Computer network communication data processing system based on artificial intelligence

Abstract

A computer network communication data processing system based on artificial intelligence particularly relates to the field of computer network communication data processing and comprises a data acquisition module, a data processing module, a resource allocation module and a feedback control module. The data acquisition module collects an old protocol communication scale value, new and old protocol data packet comparison information, protocol conversion time data, conversion error information, network flow data, a CPU utilization rate and a memory utilization rate; the data processing module determines a protocol conversion comprehensive cost based on evaluation protocol conversion, a time adjustment factor for time synchronization of a client and a server and a resource allocation adjustment coefficient allocated to data processing by an adjustment system, and the resource allocation module adjusts resource allocation of the system according to the resource allocation adjustment coefficient. And the feedback control module repeats the steps S1-S5 until the error rate reaches 0.05. According to the invention, the protocol conversion cost, time synchronization and resource allocation can be dynamically adjusted according to the change of the network environment, so that the system is more efficient when processing communication data.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of computer network communication data processing, and particularly relates to a computer network communication data processing system based on artificial intelligence. BACKGROUND

[0002] At present, in the field of computer network communication data processing, with the explosive growth of data and the increasing complexity of network environment, artificial intelligence technology is widely used to improve processing efficiency and quality. However, there are many non-ideal factors in the actual environment, which affect the data processing effect.

[0003] At present, different devices and systems adopt different communication protocol versions or types, which leads to protocol incompatibility. Specifically, old devices support old version protocols, and new devices use new version protocols. When data interaction occurs, errors will occur due to protocol differences. In the process of data processing in the prior art, the cost is not comprehensively evaluated by taking into account factors such as data scale, protocol difference degree, protocol conversion time, error rate and the like, which will lead to unreasonable resource allocation and affect processing efficiency and quality. Among them, for the time synchronization problem of the client and the server, the prior art does not fully consider the influence of protocol conversion cost and network burst traffic factors on time synchronization, which leads to inaccurate time synchronization and affects the real-time performance and accuracy of data transmission. In terms of resource allocation, the factors of time synchronization adjustment, network communication intermittence and current data processing capacity of the system are not comprehensively considered, which will also lead to insufficient or waste of resource allocation.

[0004] Therefore, the technical personnel in the field provide a computer network communication data processing system based on artificial intelligence to solve the problems raised in the background. SUMMARY

[0005] The technical problem solved by the application is to provide a computer network communication data processing system based on artificial intelligence to realize the effects of comprehensively evaluating protocol conversion cost, accurately adjusting time synchronization and optimizing resource allocation.

[0006] To solve the above problems, the application provides the following technical scheme:

[0007] The computer network communication data processing system based on artificial intelligence comprises a data acquisition module, a data processing module, a resource allocation module and a feedback control module, and the specific implementation steps are as follows:

[0008] S1, the data acquisition module collects old protocol communication scale value, new and old protocol data packet comparison information, protocol conversion time data, conversion error information, network traffic data, CPU usage and memory usage;

[0009] S2.1, the data processing module determines the protocol difference degree based on the new and old protocol packet comparison information;

[0010] S2.2, based on the conversion error information, determine the error rate;

[0011] S2.3, based on the old protocol communication scale value, the protocol difference degree, the protocol conversion time data and the weight distribution calculation of the error rate, determine the protocol conversion comprehensive cost based on the evaluation of the protocol conversion;

[0012] S3.1, based on the network traffic data, determine the burst traffic coefficient and the intermittence coefficient of network communication;

[0013] S3.2, based on the protocol conversion comprehensive cost, the protocol conversion time data and the weight distribution calculation of the burst traffic coefficient, determine the time adjustment factor of the time synchronization of the client and the server;

[0014] S4.1, based on the CPU usage and the memory usage, determine the data processing capacity coefficient;

[0015] S4.2, based on the time adjustment factor, the intermittence coefficient and the data processing capacity coefficient, determine the resource allocation adjustment coefficient of the adjustment system allocated to data processing;

[0016] S5, the resource allocation module adjusts the resource allocation of the system according to the resource allocation adjustment coefficient;

[0017] S6, the feedback control module repeats steps S1-S5 until the error rate reaches 0.05.

