PCIE data transmission analysis system

Abstract

The invention relates to a PCIE data transmission analysis system, and belongs to the technical field of data analysis. The system comprises an acquisition module, a processing module, a calculation module, an analysis module and a feedback module. Through cooperative work of a plurality of modules, the real-time performance, the accuracy and the response capability of data monitoring are improved, and compared with the current technical means, the system can provide more comprehensive data acquisition, more efficient anomaly detection and a more intelligent feedback mechanism. The improvement not only optimizes the stability and reliability of data transmission, but also reduces the risk of system failure and improves the overall operation efficiency, and a plurality of key performance indexes are incorporated into a calculation formula to calculate and obtain a transmission rate coefficient CSX, a delay analysis coefficient YCX and a verification coefficient XYX, so that the comprehensive evaluation of the data transmission state is realized, and the data transmission efficiency is improved. According to the comprehensive evaluation mode, the system can accurately reflect the performance of data transmission in multiple dimensions, and a firmer foundation is provided for subsequent anomaly detection.

Description

Technical Field

[0001] The invention belongs to the technical field of data analysis, and in particular relates to a PCIE data transmission analysis system. Background Art

[0002] PCI Express (PCIe) is a high-speed serial computer expansion bus standard that is widely used in servers, storage devices, graphics processing units, and other fields. With the growing demand for data transmission, how to achieve efficient and stable data transmission in the PCIe architecture has become a focus of attention, further promoting the demand for data transmission analysis systems, especially in the sub-fields of multi-source data monitoring and anomaly detection.

[0003] In the actual application of PCIe data transmission, traditional data monitoring methods often have problems such as poor real-time performance, insufficient accuracy and low responsiveness, which makes it difficult to detect and handle potential anomalies in data transmission in a timely manner. This not only affects the overall performance and stability of the system, but also increases the risk of failure. Existing technical means often focus on the monitoring of a single performance indicator and lack comprehensive evaluation of multi-dimensional data. Therefore, there is an urgent need for a PCIe data transmission analysis system that can integrate multiple key performance indicators, realize real-time monitoring and intelligent feedback, so as to improve the stability and reliability of data transmission.

[0004] Therefore, the present invention proposes a PCIE data transmission analysis system to solve the above-mentioned problems. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] The technical problem to be solved by the present invention is how to provide a PCIE data transmission analysis system to solve the problem that existing technical means often focus on monitoring a single performance indicator and lack comprehensive evaluation of multi-dimensional data.

[0007] (II) Technical solution

[0008] In order to solve the above technical problems, the present invention provides a PCIE data transmission analysis system, which comprises: a collection module (1), a processing module (2), a calculation module (3), an analysis module (4) and a feedback module (5);

[0009] The acquisition module (1) is used to detect and acquire data during the data transmission process, thereby generating a multi-source data set;

[0010] The processing module (2) is used to preprocess the multi-source data set and organize the preprocessed multi-source data set into a first data set, a second data set and a third data set;

[0011] The calculation module (3) is used to integrate the first data set, the second data set and the third data set to calculate the data anomaly coefficient SJY;

[0012] The analysis module (4) is used to compare the data abnormality coefficient SJY with a preset first threshold value Y, thereby generating a first comparison result, and determining whether the current data transmission is abnormal according to the first comparison result;

[0013] If the first comparison result is an abnormality, the data abnormality coefficient SJY is integrated and calculated with the preset first threshold value Y to generate the data abnormality magnitude coefficient SYL, and the data abnormality magnitude coefficient SYL is compared with the preset second threshold value R to generate a second comparison result, and the abnormal magnitude of the current data transmission is determined according to the second comparison result;

[0014] The feedback module (5) is used to send the first comparison result and the second comparison result to the terminal, and generate corresponding alarm information according to the second comparison result.

[0015] (III) Beneficial effects

[0016] The present invention provides a PCIE data transmission analysis system. Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. This system improves the real-time, accuracy and responsiveness of data monitoring through the collaborative work of multiple modules. Compared with current technical means, this system can provide more comprehensive data collection, more efficient anomaly detection and more intelligent feedback mechanism. These improvements not only optimize the stability and reliability of data transmission, but also reduce the risk of system failure and improve overall operational efficiency.

[0018] 2. By incorporating multiple key performance indicators into the calculation formula, the transmission rate coefficient CSX, delay analysis coefficient YCX and verification coefficient XYX are calculated to achieve a comprehensive evaluation of the data transmission status. This comprehensive evaluation method enables the system to accurately reflect the performance of data transmission in multiple dimensions, providing a more solid foundation for subsequent anomaly detection. By using the intervention reference coefficients of each indicator, smoothing of data changes is achieved. The use of intervention reference coefficients of each indicator reduces the impact of instantaneous fluctuations on the results, ensures the long-term stability of the calculation, and provides a more reliable basis for anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a PCIE data transmission analysis system architecture diagram of the present invention.

