, and obtaining a number of resource configuration records of the device; analyzing the resource usage recorded in the number of resource configuration records, and obtaining the peak period of resource demand of the device;

[0007] Step S200: Obtain the peak period of resource demand of each device, analyze the resource usage of each resource configuration record in the resource configuration log of each device, and predict the resource demand of each device in a unit period;

[0008] Step S300: arbitrarily selecting a unit period, analyzing the resource demand of each device, and generating an allocation order for allocating resources to each device; according to the allocation order of each device, obtaining an estimated waiting time for each device to be allocated resources;

[0009] Step S400: When a resource configuration record of a certain device is generated in real time in the intelligent modality supervision system, the resource demand situation predicted by the certain device in the adjacent resource configuration record is obtained, the actual resource demand situation of the certain device is analyzed, and it is determined whether the resource allocation situation of the certain device is abnormal and an alert is issued.

[0010] Furthermore, step S100 includes the following steps:

[0011] Step S101: arbitrarily select the i-th resource configuration record of a certain device, divide a unit period into several unit times evenly, and obtain the resource usage of the j-th unit time in the i-th resource configuration record as

[0012] Step S102: Obtain resource usage for the j+1th unit time The difference in resource usage between the jth unit time and the j+1th unit time is According to the formula: Calculate the value-added range from the jth unit time to the j+1th unit time Resource usage is the amount of resources being used per unit time, so resource usage is the amount of newly added resources used;

[0013] Step S103: Obtain the increment between any two unit times, and calculate the average value to obtain the average increment of the i-th resource configuration record. like Then the j-th unit time is set as the first characteristic time, if The j-th unit time is set as the second characteristic time; the first characteristic time is the starting point of the peak demand period. When a large increase in value occurs, it means that the resource usage in the unit time is relatively large, thus reflecting that the device has entered the peak period; the second characteristic time is the end point of the peak demand period. When the increase in value is less than 0, it means that the device's resource usage is reduced, indicating that the device is beginning to leave the peak demand period;

[0014] Step S104: arbitrarily select two adjacent second characteristic times from the unit period, extract a number of first characteristic times between the two adjacent second characteristic times, sort the several first characteristic times in chronological order, and select the first first characteristic time; select the second second characteristic time from the two second characteristic times to obtain the time period between the first first characteristic time and the second second characteristic time, and set the time period as the peak demand period in the i-th resource configuration record; a plurality of first characteristic times may be included in a first characteristic time and a second characteristic time, because there is a situation where the use of resources by the equipment is greatly increased;

[0015] Step S105: Acquire several demand peak periods in each resource configuration record of the certain device to obtain a demand peak period set of the certain device.

[0016] Further, step S200 includes the following steps:

[0017] Step S201: arbitrarily select a certain device and obtain a peak demand period set in the certain device; arbitrarily select a unit time in a unit cycle, count the number of times the unit time is in the peak demand period in the peak demand period set as M, and set the number of peak demand periods in the peak demand period set as M all , the probability that the unit time is in the peak demand period is calculated to be G = M / M all ;

[0018] Step S202: Set the unit time in a unit cycle as the jth unit time, obtain the peak demand period of the jth unit time in the i-th resource configuration record of the certain device, and obtain the maximum resource usage in the peak demand period The resource demand during a peak demand period should be determined by the maximum value, otherwise it will also lead to insufficient resources;

[0019] Step S203: Obtain the highest resource usage during the peak demand period of the j-th unit time in each resource configuration record, and obtain the average resource demand of the j-th unit time as U j ; The probability of obtaining the jth unit time being in the peak demand period is G j , the estimated resource demand for the j-th unit time is calculated to be R j =U j ×G j ; Select the estimated resource demand with the largest value in a unit period and set it as the estimated resource demand of the device.