[0018] Further: the equipment collected by the data collection module includes network monitoring equipment, protocol analysis software, system clock of client and server, network traffic monitoring equipment and system monitoring software;

[0019] The network monitoring equipment is used to collect the old protocol communication scale value and the conversion error information;

[0020] The protocol analysis software is used to collect the new and old protocol packet comparison information;

[0021] The system clock of the client and the server is used to collect the protocol conversion time data;

[0022] The network traffic monitoring equipment is used to collect the network traffic data;

[0023] The system monitoring software is used to collect the CPU usage and the memory usage.

[0024] Further, the network monitoring device obtains the old protocol communication scale value and the communication scale reference value by counting the total sum of data packet byte numbers of the old protocol and historical protocols;

[0025] The new and old protocol data packet comparison information includes the number of fields, the type of fields and the length of fields in the new protocol, and the number of fields, the type of fields and the length of fields in the old protocol;

[0026] The protocol conversion time data includes the protocol conversion start time, the protocol conversion end time, the protocol conversion average time, the client local time, the server time, and the maximum time difference allowed by the system;

[0027] The conversion error information includes the error number and the total conversion number;

[0028] The network traffic data includes the network traffic average value, the network traffic peak value, the network communication interruption number, and the duration of each interruption in the same communication duration.

[0029] Further, the specific implementation process of steps S2.1 to S2.3 is as follows:

[0030] Based on the ratio of the old protocol communication scale value to the communication scale reference value, a communication scale coefficient is determined;

[0031] The protocol analysis software analyzes the data packet format and instruction set of the new and old protocols;

[0032] The field number, the field type and the field length of different fields in the new and old protocols are compared;

[0033] According to a preset rule, the difference degree is quantified to determine the field number difference value, the field type difference value and the field length difference value;

[0034] Different weights of different types of difference values are set, including the field number difference weight, the field type difference weight and the field length difference weight;

[0035] The different weights of different types of difference values are multiplied and added to determine the protocol difference degree;

[0036] Based on the ratio of the error number to the total conversion number, the error rate is determined;

[0037] Based on the difference between the protocol conversion end time and the protocol conversion start time, the conversion time difference is determined;

[0038] Based on the ratio of the conversion time difference to the protocol conversion average time, the conversion time influence coefficient is determined;

[0039] Based on the addition result of the weight distribution of the communication scale coefficient, the protocol difference degree, the error rate and the conversion time influence coefficient, the protocol conversion comprehensive cost is determined.

[0040] Further, the specific determination process of the burst traffic coefficient and the intermittence coefficient is as follows:

[0041] Based on the ratio of the network traffic peak value to the network traffic average value, the burst traffic coefficient is determined.

[0042] The product of the network communication interruption times and the duration of each interruption within the same communication duration is summed up and divided by the same communication duration to determine the intermittence coefficient.

[0043] Further, the specific implementation process of the step S3.2 is as follows:

[0044] Based on the difference between the client local time and the server time, the system time difference is determined.

[0045] Based on the ratio of the system time difference to the maximum system time difference allowed, the system time difference influence coefficient is determined.

[0046] Based on the addition result of the weight distribution of the system time difference influence coefficient, the protocol conversion comprehensive cost and the burst traffic coefficient, the time adjustment factor is determined.

[0047] Further, the system clock of the client and the server collects the time difference samples of the client and the server within a period of time.

[0048] The time difference samples are sorted, and the maximum value after sorting is determined as the maximum system time difference allowed.

