Artificial Intelligence-Based Computer Network Communication Data Processing System

Abstract

This invention relates to an AI-based computer network communication data processing system, specifically in the field of computer network communication data processing. It includes a data acquisition module, a data processing module, a resource allocation module, and a feedback control module. The data acquisition module collects old protocol communication scale values, comparison information of old and new protocol data packets, protocol conversion time data, conversion error information, network traffic data, CPU utilization, and memory utilization. The data processing module determines the overall protocol conversion cost based on the evaluation of the protocol conversion, the time adjustment factor for client-server time synchronization, and the resource allocation adjustment coefficient for adjusting the system's allocation to data processing. The resource allocation module adjusts the system's resource allocation according to the resource allocation adjustment coefficient. The feedback control module repeats steps S1-S5 until the error rate reaches 0.05. This invention can dynamically adjust protocol conversion costs, time synchronization, and resource allocation according to changes in the network environment, making the system more efficient in processing communication data.

Description

Technical Field

[0001] This invention relates to the field of computer network communication data processing technology, specifically to a computer network communication data processing system based on artificial intelligence. Background Technology

[0002] Currently, in the field of computer network communication data processing, with the surge in data volume and the increasing complexity of network environments, artificial intelligence technology is widely used to improve processing efficiency and quality. However, many non-ideal factors exist in the real-world environment, affecting the data processing results.

[0003] Currently, different devices and systems use different versions or types of communication protocols, leading to protocol incompatibility. Specifically, older devices support older protocols while newer devices use newer ones, causing errors during data exchange due to these protocol differences. Existing technologies do not comprehensively assess costs by considering factors such as data scale, protocol differences, protocol conversion time, and error rates during data processing. This results in unreasonable resource allocation, impacting processing efficiency and quality. Specifically, regarding client-server time synchronization, existing technologies do not adequately consider the impact of protocol conversion costs and sudden network traffic spikes on time synchronization, leading to inaccurate time synchronization and affecting the real-time performance and accuracy of data transmission. Furthermore, resource allocation does not comprehensively consider factors such as time synchronization adjustments, network communication intermittency, and the system's current data processing capabilities, resulting in insufficient or wasted resources.

[0004] Therefore, those skilled in the art have provided computer network communication data processing systems based on artificial intelligence to solve the problems mentioned in the background art. Summary of the Invention

[0005] The technical problem solved by this invention is to provide an artificial intelligence-based computer network communication data processing system to achieve the effects of comprehensively evaluating protocol conversion costs, accurately adjusting time synchronization, and optimizing resource allocation.

[0006] To address the above problems, the present invention provides the following technical solution:

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

[0008] S1. The data acquisition module collects the communication scale value of the old protocol, the comparison information of the data packets of the new and old protocols, the protocol conversion time data, the conversion error information, the network traffic data, the CPU utilization rate and the memory utilization rate;

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

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

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

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

[0013] S3.2. Based on the weighted allocation calculation of the protocol conversion comprehensive cost, the protocol conversion time data, and the burst traffic coefficient, determine the time adjustment factor for client-server time synchronization;

[0014] S4.1. Based on the CPU utilization rate and the memory utilization rate, determine the data processing capability coefficient;

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

[0016] S5. The resource allocation module adjusts the system's resource allocation 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] Furthermore: the data acquisition module includes network monitoring equipment, protocol analysis software, system clocks for the client and server, network traffic monitoring equipment, and system monitoring software;

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

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

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

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

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

[0024] Further: The network monitoring device obtains the communication scale value and communication scale reference value of the old protocol by statistically analyzing the total number of bytes in data packets of the old protocol and the historical protocol;

[0025] 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, as well as the number of fields, field types, and field lengths in the old protocol;

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

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

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

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

[0030] The communication scale coefficient is determined based on the ratio of the old protocol communication scale value to the communication scale reference value;

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

[0032] Compare the number of fields, field types, and field lengths in the new and old protocols.