[0020] In the figure: 1. acquisition module; 11. first acquisition unit; 12. second acquisition unit; 13. third acquisition unit; 2. processing module; 21. preprocessing unit; 22. sorting unit; 3. calculation module; 31. first calculation unit; 32. second calculation unit; 4. analysis module; 41. first analysis unit; 42. second analysis unit; 5. feedback module. DETAILED DESCRIPTION

[0021] In order to make the purpose, content and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below in conjunction with the drawings and examples.

[0022] The purpose of the present invention is to provide a PCIE data transmission analysis system to solve the problems of poor real-time performance, insufficient accuracy and low responsiveness in traditional data monitoring methods proposed in the above background technology, which makes it difficult to timely discover and handle potential anomalies in data transmission. This not only affects the overall performance and stability of the system, but also increases the risk of failure. Existing technical means often focus on the monitoring of a single performance indicator and lack the problem of comprehensive evaluation of multi-dimensional data.

[0023] The present invention relates to the field of data analysis technology, and discloses a PCIE data transmission analysis system. The system improves the real-time, accuracy and responsiveness of data monitoring through the collaborative work of multiple modules. Compared with current technical means, the system can provide more comprehensive data collection, more efficient anomaly detection and more intelligent feedback mechanism. These improvements not only optimize the stability and reliability of data transmission, but also reduce the risk of system failure and improve overall operational efficiency. By incorporating multiple key performance indicators into the calculation formula, the transmission rate coefficient CSX, the delay analysis coefficient YCX and the validation coefficient XYX are calculated to achieve a comprehensive evaluation of the data transmission status. This comprehensive evaluation method enables the system to accurately reflect the performance of data transmission in multiple dimensions, providing a more solid foundation for subsequent anomaly detection.

[0024] To achieve the above object, the present invention provides the following technical solutions:

[0025] A PCIE data transmission analysis system comprises: a collection module (1), a processing module (2), a calculation module (3), an analysis module (4) and a feedback module (5);

[0026] The acquisition module (1) is used to detect and acquire data during the data transmission process, thereby generating a multi-source data set;

[0027] The processing module (2) is used to preprocess the multi-source data set and organize the preprocessed multi-source data set into a first data set, a second data set and a third data set;

[0028] The calculation module (3) is used to integrate the first data set, the second data set and the third data set to calculate the data anomaly coefficient SJY;

[0029] The analysis module (4) is used to compare the data abnormality coefficient SJY with a preset first threshold value Y, thereby generating a first comparison result, and determining whether the current data transmission is abnormal according to the first comparison result;

[0030] If the first comparison result is an abnormality, the data abnormality coefficient SJY is integrated and calculated with the preset first threshold value Y to generate the data abnormality magnitude coefficient SYL, and the data abnormality magnitude coefficient SYL is compared with the preset second threshold value R to generate a second comparison result, and the abnormal magnitude of the current data transmission is determined according to the second comparison result;

[0031] The feedback module (5) is used to send the first comparison result and the second comparison result to the terminal, and generate corresponding alarm information according to the second comparison result.

[0032] Furthermore, the acquisition module (1) comprises a first acquisition unit (11), a second acquisition unit (12) and a third acquisition unit (13);

[0033] The first acquisition unit (11) is used to acquire bus rate, effective data rate, channel width and peak throughput;

[0034] The second collection unit (12) is used to collect link delay, data packet response time, queue depth and response waiting time;

[0035] The third collection unit (13) is used to collect data packet error rate, retransmission rate, CRC check error rate and packet loss rate;

[0036] The bus rate, effective data rate, channel width, peak throughput, link delay, data packet response time, queue depth, response waiting time, data packet error rate, retransmission rate, CRC check error rate and packet loss rate constitute a multi-source data set.

[0037] Furthermore, the processing module (2) comprises a pre-processing unit (21) and a finishing unit (22);

[0038] The preprocessing unit (21) is used to preprocess the multi-source data set and make it dimensionless, and the sorting unit (22) is used to sort the processed multi-source data set to generate a first data set, a second data set and a third data set;

[0039] The first data set includes bus rate, effective data rate, channel width, and peak throughput;

[0040] The bus rates are denoted as A1, A2, A3, ..., An according to the timestamps;

[0041] The effective data rates are respectively marked as B1, B2, B3, ..., Bn according to the time stamp;

[0042] The channel widths are marked as C1, C2, C3, ..., Cn according to the timestamps;

[0043] The peak throughputs are denoted as D1, D2, D3, ..., Dn according to the timestamps;

[0044] The second data set includes link latency, packet response time, queue depth, and response latency;

[0045] The link delays are denoted as E1, E2, E3, ..., En according to the timestamps;