[0049] Further, the specific process of the steps S3.1 to S3.2 is as follows:

[0050] The system monitoring software collects the data processing capacity samples of the system under normal operation state;

[0051] The distribution of the data processing capacity samples is analyzed, and the average value of the data processing capacity samples is selected as the data processing capacity reference coefficient;

[0052] Wherein, the data processing capacity coefficient and the data processing capacity reference coefficient are obtained by the respective CPU usage rate and the memory usage rate, and the specific process is as follows:

[0053] The CPU available rate is subtracted by the CPU usage rate and the weight distribution calculation is performed to determine the CPU available coefficient;

[0054] The memory available rate is subtracted from the memory usage rate and a weight distribution calculation is performed to determine a memory available usage coefficient;

[0055] Based on the addition of the CPU available usage coefficient and the memory available usage coefficient, the data processing capability coefficient and the data processing capability reference coefficient are respectively determined;

[0056] Based on the ratio of the data processing capability coefficient and the data processing capability reference coefficient, a data processing capability influence coefficient is determined;

[0057] Based on the addition result of the weight distribution of the time adjustment factor, the intermittence coefficient and the data processing capability influence coefficient, the resource allocation adjustment coefficient is determined.

[0058] Further: based on the result of the resource allocation adjustment coefficient, if the resource allocation adjustment coefficient increases, the resource allocation is increased; if the resource allocation adjustment coefficient decreases, the resource allocation is reduced.

[0059] Further: the error rate is determined by using the single ratio of the initial error number and the total conversion number in the initial stage;

[0060] With the repetition number of the feedback control module, the average value of the ratio of the error number and the total conversion number of the repeated number is used to determine the error rate.

[0061] The effect of the above scheme is as follows:

[0062] The protocol conversion comprehensive cost based on the evaluation of the protocol conversion is determined by comprehensively considering the weight distribution of the old protocol communication scale value, the protocol difference degree, the protocol conversion time data and the error rate factor, so that the system can comprehensively evaluate the cost in the protocol conversion process by comprehensively considering the characteristics of the communication behavior and the protocol interaction mode, so as to avoid the waste of resources or low processing efficiency caused by the incomplete consideration of a single factor.

[0063] The time adjustment factor is determined by considering the protocol conversion comprehensive cost, the protocol conversion time data and the burst traffic coefficient factor, wherein the classified data is further integrated and optimized according to the data fusion and feature enhancement method, and more representative and effective time factors and burst traffic coefficients related to communication are extracted to enhance the data features, more accurately reflect the device state, and through the time adjustment factor, the system can more accurately adjust the time synchronization of the client and the server, thereby improving the real-time performance and accuracy of data transmission, and reducing the data transmission errors caused by different time synchronization.

[0064] According to the resource allocation adjustment coefficient determined by comprehensively considering the time adjustment factor, the intermittence coefficient and the data processing capacity coefficient, the system can dynamically adjust the resource allocation, so as to ensure that the CPU, memory and network bandwidth resources can be reasonably allocated under different network environments and system states, improve the resource utilization, and enhance the stability and processing capacity of the system.

[0065] In addition, in the cyclic influence process, the cyclic feedback is ended when the error rate reaches 0.05, so that the system can be prevented from continuing to consume resources in the case that the error rate cannot be effectively reduced, the resources can be used for other more meaningful tasks, the processing efficiency is improved, and the system stability is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0066] Figure 1 A flowchart of the intelligent system resource allocation of the computer network communication data processing system is shown in the figure.

[0067] Figure 2 A schematic diagram of the overall module in the system is shown in the figure.

[0068] Figure 3 A schematic diagram of the equipment used by the data acquisition module and the corresponding data in the application is shown in the figure. DETAILED DESCRIPTION

[0069] The technical solutions in the embodiments of the application will be clearly and completely introduced below with reference to the drawings in the embodiments of the application.