[0033] The degree of difference is quantified according to preset rules, and the differences in the number of fields, the types of fields, and the length of fields are determined.

[0034] Set different weights for different types of difference values, including weights for differences in the number of fields, differences in the type of fields, and differences in the length of fields;

[0035] The protocol difference degree is determined by multiplying and adding the different weights of different types of difference values;

[0036] The error rate is determined based on the ratio of the number of errors to the total number of conversions.

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

[0038] The conversion time impact coefficient is determined based on the ratio of the conversion time difference to the average conversion time of the protocol.

[0039] The overall cost of protocol conversion is determined by summing the weighted distributions of the communication scale coefficient, the protocol difference, the error rate, and the conversion time impact coefficient.

[0040] Further: The specific determination process for the burst flow coefficient and the intermittency coefficient is as follows:

[0041] The burst traffic coefficient is determined based on the ratio of the peak network traffic to the average network traffic.

[0042] The intermittency coefficient is determined by summing the product of the number of network communication interruptions within the same communication duration and the duration of each interruption, and then dividing by the same communication duration.

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

[0044] The system time difference is determined based on the difference between the client's local time and the server's time.

[0045] The system time difference influence coefficient is determined based on the ratio of the system time difference to the maximum allowable time difference of the system.

[0046] The time adjustment factor is determined by summing the system time difference impact coefficient, the protocol conversion comprehensive cost, and the burst traffic coefficient weight allocation.

[0047] Furthermore: the system clocks of the client and server collect time difference samples between the client and server over a period of time;

[0048] The time difference samples are sorted, and the maximum value after sorting is taken to determine the maximum allowable time difference of the system.

[0049] Further: The specific processes of steps S3.1 to S3.2 are as follows:

[0050] The system monitoring software is used to collect data processing capability samples of the system under normal operating conditions.

[0051] Analyze the distribution of the data processing capability samples and select the average value of the data processing capability samples as the data processing capability reference coefficient;

[0052] The data processing capability coefficient and the data processing capability reference coefficient are both obtained through their respective CPU utilization and memory utilization, as detailed below:

[0053] The CPU availability coefficient is determined by subtracting the CPU utilization rate from the CPU availability rate and then performing weighted calculations.

[0054] The memory availability rate is subtracted from the memory utilization rate, and a weighted calculation is performed to determine the memory availability coefficient.

[0055] Based on the sum of the CPU availability coefficient and the memory availability coefficient, the data processing capability coefficient and the data processing capability reference coefficient are determined respectively.

[0056] The data processing capability impact coefficient is determined based on the ratio of the data processing capability coefficient to the data processing capability reference coefficient.

[0057] The resource allocation adjustment coefficient is determined by summing the weighted results of the time adjustment factor, the intermittency coefficient, and the data processing capability influence coefficient.

[0058] Furthermore: Based on the result of the resource allocation adjustment coefficient, if the resource allocation adjustment coefficient increases, then resource allocation is increased; if the resource allocation adjustment coefficient decreases, then resource allocation is decreased.

[0059] Further: In the initial stage, the error rate is determined based on the ratio of the current number of errors to the total number of transformations;

[0060] The error rate is determined by the average of the ratio of the number of errors that have occurred to the total number of conversions, based on the number of repetitions of the feedback control module.

[0061] The effects of the above solution are as follows:

[0062] This invention determines the comprehensive cost of protocol conversion based on the evaluation of protocol conversion by comprehensively considering the weight allocation of factors such as the communication scale of the old protocol, the degree of protocol difference, the protocol conversion time data, and the error rate. This enables the system to comprehensively evaluate the cost in the protocol conversion process by taking into account the characteristics of communication behavior and protocol interaction mode, thereby avoiding resource waste or low processing efficiency caused by incomplete consideration of a single factor.