[0020] Further, step S300 includes the following steps:

[0021] Step S301: arbitrarily select a device and set the estimated resource demand of the device in the jth unit time in a unit cycle as R j If R j ≠0, the j-th unit time is set as the target time and extracted, and the extracted target times are divided according to the demand peak period set of the certain equipment;

[0022] Step S302: Obtain the number of target times contained in the first demand peak period in the demand peak period set as N1, set the first target time in the first demand peak period as the jth unit time, according to the formula:

[0023]

[0024] Wherein, k is an integer, and k∈(1,s), s is the number of peak demand periods in the peak demand period set of the device, and t k is the length of the time period of the kth peak demand period; the priority value of allocating resources to the device is calculated; the allocation order of resources to each device is generated from high to low according to the priority value; the priority of allocating resources to the device depends on the number of peak demand periods of the device, the time when the resources start to be used in a unit cycle, and the length of time in the peak demand period, which can directly reflect the resource demand;

[0025] Step S303: Set the number of unit time obtained in a unit cycle as P, obtain the number of target time of a certain device in a unit cycle as Q, and calculate the resource usage time of the device in a unit cycle as T use =T×(Q / P), where T is the duration of a unit cycle;

[0026] Step S304: Obtain the resource usage duration of each device in a unit cycle, set the position of the device in the allocation sequence to w, according to the formula:

[0027]

[0028] Where g is an integer and g∈(1,w-1), (T use ) g is the resource usage time of the device with the position order g in a unit cycle; the estimated waiting time t for the resource to be allocated to the device is calculated wait .

[0029] Furthermore, step S400 includes the following steps:

[0030] Step S401: when a resource configuration record of a certain device is generated in real time, an adjacent resource configuration record of the resource configuration record is obtained, and an estimated resource demand of the certain device in the adjacent resource configuration record is obtained as C;

[0031] Step S402: obtaining the actual resource demand of the certain device recorded in the resource configuration record as C', setting a demand error threshold as θ, and if |C-C'| / C>θ, an abnormal reminder is issued to the certain device.

[0032] In order to better implement the above method, an intelligent modality supervision system based on network function virtualization is also proposed. The supervision system includes a historical record analysis module, a resource demand analysis module, a resource allocation module and an abnormality analysis module;

[0033] The historical record analysis module is used to build an intelligent modal supervision system, store the resource usage process of any device in each unit cycle, and obtain several resource configuration records of the device; analyze the resource usage recorded in the several resource configuration records to obtain the peak period of resource demand of the device;

[0034] The resource demand analysis module is used to obtain the peak period of resource demand of each device, analyze the resource usage of each resource configuration record in the resource configuration log of each device, and predict the resource demand of each device in a unit cycle;

[0035] The resource allocation module is used to select a unit period at random, analyze the resource demand of each device, and generate an allocation order for allocating resources to each device; according to the allocation order of each device, obtain the estimated waiting time for each device to be allocated resources;

[0036] The abnormality analysis module is used to obtain the resource demand situation predicted by the device in the adjacent resource configuration record when a resource configuration record of a certain device is generated in real time in the intelligent modal supervision system, analyze the actual resource demand situation of the certain device, and determine whether the resource allocation situation of the certain device is abnormal and prompt.

[0037] Further, the historical record analysis module includes a configuration record building unit and a demand peak acquisition unit;

[0038] The configuration record construction unit is used to construct an intelligent modal supervision system, store the resource usage process of any device in each unit cycle, and obtain a number of resource configuration records of the device; the demand peak acquisition unit is used to analyze the resource usage recorded in the several resource configuration records to obtain the peak period of resource demand of the device.

[0039] Further, the resource demand analysis module includes a resource usage analysis unit and a demand prediction unit;

[0040] The resource usage analysis unit is used to obtain the peak period of resource demand of each device and analyze the resource usage of each resource configuration record in the resource configuration log of each device; the demand prediction unit is used to predict the resource demand of each device in a unit period based on the resource usage of each resource configuration record in the resource configuration log of each device.

[0041] Further, the resource allocation module includes an allocation order setting unit and a waiting time estimation unit;

[0042] The allocation sequence setting unit is used to arbitrarily select a unit period, analyze the resource demand of each device, and generate an allocation sequence for allocating resources to each device; the waiting time estimation unit is used to obtain the estimated waiting time for each device to be allocated resources based on the allocation sequence of each device.

[0043] Furthermore, the abnormality analysis module includes a real-time demand analysis unit and an abnormality judgment unit;

[0044] The real-time demand analysis unit is used to obtain the resource demand situation predicted by the device in the adjacent resource configuration record when a resource configuration record of a device is generated in real time in the intelligent modal supervision system; the abnormality judgment unit is used to analyze the actual resource demand situation of the device and judge whether the resource allocation situation of the device is abnormal.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The present invention can more accurately predict the resource requirements of each device based on the resource usage of adjacent unit cycles, thereby reducing the impact of other factors and conducting more accurate demand analysis; this is conducive to meeting the requirements of each device during subsequent resource allocation;