[0070] Embodiment one, please refer to Figures 1-3 The computer network communication data processing system based on artificial intelligence comprises a data acquisition module, a data processing module, a resource allocation module and a feedback control module, and the specific implementation process is as follows:

[0071] S1, the data acquisition module collects old protocol communication scale values, new and old protocol data packet comparison information, protocol conversion time data, conversion error information, network traffic data, CPU usage and memory usage;

[0072] S2.1, the data processing module determines the protocol difference degree based on the new and old protocol data packet comparison information;

[0073] S2.2, the error rate is determined based on the conversion error information;

[0074] S2.3, the protocol conversion comprehensive cost based on the evaluation of the protocol conversion is determined based on the weight distribution calculation of the old protocol communication scale value, the protocol difference degree, the protocol conversion time data and the error rate;

[0075] S3.1, the burst traffic coefficient and the intermittence coefficient of network communication are determined based on the network traffic data;

[0076] S3.2, determining the time adjustment factor of the time synchronization between the client and the server based on the weight distribution calculation of the protocol conversion comprehensive cost, the protocol conversion time data and the burst traffic coefficient;

[0077] S4.1, determining the data processing capacity coefficient based on the CPU usage and the memory usage;

[0078] S4.2, determining the resource allocation adjustment coefficient of the adjustment system allocated to the data processing based on the time adjustment factor, the intermittent coefficient and the data processing capacity coefficient;

[0079] S5, the resource allocation module adjusts the resource allocation of the system according to the resource allocation adjustment coefficient;

[0080] S6, the feedback control module repeats steps S1-S5 until the error rate reaches 0.05.

[0081] The equipment collected by the data collection module includes network monitoring equipment, protocol analysis software, system clock of the client and the server, network traffic monitoring equipment and system monitoring software;

[0082] The network monitoring equipment is used to collect the old protocol communication scale value and the conversion error information, the network monitoring equipment obtains the old protocol communication scale value and the communication scale reference value by counting the total number of data packet bytes of the old protocol and the historical protocol, and the conversion error information includes the error number and the total conversion number;

[0083] The protocol analysis software is used to collect the new and old protocol data packet comparison information, and the new and old protocol data packet comparison information includes the number of fields, the type of fields and the length of fields in the new protocol, and the number of fields, the type of fields and the length of fields in the old protocol;

[0084] The system clock of the client and the server is used to collect the protocol conversion time data, and the protocol conversion time data includes the protocol conversion start time, the protocol conversion end time, the protocol conversion average time, the client local time, the server time and the maximum time difference allowed by the system;

[0085] The network traffic monitoring equipment is used to collect the network traffic data, and the network traffic data includes the network traffic average value, the network traffic peak value, the network communication interruption number and the duration of each interruption in the same communication duration;

[0086] The system monitoring software is used to collect the CPU usage and the memory usage.

[0087] The embodiment obtains parameters through a data acquisition module and related equipment, and finally determines the calculation results including protocol conversion comprehensive cost, time adjustment factor and resource allocation adjustment coefficient through processing and calculation of the data processing module. These calculation results guide the resource allocation module to optimize resource allocation, improve processing efficiency, guarantee time synchronization accuracy and system stability, enable the system to effectively cope with complex network environment, and improve overall performance.

[0088] Please refer to Figure 1 The specific implementation process of steps S2.1 to S2.3 is as follows:

[0089] Determine the communication scale coefficient based on the ratio of the old protocol communication scale value and the communication scale reference value.

[0090] The protocol analysis software analyzes the data packet format and instruction set of the new and old protocols.

[0091] Compare the field quantity, field type and field length of different fields in the new and old protocols.

[0092] Quantify the difference degree according to the preset rule to determine the field quantity difference value, field type difference value and field length difference value.

[0093] Set different weights for different types of difference values, including field quantity difference weight, field type difference weight and field length difference weight.

[0094] Multiply and add the different weights of different types of difference values to determine the protocol difference degree.

[0095] The calculation formula of the protocol difference degree is as follows:

[0096] CY=a×c1+b×c2+c×c3.

[0097] CY is the protocol difference degree, c1 is the field quantity difference value, c2 is the field type difference value, c3 is the field length difference value, a is the field quantity difference weight, b is the field type difference weight, and c is the field length difference weight.

[0098] In addition, a+b+c=1.