[0063] The time adjustment factor is determined by considering the overall cost of protocol conversion, protocol conversion time data, and burst traffic coefficient. In particular, based on data fusion and feature enhancement methods, the classified data is further integrated and optimized to extract more representative and effective time factors and burst traffic coefficients related to communication, so as to enhance data features and more accurately reflect the device status. Moreover, through this time adjustment factor, the system can more accurately adjust the time synchronization between the client and the server, thereby improving the real-time performance and accuracy of data transmission and reducing data transmission errors caused by time asynchrony.

[0064] Based on the resource allocation adjustment coefficient determined by comprehensively considering factors such as time adjustment factor, intermittency coefficient, and data processing capacity coefficient, the system can dynamically adjust resource allocation to ensure that CPU, memory, and network bandwidth resources can be reasonably allocated under different network environments and system states, thereby improving resource utilization and enhancing system stability and processing capacity.

[0065] In addition, during the cyclical feedback process, the feedback loop ends when the error rate reaches 0.05. This prevents the system from continuing to consume resources when it cannot effectively reduce the error rate, allowing the resources to be used for other more meaningful tasks, thereby improving processing efficiency and ensuring system stability. Attached Figure Description

[0066] Figure 1 A flowchart illustrating the intelligent allocation of system resources for this computer network communication data processing system;

[0067] Figure 2 This is a schematic diagram of the overall modules in the system of the present invention;

[0068] Figure 3 This is a schematic diagram of the equipment and corresponding data collected by the data acquisition module in this invention. Detailed Implementation

[0069] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings.

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

[0071] S1. The data acquisition module collects the communication scale value of the old protocol, the comparison information of the data packets of the old and new protocols, the protocol conversion time data, the conversion error information, the network traffic data, the CPU utilization rate and the memory utilization rate;

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

[0073] S2.2 Determine the error rate based on the conversion error information;

[0074] S2.3. Based on the weighted allocation of the old protocol communication scale value, protocol difference degree, protocol conversion time data and error rate, determine the comprehensive cost of protocol conversion based on the evaluation protocol conversion;

[0075] S3.1. Based on network traffic data, determine the burst traffic coefficient and intermittency coefficient of network communication;

[0076] S3.2. Based on the weighted calculation of the comprehensive cost of protocol conversion, protocol conversion time data, and burst traffic coefficient, determine the time adjustment factor for client-server time synchronization;

[0077] S4.1 Determine the data processing capability coefficient based on CPU utilization and memory utilization;

[0078] S4.2. Based on the time adjustment factor, intermittency coefficient, and data processing capacity coefficient, determine the resource allocation adjustment coefficient allocated to data processing by the adjustment system;

[0079] S5. The resource allocation module adjusts the system's resource allocation based on the resource allocation adjustment coefficient.

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

[0081] The data acquisition module includes network monitoring equipment, protocol analysis software, system clocks for clients and servers, network traffic monitoring equipment, and system monitoring software;

[0082] Network monitoring equipment is used to collect old protocol communication scale values ​​and conversion error information. The network monitoring equipment obtains the old protocol communication scale value and communication scale reference value by statistically summing the number of data packets bytes of the old protocol and the historical protocol. The conversion error information includes the number of errors and the total number of conversions.

[0083] Protocol analysis software is used to collect comparison information of data packets between new and old protocols. The comparison information includes the number of fields, field types, and field lengths in the new protocol, as well as the number of fields, field types, and field lengths in the old protocol.

[0084] The system clocks of the client and server are used to collect protocol conversion time data, which includes protocol conversion start time, protocol conversion end time, average protocol conversion time, client local time, server time, and the maximum allowed time difference of the system.

[0085] Network traffic monitoring devices are used to collect network traffic data, which includes the average network traffic over the same communication duration, peak network traffic, number of network communication interruptions, and duration of each interruption.

[0086] System monitoring software is used to collect CPU usage and memory usage.