[0047] 2. The present invention sets the resource requirements of each device according to the priority of allocation, so as to help the devices with greater needs to obtain resources and complete tasks in a timely manner; thus, the failure of tasks caused by the lack of resource allocation and the impact on the operation process are avoided, thereby improving the operation efficiency of the system;

[0048] 3. The present invention can promptly discover serious deviations in the prediction results, help the device solve the problem of insufficient resource allocation in a timely manner, and help the device continue to operate normally as quickly as possible to avoid affecting the operating performance of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the steps of an intelligent modal supervision method based on network function virtualization;

[0050] Figure 2 A structural diagram of an intelligent modal supervision system based on network function virtualization. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Example: Figure 1 to Figure 2 As shown, the present invention provides an intelligent modality supervision method based on network function virtualization, and the supervision method comprises the following steps:

[0053] Step S100: constructing an intelligent modal supervision system, storing the resource usage process of any device in each unit cycle, and obtaining a number of resource configuration records of the device; analyzing the resource usage recorded in the number of resource configuration records, and obtaining the peak period of resource demand of the device;

[0054] Wherein, step S100 includes the following steps:

[0055] Step S101: arbitrarily select the i-th resource configuration record of a certain device, divide a unit period into several unit times evenly, and obtain the resource usage of the j-th unit time in the i-th resource configuration record as

[0056] Step S102: Obtain resource usage for the j+1th unit time The difference in resource usage between the jth unit time and the j+1th unit time is According to the formula: Calculate the value-added range from the jth unit time to the j+1th unit time

[0057] Step S103: Obtain the increment between any two unit times, and calculate the average value to obtain the average increment of the i-th resource configuration record. like Then the j-th unit time is set as the first characteristic time, if Then the j-th unit time is set as the second characteristic time;

[0058] Example 1: Set the resource usage in the first hour to 5, and the resource usage in the second hour to 10, and obtain the resource usage difference to be 5, and obtain the value-added range between the two hours to be 100%; set the average value-added range to 50%, because 100%>50%, so the first hour is set as the first characteristic time, and is in a demand peak period;

[0059] Step S104: arbitrarily select two adjacent second characteristic times from the unit period, extract a number of first characteristic times between the two adjacent second characteristic times, sort the first characteristic times in chronological order, and select the first first characteristic time; select the second second characteristic time from the two second characteristic times to obtain the time period between the first first characteristic time and the second second characteristic time, and set the time period as the peak demand period in the i-th resource configuration record;

[0060] Embodiment 2: Assume that there are 3 unit times, and the first characteristic time is A, the second characteristic time is B, and the 3 unit times are AAB. Then, the first A is used as the starting time of the peak demand period, and B is used as the ending time of the peak demand period, and a peak demand period is obtained;

[0061] Step S105: Acquire several demand peak periods in each resource configuration record of the certain device to obtain a demand peak period set of the certain device.

[0062] Step S200: Obtain the peak period of resource demand of each device, analyze the resource usage of each resource configuration record in the resource configuration log of each device, and predict the resource demand of each device in a unit period;

[0063] Wherein, step S200 includes the following steps:

[0064] Step S201: arbitrarily select a certain device and obtain a peak demand period set in the certain device; arbitrarily select a unit time in a unit cycle, count the number of times the unit time is in the peak demand period in the peak demand period set as M, and set the number of peak demand periods in the peak demand period set as M all , the probability that the unit time is in the peak demand period is calculated to be G = M / M all ;

[0065] Step S202: Set the unit time in a unit cycle as the jth unit time, obtain the peak demand period of the jth unit time in the i-th resource configuration record of the certain device, and obtain the maximum resource usage in the peak demand period

[0066] Step S203: Obtain the highest resource usage during the peak demand period of the j-th unit time in each resource configuration record, and obtain the average resource demand of the j-th unit time as U j ; The probability of obtaining the jth unit time being in the peak demand period is G j , the estimated resource demand for the j-th unit time is calculated to be R j =U j ×G j ; Select the estimated resource demand with the largest value in a unit period and set it as the estimated resource demand of the device.