[0099] Determine the error rate based on the ratio of the number of errors and the total number of conversions.

[0100] Determine the conversion time difference based on the difference between the protocol conversion end time and the protocol conversion start time.

[0101] Determine the conversion time influence coefficient based on the ratio of the conversion time difference and the protocol conversion average time.

[0102] Determine the protocol conversion comprehensive cost based on the addition result of the weight distribution of the communication scale coefficient, the protocol difference degree, the error rate and the conversion time influence coefficient;

[0103] Firstly, the calculation formula of the protocol conversion comprehensive cost is as follows:

[0104]

[0105] The calculation formula of the error rate is as follows:

[0106] CC=C err / C total ;

[0107] Wherein:

[0108] XC is the protocol conversion comprehensive cost;

[0109] G old is the old protocol communication scale value, G norm is the communication scale reference value, T trans is the conversion time difference, T norm is the protocol conversion average time, CC is the error rate, w1, w2, w3 and w4 are weight coefficients, and w1+w2+w3+w4=1;

[0110] C err is the error times, C total is the total conversion times.

[0111] In this embodiment, the larger the data scale is, the more resources and time required for protocol conversion are, The larger the part of the calculation result is, the larger the protocol conversion comprehensive cost XC will be, which reflects the positive influence of the data scale on the protocol conversion comprehensive cost XC;

[0112] The larger the protocol difference degree CY is, the higher the difficulty and cost of protocol conversion are, the larger the part of the calculation result w2×CY is, which will make the protocol conversion comprehensive cost XC increase, which reflects the important influence of the protocol difference degree on the protocol conversion comprehensive cost XC;

[0113] The longer the conversion time difference is, the longer the system resources are occupied, the higher the cost is, The larger the part of the calculation result is, the larger the protocol conversion comprehensive cost XC will be, which reflects the influence of the protocol conversion time on the protocol conversion comprehensive cost XC;

[0114] The higher the error rate CC is, the more time and resources are required to handle errors, the higher the protocol conversion cost is, the larger the part of the calculation result w4×CC is, which will make the protocol conversion comprehensive cost XC increase, which shows that the error rate has a positive influence on the protocol conversion comprehensive cost XC.

[0115] Please refer to Figure 1 and Figure 3 The specific determination process of the burst traffic coefficient and the intermittence coefficient is as follows:

[0116] The burst traffic coefficient is determined based on the ratio of the network traffic peak value to the network traffic average value, and the calculation formula of the burst traffic coefficient is as follows:

[0117] LT=F peak / F avg ;

[0118] LT is the burst traffic coefficient, F peak is the network traffic peak value, and F avg is the network traffic average value.

[0119] The specific implementation process of step S3.2 is as follows:

[0120] The system time difference is determined based on the difference between the client local time and the server time.

[0121] The system time difference impact coefficient is determined based on the ratio of the system time difference to the maximum system allowed time difference.

[0122] The time adjustment factor is determined based on the weighted sum of the system time difference impact coefficient, the protocol conversion comprehensive cost, and the burst traffic coefficient.

[0123] The calculation formula of the time adjustment factor is as follows:

[0124]

[0125] Wherein:

[0126] TB is the time adjustment factor.

[0127] T k is the client local time, T f is the server time, T norm-sync is the maximum system allowed time difference, w5, w6, and w7 are weight coefficients, and w5+w6+w7=1.

[0128] Wherein, the larger the system time difference, the more out of sync the client and the server, the larger the adjustment amplitude needed, the larger the part of the calculation result, the larger the time adjustment factor TB, which reflects the importance of the time difference to the time synchronization adjustment;

[0129] Similarly, the higher the protocol conversion comprehensive cost XC, the greater the impact on time synchronization, and a larger adjustment factor is needed to ensure the accuracy of time synchronization, and the larger the part of the calculation result w6×XC, the larger the time adjustment factor TB.