[0087] This embodiment acquires parameters through a data acquisition module and related equipment. The data processing module then processes and calculates the cost assessment, time synchronization adjustment factor, and resource allocation adjustment coefficient. Finally, it determines the calculation results, including the comprehensive cost of protocol conversion, time adjustment factor, and resource allocation adjustment coefficient. These calculation results guide the resource allocation module to optimize resource allocation, improve processing efficiency, ensure the accuracy of time synchronization and system stability, enable the system to effectively cope with complex network environments, and improve overall performance.

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

[0089] The communication scale coefficient is determined based on the ratio of the old protocol communication scale value to the communication scale reference value;

[0090] Protocol analysis software analyzes the data packet format and instruction set of both new and old protocols;

[0091] Compare the number of fields, field types, and field lengths of different fields in the old and new protocols;

[0092] The degree of difference is quantified according to preset rules, and the differences in the number of fields, the types of fields, and the length of fields are determined.

[0093] Set different weights for different types of difference values, including weights for differences in the number of fields, differences in the type of fields, and differences in the length of fields;

[0094] The protocol difference degree is determined by multiplying and adding the different weights of different types of difference values;

[0095] The formula for calculating the protocol difference is as follows:

[0096] CY = a × c1 + b × c2 + c × c3;

[0097] CY represents the protocol difference degree, c1 represents the difference in the number of fields, c2 represents the difference in the field type, c3 represents the difference in the field length, a represents the weight of the difference in the number of fields, b represents the weight of the difference in the field type, and c represents the weight of the difference in the field length.

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

[0099] The error rate is determined based on the ratio of the number of errors to the total number of conversions.

[0100] The conversion time difference is determined based on the difference between the end time and the start time of the protocol conversion.

[0101] The conversion time impact coefficient is determined based on the ratio of the conversion time difference to the average protocol conversion time.

[0102] The overall cost of protocol conversion is determined by summing the weighted results of the communication scale coefficient, protocol difference, error rate, and conversion time impact coefficient.

[0103] First, in this embodiment, the formula for calculating the overall cost of protocol conversion is as follows:

[0104] ;

[0105] The formula for calculating the error rate is as follows:

[0106] CC=C err / C total ;

[0107] in:

[0108] XC represents the overall cost of protocol conversion;

[0109] G old For the old protocol communication scale value, G norm T is a reference value for communication scale. trans To convert the time difference, T norm The average time for protocol conversion is given by , CC is the error rate, and w1, w2, w3, and w4 are all weighting coefficients, with w1+w2+w3+w4=1.

[0110] C err C represents the number of errors. total This represents the total number of conversions.

[0111] In this embodiment, the larger the data volume, the more resources and time are typically required for protocol conversion. The larger the partial calculation result, the greater the protocol conversion comprehensive cost XC, reflecting the positive impact of data size on the protocol conversion comprehensive cost XC;

[0112] The greater the protocol difference (CY), the higher the difficulty and cost of protocol conversion. The larger the partial calculation result, the greater the protocol conversion comprehensive cost XC, which reflects the significant impact of protocol difference on the protocol conversion comprehensive cost XC.

[0113] A longer conversion time lag means a longer period of time system resources are occupied, resulting in higher costs. The larger the partial calculation result, the greater the protocol conversion comprehensive cost XC, reflecting the impact of protocol conversion time on the protocol conversion comprehensive cost XC;

[0114] A higher error rate (CC) requires more time and resources to handle errors, resulting in higher protocol conversion costs. The larger the partial calculation result, the greater the protocol conversion comprehensive cost XC, indicating that the error rate has a positive impact on the protocol conversion comprehensive cost XC.

[0115] Please see Figure 1 and Figure 3 The specific process for determining the burst flow coefficient and the intermittent flow coefficient is as follows:

[0116] The burst traffic coefficient is determined based on the ratio of peak network traffic to average network traffic. The formula for calculating the burst traffic coefficient is as follows:

[0117] LT=F peak / F avg ;

[0118] LT is the burst flow coefficient, F peak For peak network traffic, F avg This represents the average network traffic.