[0067] Step S300: arbitrarily selecting a unit period, analyzing the resource demand of each device, and generating an allocation order for allocating resources to each device; according to the allocation order of each device, obtaining an estimated waiting time for each device to be allocated resources;

[0068] Wherein, step S300 includes the following steps:

[0069] Step S301: arbitrarily select a device and set the estimated resource demand of the device in the jth unit time in a unit cycle as R j If R j ≠0, the j-th unit time is set as the target time and extracted, and the extracted target times are divided according to the demand peak period set of the certain equipment;

[0070] Step S302: Obtain the number of target times contained in the first demand peak period in the demand peak period set as N1, set the first target time in the first demand peak period as the jth unit time, according to the formula:

[0071]

[0072] Wherein, k is an integer, and k∈(1,s), s is the number of peak demand periods in the peak demand period set of the device, and t k is the length of the time period of the kth peak demand period; calculating the priority value of allocating resources to the certain device; generating an allocation order of resources to each device according to the priority value from high to low;

[0073] Example 3: The number of target times included in the first demand peak period is obtained to be 5, and the first target time is the first hour; the number of demand peak periods in the demand peak period set of the device is set to be 3 times, and the time period lengths are 2h, 3h, and 1h respectively, and the priority value of the allocated resources is calculated to be Y=5 / 1×6=30;

[0074] Step S303: Set the number of unit time obtained by dividing a unit cycle as P, obtain the number of target time of the device in a unit cycle as Q, and calculate the resource usage time of the device in a unit cycle as T use =T×(Q / P), where T is the duration of a unit cycle;

[0075] Step S304: Obtain the resource usage duration of each device in a unit cycle, set the position of the device in the allocation sequence to w, according to the formula:

[0076]

[0077] Where g is an integer and g∈(1,w-1), (T use ) g is the resource usage time of the device with the position order g in a unit cycle; the estimated waiting time t for the resource to be allocated to the device is calculated wait .

[0078] Step S400: when a resource configuration record of a certain device is generated in real time in the intelligent modality supervision system, the resource demand of the certain device predicted in the adjacent resource configuration record is obtained, the actual resource demand of the certain device is analyzed, and it is determined whether the resource allocation of the certain device is abnormal and an alert is issued;

[0079] Wherein, step S400 includes the following steps:

[0080] Step S401: when a resource configuration record of a certain device is generated in real time, an adjacent resource configuration record of the resource configuration record is obtained, and an estimated resource demand of the certain device in the adjacent resource configuration record is obtained as C;

[0081] Step S402: obtaining the actual resource demand of the certain device recorded in the resource configuration record as C', setting a demand error threshold as θ, and if |C-C'| / C>θ, an abnormal reminder is issued to the certain device.

[0082] An intelligent modality supervision system based on network function virtualization, the supervision system includes a historical record analysis module, a resource demand analysis module, a resource allocation module and an abnormality analysis module;

[0083] The historical record analysis module is used to build an intelligent modal supervision system, store the resource usage process of any device in each unit cycle, and obtain several resource configuration records of the device; analyze the resource usage recorded in the several resource configuration records to obtain the peak period of resource demand of the device;

[0084] The resource demand analysis module is used to obtain the peak period of resource demand of each device, analyze the resource usage of each resource configuration record in the resource configuration log of each device, and predict the resource demand of each device in a unit cycle;

[0085] The resource allocation module is used to select a unit period at random, analyze the resource demand of each device, and generate an allocation order for allocating resources to each device; according to the allocation order of each device, obtain the estimated waiting time for each device to be allocated resources;

[0086] The abnormality analysis module is used to obtain the resource demand situation predicted by the device in the adjacent resource configuration record when a resource configuration record of a certain device is generated in real time in the intelligent modal supervision system, analyze the actual resource demand situation of the certain device, and determine whether the resource allocation situation of the certain device is abnormal and prompt.

[0087] Among them, the historical record analysis module includes a configuration record building unit and a demand peak acquisition unit;

[0088] The configuration record construction unit is used to construct an intelligent modal supervision system, store the resource usage process of any device in each unit cycle, and obtain a number of resource configuration records of the device; the demand peak acquisition unit is used to analyze the resource usage recorded in the several resource configuration records to obtain the peak period of resource demand of the device.

[0089] Among them, the resource demand analysis module includes a resource usage analysis unit and a demand prediction unit;

[0090] The resource usage analysis unit is used to obtain the peak period of resource demand of each device and analyze the resource usage of each resource configuration record in the resource configuration log of each device; the demand prediction unit is used to predict the resource demand of each device in a unit period based on the resource usage of each resource configuration record in the resource configuration log of each device.