[0130] The greater the burst traffic coefficient LT, the greater the possibility of increasing network delay and jitter, affecting the accuracy of time synchronization, and requiring a larger adjustment factor for compensation, so the greater the calculation result of this part, the greater the time adjustment factor TB.

[0131] Please refer to Figure 1 and Figure 3 , the system clock of the client and the server collects the time difference samples of the client and the server within a period of time;

[0132] The time difference samples are sorted, and the maximum value is taken after sorting to determine the maximum time difference allowed by the system;

[0133] The product of the number of network communication interruptions and the duration of each interruption within the same communication duration is summed up and divided by the same communication duration to determine the intermittency coefficient;

[0134] The specific process of steps S3.1 to S3.2 is as follows:

[0135] The system monitoring software collects data processing capacity samples of the system in a normal operating state;

[0136] The distribution of the data processing capacity samples is analyzed, and the average value of the data processing capacity samples is selected as the data processing capacity reference coefficient;

[0137] Among them, the data processing capacity coefficient and the data processing capacity reference coefficient are obtained through the respective CPU usage rate and memory usage rate, as follows:

[0138] The CPU available rate is subtracted from the CPU usage rate and a weight distribution calculation is performed to determine the CPU available coefficient;

[0139] The memory available rate is subtracted from the memory usage rate and a weight distribution calculation is performed to determine the memory available coefficient;

[0140] Based on the addition of the CPU available coefficient and the memory available coefficient, the data processing capacity coefficient and the data processing capacity reference coefficient are determined respectively;

[0141] Based on the ratio of the data processing capacity coefficient and the data processing capacity reference coefficient, the data processing capacity influence coefficient is determined;

[0142] Based on the addition of the weight distribution of the time adjustment factor, the intermittency coefficient and the data processing capacity influence coefficient, the resource allocation adjustment coefficient is determined.

[0143] First, the intermittency coefficient calculation formula is as follows:

[0144]

[0145] J is an intermittence coefficient, d is the number of network communication interruptions, dt i is the ith interruption duration, and T is the same communication duration;

[0146] The number of network communication interruptions d includes the number of short-time jitter interruptions;

[0147] T norm-sync = max (△t1,△t2,...,△tq) ;

[0148] Wherein:

[0149] △t1 is the first sample time difference,△t2 is the second sample time difference, and△tq is the qth sample time difference;

[0150] The calculation formula of the resource allocation adjustment coefficient is as follows:

[0151]

[0152] The calculation formula of the data processing capacity coefficient is as follows:

[0153] C proc = w cpu × (1-u cpu )+ w mem × (1-u mem ) ;

[0154] Wherein:

[0155] FZ is the resource allocation adjustment coefficient;

[0156] C proc is the data processing capacity coefficient, C norm is the data processing capacity reference coefficient, w8, w9 and w10 are weight coefficients, and w8+w9+w10=1;

[0157] u cpu is the CPU usage, u mem is the memory usage, w cpu is the CPU weight coefficient, w mem is the memory weight coefficient, and w cpu +w mem =1;

[0158] Wherein, the larger the time synchronization adjustment factor TB is, the more resources are needed for time synchronization adjustment, and the larger the w8×TB part calculation result is, which will increase the resource allocation adjustment coefficient FZ, reflecting the influence of time synchronization on resource allocation;

[0159] The larger the intermittence coefficient J is, the more unstable the network communication is, and more resources are needed to deal with the instability, and the larger the w9xJ part calculation result is, the larger the resource allocation adjustment coefficient FZ is, which reflects the influence of network intermittence on resource allocation.

[0160] When the current data processing capacity of the system is low, the value of CC is small, and thus more resources are needed to improve the data processing capacity, and the part calculation result will affect the size of the resource allocation adjustment coefficient FZ, so as to adjust the resource allocation.

[0161] Embodiment two, please refer to Figure 1 Based on the result of the resource allocation adjustment coefficient, if the resource allocation adjustment coefficient increases, the resource allocation is increased, and if the resource allocation adjustment coefficient decreases, the resource allocation is reduced.