[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's local time and the server time.

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

[0122] The time adjustment factor is determined by summing the weights of the system time difference impact coefficient, the comprehensive cost of protocol conversion, and the burst traffic coefficient.

[0123] In this embodiment, the formula for calculating the time adjustment factor is as follows:

[0124] ;

[0125] in:

[0126] TB is the time adjustment factor;

[0127] T k T is the client's local time. f For server time, T norm-sync To represent the maximum allowable time difference for the system, w5, w6, and w7 are all weighting coefficients, and w5+w6+w7=1.

[0128] The larger the system time difference, the more out of sync the client and server times are, and the greater the adjustments required. The larger the partial calculation result, the larger the time adjustment factor TB will be, which reflects the importance of the time difference for time synchronization adjustment;

[0129] Similarly, the higher the overall cost of protocol conversion (XC), the greater the impact on time synchronization, requiring a larger adjustment factor to ensure the accuracy of time synchronization. The larger the partial calculation result, the larger the time adjustment factor TB will be;

[0130] A larger burst traffic coefficient LT may lead to increased network latency and jitter, affecting the accuracy of time synchronization. This requires a larger adjustment factor to compensate for the loss. Therefore, the larger the result of this part of the calculation, the larger the time adjustment factor TB will be.

[0131] Please see Figure 1 and Figure 3 The system clocks of the client and server collect samples of the time difference between the client and the server over a period of time.

[0132] Sort the time difference samples, take the maximum value after sorting, and determine the maximum allowable time difference of the system;

[0133] The intermittency coefficient is determined by summing the product of the number of network communication interruptions within the same communication duration and the duration of each interruption, and then dividing by the same communication duration.

[0134] The specific processes of steps S3.1 to S3.2 are as follows:

[0135] Collect data processing capability samples of the system under normal operating conditions using system monitoring software;

[0136] Analyze the distribution of data processing capability samples and select the average value of the data processing capability samples as the reference coefficient for data processing capability.

[0137] The data processing capability coefficient and the data processing capability reference coefficient are obtained respectively through CPU utilization and memory utilization, as detailed below:

[0138] The CPU availability coefficient is determined by subtracting the CPU utilization rate from the CPU availability rate and then performing weighted calculations.

[0139] The memory availability coefficient is determined by subtracting the memory utilization rate from the memory availability rate and then applying weighted calculations.

[0140] Based on the sum of the CPU availability coefficient and the memory availability coefficient, the data processing capability coefficient and the data processing capability reference coefficient are determined respectively.

[0141] The data processing capability impact coefficient is determined based on the ratio of the data processing capability coefficient to the data processing capability reference coefficient.

[0142] The resource allocation adjustment coefficient is determined by summing the weighted results of the time adjustment factor, the intermittency coefficient, and the data processing capability influence coefficient.

[0143] First, in this embodiment, the formula for calculating the intermittency coefficient is as follows:

[0144] ;

[0145] J is the intermittency coefficient, d is the number of network communication interruptions, and dt is the intermittency coefficient. i Let T be the duration of the i-th interrupt in each interruption, and T be the duration of the same communication.

[0146] Among them, the number of network communication interruptions d includes the number of short-term jitter interruptions;

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

[0148] in:

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

[0150] The formula for calculating the resource allocation adjustment coefficient is as follows:

[0151] ;

[0152] The formula for calculating the data processing capability coefficient is as follows:

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

[0154] in:

[0155] FZ is the resource allocation adjustment coefficient;

[0156] C proc C represents the data processing capability coefficient. norm The data processing capability reference coefficients are w8, w9, and w10, which are all weighting coefficients, and w8+w9+w10=1;

[0157] u cpu For CPU utilization, u mem For memory usage, w cpu w is the CPU weighting coefficient. mem Let w be the memory weight coefficient. cpu +w mem =1;