[0091] Among them, the resource allocation module includes an allocation order setting unit and a waiting time estimation unit;

[0092] The allocation sequence setting unit is used to arbitrarily select a unit period, analyze the resource demand of each device, and generate an allocation sequence for allocating resources to each device; the waiting time estimation unit is used to obtain the estimated waiting time for each device to be allocated resources based on the allocation sequence of each device.

[0093] Among them, the abnormal analysis module includes a real-time demand analysis unit and an abnormal judgment unit;

[0094] The real-time demand analysis unit is used to obtain the resource demand situation predicted by the device in the adjacent resource configuration record when a resource configuration record of a device is generated in real time in the intelligent modal supervision system; the abnormality judgment unit is used to analyze the actual resource demand situation of the device and judge whether the resource allocation situation of the device is abnormal.

[0095] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An intelligent modality supervision method based on network function virtualization, characterized in that: The supervision method comprises the following steps: Step S100: constructing an intelligent modal supervision system, storing the resource usage process of any device in each unit cycle, and obtaining a number of resource configuration records of the device; analyzing the resource usage recorded in the number of resource configuration records, and obtaining the peak period of resource demand of the device; Step S200: Obtain the peak period of resource demand of each device, analyze the resource usage of each resource configuration record in the resource configuration log of each device, and predict the resource demand of each device in a unit period; Step S300: arbitrarily selecting a unit period, analyzing the resource demand of each device, and generating an allocation order for allocating resources to each device; according to the allocation order of each device, obtaining an estimated waiting time for each device to be allocated resources; Step S400: When a resource configuration record of a certain device is generated in real time in the intelligent modality supervision system, the resource demand situation predicted by the certain device in the adjacent resource configuration record is obtained, the actual resource demand situation of the certain device is analyzed, and it is determined whether the resource allocation situation of the certain device is abnormal and an alert is issued.

2. According to claim 1, the intelligent modality supervision method based on network function virtualization is characterized in that: The step S100 includes the following steps: Step S101: arbitrarily select the i-th resource configuration record of a certain device, divide a unit period into several unit times evenly, and obtain the resource usage of the j-th unit time in the i-th resource configuration record as Step S102: Obtain resource usage for the j+1th unit time The difference in resource usage between the jth unit time and the j+1th unit time is According to the formula: Calculate the value-added range from the jth unit time to the j+1th unit time Step S103: Obtain the increment between any two unit times, and calculate the average value to obtain the average increment of the i-th resource configuration record. like Then the j-th unit time is set as the first characteristic time, if Then the j-th unit time is set as the second characteristic time; Step S104: arbitrarily select two adjacent second characteristic times from the unit period, extract a number of first characteristic times between the two adjacent second characteristic times, sort the first characteristic times in chronological order, and select the first first characteristic time; select the second second characteristic time from the two second characteristic times to obtain the time period between the first first characteristic time and the second second characteristic time, and set the time period as the peak demand period in the i-th resource configuration record; Step S105: Acquire several demand peak periods in each resource configuration record of the certain device to obtain a demand peak period set of the certain device.

3. The intelligent modality supervision method based on network function virtualization according to claim 2 is characterized in that: The step S200 includes the following steps: Step S201: arbitrarily select a certain device and obtain a peak demand period set in the certain device; arbitrarily select a unit time in a unit cycle, count the number of times the unit time is in the peak demand period in the peak demand period set as M, and set the number of peak demand periods in the peak demand period set as M all , the probability that the unit time is in the peak demand period is calculated to be G = M / M all ; Step S202: Set the unit time in a unit cycle as the jth unit time, obtain the peak demand period of the jth unit time in the i-th resource configuration record of the certain device, and obtain the maximum resource usage in the peak demand period Step S203: Obtain the highest resource usage during the peak demand period of the j-th unit time in each resource configuration record, and obtain the average resource demand of the j-th unit time as U j ; The probability of obtaining the jth unit time being in the peak demand period is G j , the estimated resource demand for the j-th unit time is calculated to be R j =U j ×G j ; Select the estimated resource demand with the largest value in a unit period and set it as the estimated resource demand of the device.