[0162] The error rate is determined by the single ratio of the initial error number to the total conversion number in the initial stage.

[0163] With the repetition number of the feedback control module, the average value of the ratio of the error number to the total conversion number of the repeated number is used to determine the error rate.

[0164] In this embodiment, if the error rate CC is always high, the continuous loop feedback will waste a large amount of system resources, and the setting of ending the loop feedback when the error rate reaches 0.05 can avoid the system from continuing to consume resources in the case of being unable to effectively reduce the error rate, and the resources can be used for other more meaningful tasks, and the timely ending of the loop feedback can avoid the system from being trapped in an invalid loop, so that the system can adjust the processing strategy more quickly, improve the overall processing efficiency, at the same time, after ending the loop feedback, the system can take other measures to further reduce the error rate CC according to the current error rate CC, in addition, the error rate CC is too high, which may affect the stability of the system, and ending the loop feedback when the error rate CC reaches 0.05 is helpful to ensure the stability of the system and reduce the negative effects caused by the high error rate CC.

[0165] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art, without departing from the spirit and scope of the present application, can make various changes and modifications, therefore the protection scope of the present application should be limited by the scope defined by the claims.

Claims

1. A computer network communication data processing system based on artificial intelligence, characterized by: It includes data acquisition module, data processing module, resource allocation module and feedback control module. The specific implementation steps are as follows: S1, the data acquisition module collects the old protocol communication scale value, the new and old protocol data packet comparison information, the protocol conversion time data, the conversion error information, the network traffic data, the CPU usage rate and the memory usage rate; S2.

1. The data processing module determines the degree of protocol difference based on the comparison information of the new and old protocol data packets; S2.

2. Determine an error rate based on the conversion error information; S2.

3. Determine a comprehensive protocol conversion cost based on the evaluation protocol conversion based on a weighted calculation of the old protocol communication scale value, the protocol difference degree, the protocol conversion time data, and the error rate; S3.

1. Determine a burst flow coefficient and an intermittent flow coefficient of network communication based on the network flow data; S3.

2. Determine a time adjustment factor for client-server time synchronization based on the protocol conversion comprehensive cost, the protocol conversion time data, and a weighted distribution calculation of the burst traffic coefficient; S4.

1. Determine a data processing capacity coefficient based on the CPU usage and the memory usage; S4.

2. Determine, based on the time adjustment factor, the intermittency coefficient, and the data processing capacity coefficient, a resource allocation adjustment coefficient for adjusting the system allocated to data processing; S5. The resource allocation module adjusts the resource allocation of the system according to the resource allocation adjustment coefficient; S6. The feedback control module repeats steps S1-S5 until the error rate reaches 0.

05.

2. The computer network communication data processing system based on artificial intelligence according to claim 1, characterized in that: The equipment collected by the data acquisition module includes network monitoring equipment, protocol analysis software, system clocks of clients and servers, network traffic monitoring equipment and system monitoring software; The network monitoring device is used to collect the old protocol communication scale value and the conversion error information; The protocol analysis software is used to collect comparison information of the new and old protocol data packets; The system clocks of the client and the server are used to collect the protocol conversion time data; The network traffic monitoring device is used to collect the network traffic data; The system monitoring software is used to collect the CPU usage and the memory usage.

3. The computer network communication data processing system based on artificial intelligence according to claim 2, characterized in that: The network monitoring device obtains the communication scale value and the communication scale reference value of the old protocol by counting the total number of bytes of data packets of the old protocol and the historical protocol; The comparison information of the new and old protocol data packets includes the number of fields, field types and field lengths in the new protocol, and the number of fields, field types and field lengths in the old protocol; The protocol conversion time data includes the protocol conversion start time, protocol conversion end time, protocol conversion average time, client local time, server time, and the maximum time difference allowed by the system; The conversion error information includes the number of errors and the total number of conversions; The network traffic data includes the average network traffic volume, the peak network traffic volume, the number of network communication interruptions, and the duration of each interruption within the same communication duration.