[0158] The larger the time synchronization adjustment factor TB, the more resources are needed for time synchronization adjustment. The larger the partial calculation result, the larger the resource allocation adjustment coefficient FZ will be, reflecting the impact of time synchronization on resource allocation;

[0159] The larger the intermittency coefficient J, the more unstable the network communication, and the more resources are needed to cope with this instability. The larger the partial calculation result, the larger the resource allocation adjustment coefficient FZ will be, reflecting the impact of network intermittency on resource allocation;

[0160] When the system's current data processing capacity is low The value is relatively small, therefore more resources are needed to improve data processing capabilities. Some calculation results can affect the size of the resource allocation adjustment coefficient FZ, thereby adjusting resource allocation.

[0161] Example 2, please refer to Figure 1 Based on the results of the resource allocation adjustment coefficient, if the resource allocation adjustment coefficient increases, the resource allocation will increase; if the resource allocation adjustment coefficient decreases, the resource allocation will decrease.

[0162] In the initial stage, the error rate is determined based on the ratio of the current number of errors to the total number of conversions;

[0163] The error rate is determined by the average of the ratio of the number of errors that have occurred to the total number of conversions, based on the number of repetitions of the feedback control module.

[0164] In this embodiment, if the error rate (CC) remains high, continuing the loop feedback would waste a significant amount of system resources. Setting the loop feedback to end when the error rate reaches 0.05 prevents the system from consuming resources when it cannot effectively reduce the error rate, allowing resources to be used for other more meaningful tasks. Furthermore, ending the loop feedback in a timely manner prevents the system from getting stuck in an ineffective loop, enabling the system to adjust its processing strategy more quickly and improve overall processing efficiency. Simultaneously, after ending the loop feedback, the system can take other measures to further reduce the error rate (CC) based on the current CC status. In addition, an excessively high error rate (CC) may affect system stability; ending the loop feedback when the error rate (CC) reaches 0.05 helps ensure system stability and reduces the negative impact of an excessively high error rate (CC).

[0165] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A computer network communication data processing system based on artificial intelligence, characterized in that: The computer network communication data processing system includes a data acquisition module, a data processing module, a resource allocation module, and a feedback control module. The specific implementation steps are as follows: S1. The data acquisition module collects the communication scale value of the old protocol, the comparison information of the data packets of the new and old protocols, the protocol conversion time data, the conversion error information, the network traffic data, the CPU utilization rate and the memory utilization rate; S2.1 The data processing module determines the degree of protocol difference based on the comparison information of the old and new protocol data packets; S2.

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

3. The old protocol communication scale value, protocol difference degree, protocol conversion time data and error rate are weighted and summed to obtain the comprehensive cost of protocol conversion; S3.1 Based on the network traffic data, determine the burst traffic coefficient and intermittency coefficient of network communication; The formula for calculating the intermittency coefficient is as follows: ； Where J is the intermittency coefficient, d is the number of network communication interruptions, and dt is the frequency of network communication interruptions. i Let T be the duration of the i-th interrupt in each interruption, and T be the duration of the same communication. S3.2, through calculation formula The time adjustment factor is obtained, where TB is the time adjustment factor and T is the time adjustment factor. k T is the client's local time. f For server time, T norm-sync The maximum allowable time difference for the system is LT, which is the burst traffic coefficient, the comprehensive cost of protocol conversion is XC, and w5, w6 and w7 are all weighting coefficients, and w5+w6+w7=1; S4.

1. Based on the CPU utilization rate and the memory utilization rate, determine the data processing capability coefficient; S4.