4. The intelligent modality supervision method based on network function virtualization according to claim 3 is characterized in that: The step S300 includes the following steps: Step S301: arbitrarily select a device and set the estimated resource demand of the device in the jth unit time in a unit cycle as R j If R j ≠0, the j-th unit time is set as the target time and extracted, and the extracted target times are divided according to the demand peak period set of the certain equipment; Step S302: Obtain the number of target times contained in the first demand peak period in the demand peak period set as N1, set the first target time in the first demand peak period as the jth unit time, according to the formula: Wherein, k is an integer, and k∈(1,s), s is the number of peak demand periods in the peak demand period set of the device, and t k is the length of the time period of the kth peak demand period; calculating the priority value of allocating resources to the certain device; generating an allocation order of resources to each device according to the priority value from high to low; Step S303: Set the number of unit time obtained in a unit cycle as P, obtain the number of target time of a certain device in a unit cycle as Q, and calculate the resource usage time of the device in a unit cycle as T use =T×(Q / P), where T is the duration of a unit cycle; Step S304: Obtain the resource usage duration of each device in a unit cycle, set the position of the device in the allocation sequence to w, according to the formula: Where g is an integer and g∈(1,w-1), (T use ) g is the resource usage time of the device with the position order g in a unit cycle; the estimated waiting time t for the resource to be allocated to the device is calculated wait .

5. The intelligent modality supervision method based on network function virtualization according to claim 4 is characterized in that: The step S400 includes the following steps: Step S401: when a resource configuration record of a certain device is generated in real time, an adjacent resource configuration record of the resource configuration record is obtained, and an estimated resource demand of the certain device in the adjacent resource configuration record is obtained as C; Step S402: obtaining the actual resource demand of the certain device recorded in the resource configuration record as C', setting a demand error threshold as θ, and if |C-C'| / C>θ, an abnormal reminder is issued to the certain device.

6. An intelligent modality supervision system, used to execute the intelligent modality supervision method based on network function virtualization according to any one of claims 1 to 5, characterized in that: The supervision system includes a historical record analysis module, a resource demand analysis module, a resource allocation module and an abnormality analysis module; The historical record analysis module is used to construct an intelligent modal supervision system, store the resource usage process of any device in each unit cycle, and obtain several resource configuration records of the device; analyze the resource usage recorded in the several resource configuration records to obtain the peak period of resource demand of the device; The resource demand analysis module is used to obtain the peak period of resource demand of each device, analyze the resource usage of each resource configuration record in the resource configuration log of each device, and predict the resource demand of each device in a unit period; The resource allocation module is used to arbitrarily select a unit period, analyze the resource demand of each device, and generate an allocation order for allocating resources to each device; according to the allocation order of each device, the estimated waiting time for each device to be allocated resources is obtained; The abnormal analysis module is used to obtain the resource demand situation predicted by the device in the adjacent resource configuration record when a resource configuration record of a certain device is generated in real time in the intelligent modal supervision system, analyze the actual resource demand situation of the certain device, and determine whether the resource allocation situation of the certain device is abnormal. Prompt.

7. The intelligent modality supervision system according to claim 6, characterized in that: The historical record analysis module includes a configuration record construction unit and a demand peak acquisition unit; The configuration record construction unit is used to construct an intelligent modal supervision system, store the resource usage process of any device in each unit cycle, and obtain a number of resource configuration records of the device; the demand peak acquisition unit is used to analyze the resource usage recorded in the several resource configuration records to obtain the peak period of resource demand of the device.

8. The intelligent modality supervision system according to claim 6, characterized in that: The resource demand analysis module includes a resource usage analysis unit and a demand prediction unit; The resource usage analysis unit is used to obtain the peak period of resource demand of each device and analyze the resource usage of each resource configuration record in the resource configuration log of each device; the demand prediction unit is used to predict the resource demand of each device in a unit period based on the resource usage of each resource configuration record in the resource configuration log of each device.

9. The intelligent modality supervision system according to claim 6, characterized in that: The resource allocation module includes an allocation order setting unit and a waiting time estimation unit; The allocation order setting unit is used to arbitrarily select a unit period, analyze the resource requirements of each device, and generate an allocation order for allocating resources to each device; the waiting time estimation unit is used to obtain the estimated waiting time for each device to be allocated resources based on the allocation order of each device.

10. The intelligent modality supervision system according to claim 6, characterized in that: The abnormality analysis module includes a real-time demand analysis unit and an abnormality judgment unit; The real-time demand analysis unit is used to obtain the resource demand situation predicted by the device in the adjacent resource configuration record when a resource configuration record of a device is generated in real time in the intelligent modal supervision system; the abnormality judgment unit is used to analyze the actual resource demand situation of the device and determine whether the resource allocation situation of the device should be abnormal.

Images