4. The computer network communication data processing system based on artificial intelligence according to claim 3, characterized in that: The specific implementation process of steps S2.1 to S2.3 is as follows: Determining a communication scale coefficient based on a ratio of the old protocol communication scale value to the communication scale reference value; The protocol analysis software analyzes the data packet format and instruction set of the new and old protocols; Comparing the number of fields, the field type, and the field length of different fields in the new and old protocols; Quantify the difference according to the preset rules to determine the difference value of field quantity, field type and field length; Set different weights for different types of difference values, including field quantity difference weight, field type difference weight, and field length difference weight; Multiplying and adding different weights of different types of difference values ​​to determine the agreement difference degree; determining the error rate based on a ratio of the number of errors to the total number of conversions; Determining a conversion time difference based on a difference between the protocol conversion end time and the protocol conversion start time; Determining a conversion time influence coefficient based on a ratio of the conversion time difference to the protocol conversion average time; The comprehensive cost of the protocol conversion is determined based on the sum of the weights assigned to the communication scale coefficient, the protocol difference, the error rate, and the conversion time impact coefficient.

5. The computer network communication data processing system based on artificial intelligence according to claim 4, characterized in that: The specific determination process of the burst flow coefficient and the intermittent coefficient is as follows: Determining a burst flow coefficient based on a ratio of the network flow peak value to the network flow average value; The intermittent coefficient is determined by summing up the product of the number of network communication interruptions within the same communication duration and the duration of each interruption, and dividing the product by the same communication duration.

6. The computer network communication data processing system based on artificial intelligence according to claim 5, characterized in that: The specific implementation process of step S3.2 is as follows: Determine a system time difference based on a difference between the client local time and the server time; Determining a system time difference influence coefficient based on a ratio of the system time difference to a maximum time difference allowed by the system; The time adjustment factor is determined based on the sum of the system time difference influence coefficient, the protocol conversion comprehensive cost, and the burst traffic coefficient weight distribution.

7. The computer network communication data processing system based on artificial intelligence according to claim 6, characterized in that: The system clocks of the client and the server collect time difference samples between the client and the server over a period of time; The time difference samples are sorted, and the maximum value is taken after sorting to determine the maximum time difference allowed by the system.

8. The computer network communication data processing system based on artificial intelligence according to claim 6, characterized in that: The specific process of steps S3.1 to S3.2 is as follows: The system monitoring software collects samples of the data processing capacity of the system under normal operating conditions; Analyzing the distribution of the data processing capability samples and selecting the average value of the data processing capability samples as a data processing capability reference coefficient; The data processing capability coefficient and the data processing capability reference coefficient are obtained by using the CPU usage and the memory usage, respectively, as follows: The CPU availability rate is deducted from the CPU usage rate and a weight distribution calculation is performed to determine a CPU available coefficient; The memory availability rate is subtracted from the memory usage rate and a weight distribution calculation is performed to determine a memory availability coefficient; Determining the data processing capacity coefficient and the data processing capacity reference coefficient based on the sum of the CPU available coefficient and the memory available coefficient; determining a data processing capability impact coefficient based on a ratio of the data processing capability coefficient to the data processing capability reference coefficient; The resource allocation adjustment coefficient is determined based on the sum of the weight distribution of the time adjustment factor, the intermittent coefficient and the data processing capacity influence coefficient.

9. The computer network communication data processing system based on artificial intelligence according to claim 8, characterized in that: Based on the result of the resource allocation adjustment coefficient, if the resource allocation adjustment coefficient increases, resource allocation is increased; if the resource allocation adjustment coefficient decreases, resource allocation is reduced.

10. The computer network communication data processing system based on artificial intelligence according to claim 4, characterized in that: The error rate is determined in an initial stage by using a single ratio of the initial number of errors to the total number of conversions; The error rate is determined according to the number of repetitions of the feedback control module and the average value of the ratio of the number of errors that have occurred in the number of repetitions to the total number of conversions.

Images