2. The time adjustment factor, intermittency coefficient, and data processing capability coefficient are weighted and summed to obtain the resource allocation adjustment coefficient; The formula for calculating the resource allocation adjustment coefficient is as follows: ； Where FZ is the resource allocation adjustment coefficient, and C proc C represents the data processing capability coefficient. norm TB is the reference coefficient for data processing capability, J is the time adjustment factor, w8, w9 and w10 are all weighting coefficients, and w8+w9+w10=1; S5. The resource allocation module adjusts the system's resource allocation 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 data acquisition module includes network monitoring equipment, protocol analysis software, system clocks for 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 old and new protocol data packets; The system clocks of the client and 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 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 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, as well as the number of fields, field types, and field lengths in the old protocol; The protocol conversion time data includes protocol conversion start time, protocol conversion end time, average protocol conversion time, client local time, server time, and the system's maximum allowed time difference; The conversion error information includes the number of errors and the total number of conversions; The network traffic data includes the average network traffic, peak network traffic, number of network communication interruptions, and 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: The communication scale coefficient is determined based on the 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 old and new protocols; Compare the number of fields, field types, and field lengths in the new and old protocols. The degree of difference is quantified according to preset rules, and the differences in the number of fields, the types of fields, and the length of fields are determined. Set different weights for different types of difference values, including weights for differences in the number of fields, differences in the type of fields, and differences in the length of fields; The protocol difference degree is determined by the formula CY=a×c1+b×c2+c×c3, where CY is the protocol difference degree, c1 is the difference value of the number of fields, c2 is the difference value of the field type, c3 is the difference value of the field length, a is the weight of the difference of the number of fields, b is the weight of the difference of the field type, and c is the weight of the difference of the field length. The error rate is determined based on the ratio of the number of errors to the total number of conversions. The conversion time difference is determined based on the difference between the protocol conversion end time and the protocol conversion start time; The conversion time impact coefficient is determined based on the ratio of the conversion time difference to the average conversion time of the protocol. The overall cost of protocol conversion is obtained by weighting and summing the communication scale coefficient, protocol difference, error rate, and 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 for the burst flow coefficient and the intermittent flow coefficient is as follows: The burst traffic coefficient is determined based on the ratio of the peak network traffic to the average network traffic. The intermittency coefficient is determined by summing the product of the number of network communication interruptions within the same communication duration and the duration of each interruption, and then dividing 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: The system time difference is determined based on the difference between the client's local time and the server's time. The system time difference influence coefficient is determined based on the ratio of the system time difference to the maximum allowable time difference of the system. The time adjustment factor is obtained by weighting and summing the system time difference impact coefficient, the comprehensive cost of protocol conversion, and the burst traffic coefficient.

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 server collect time difference samples between the client and server over a period of time; The time difference samples are sorted, and the maximum value after sorting is taken to determine the maximum allowable time difference of the system.

8. The computer network communication data processing system based on artificial intelligence according to claim 6, characterized in that: The specific processes of steps S3.1 to S3.2 are as follows: The system monitoring software is used to collect data processing capability samples of the system under normal operating conditions. Analyze the distribution of the data processing capability samples and select the average value of the data processing capability samples as the data processing capability reference coefficient; The data processing capability coefficient and the data processing capability reference coefficient are both obtained through their respective CPU utilization and memory utilization, as detailed below: The CPU availability coefficient is determined by subtracting the CPU utilization rate from the CPU availability rate and then performing weighted calculations. The memory availability rate is subtracted from the memory utilization rate, and a weighted calculation is performed to determine the memory availability coefficient. Based on the sum of the CPU availability coefficient and the memory availability coefficient, the data processing capability coefficient and the data processing capability reference coefficient are determined respectively. The data processing capability impact coefficient is determined based on the ratio of the data processing capability coefficient to the data processing capability reference coefficient. The resource allocation adjustment coefficient is obtained by weighting and summing the time adjustment factor, the intermittency coefficient, and the data processing capability impact coefficient.

9. The computer network communication data processing system based on artificial intelligence according to claim 8, characterized in that: Resource allocation is adjusted according to the magnitude 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 decreased.

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

Images