[0046] The data packet response times are marked as F1, F2, F3, ..., Fn according to the timestamps;

[0047] The queue depths are denoted by timestamp as G1, G2, G3, ..., Gn;

[0048] The response waiting times are marked as H1, H2, H3, ..., Hn according to the timestamps;

[0049] The third data set includes data packet error rate, retransmission rate, CRC check error rate and packet loss rate;

[0050] The packet error rates are denoted as I1, I2, I3, ..., In according to the timestamps;

[0051] The retransmission rates are denoted as J1, J2, J3, ..., Jn according to the timestamps;

[0052] The CRC check error rates are respectively denoted as K1, K2, K3, ..., Kn according to the timestamp;

[0053] The packet loss rates are recorded as L1, L2, L3, ..., Ln according to the timestamps.

[0054] Furthermore, the calculation module (3) comprises a first calculation unit (31) and a second calculation unit (32);

[0055] The first calculation unit (31) is used to calculate the first data set, the second data set and the third data set into a transmission rate coefficient CSX, a delay analysis coefficient YCX and a validation coefficient XYX respectively;

[0056] The second calculation unit (32) is used to calculate the transmission rate coefficient CSX, the delay analysis coefficient YCX and the validation coefficient XYX into a data anomaly coefficient SJY.

[0057] Furthermore, the transmission rate coefficient CSX, the delay analysis coefficient YCX and the validation coefficient XYX are calculated and obtained by the following formulas respectively:

[0058]

[0059] Where: An, Bn, Cn, Dn, En, Fn, Gn, Hn, In, Jn, Kn and Ln are the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate and packet loss rate at the nth timestamp respectively;

[0060] PA is the bus rate intervention reference coefficient, PB is the effective data rate intervention reference coefficient, PC is the channel width intervention reference coefficient, PD is the peak throughput intervention reference coefficient, PE is the link delay intervention reference coefficient, PF is the packet response time intervention reference coefficient, PG is the queue depth intervention reference coefficient, PH is the response waiting time intervention reference coefficient, PI is the packet error rate intervention reference coefficient, PJ is the retransmission rate intervention reference coefficient, PK is the check error rate intervention reference coefficient, and PL is the packet loss rate intervention reference coefficient;

[0061] a1, a2, a3 and a4 are weight values, and the values ​​of a1, a2, a3 and a4 are adjusted and set by the user, and ln is a logarithmic function.

[0062] Further, the bus rate intervention reference coefficient PA, the effective data rate intervention reference coefficient PB, the channel width intervention reference coefficient PC, the peak throughput intervention reference coefficient PD, the link delay intervention reference coefficient PE, the data packet response time intervention reference coefficient PF, the queue depth intervention reference coefficient PG, the response waiting time intervention reference coefficient PH, the data packet error rate intervention reference coefficient PI, the retransmission rate intervention reference coefficient PJ, the check error rate intervention reference coefficient PK, and the packet loss rate intervention reference coefficient PL are respectively calculated and obtained by the following formulas:

[0063]

[0064]

[0065] Where: Ai, Bi, Ci, Di, Ei, Fi, Gi, Hi, Ii, Ji, Ki, and Li are respectively the bus rate, effective data rate, channel width, peak throughput, link latency, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate, and packet loss rate at the i-th timestamp;

[0066] An, Bn, Cn, Dn, En, Fn, Gn, Hn, In, Jn, Kn, and Ln are respectively the bus rate, effective data rate, channel width, peak throughput, link latency, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate, and packet loss rate at the n-th timestamp;

[0067] m is the size of the window, indicating that there are m data points in each window. m consecutive data points are taken to calculate the mean, and ln is the logarithmic function.

[0068] Furthermore, the data anomaly coefficient SJY is calculated and obtained through the following formula:

[0069] SJY = b1 × ln(CSX) + b2 × ln(YCX) + b3 × ln(XYX);

[0070] Where: b1, b2, and b3 are weight values, and the values of b1, b2, and b3 are adjusted and set by the user. CSX is the transmission rate coefficient, YCX is the delay analysis coefficient, and XYX is the check coefficient.

[0071] Furthermore, the analysis module (4) includes a first analysis unit (41) and a second analysis unit (42). The first analysis unit (41) is used to generate a first comparison result, and the second analysis unit (42) is used to generate a second comparison result;

[0072] The first comparison result is as follows:

[0073] When SJY < Y, it represents that the current data transmission is normal;

[0074] When SJY ≥ Y, it represents that the current data transmission has an anomaly;

[0075] The second comparison result is as follows:

[0076] When SYL < R, it represents that the current data transmission is a first-level anomaly;

[0077] When R ≤ SYL < R × 115%, it represents that the current data transmission is a second-level anomaly;

[0078] When SYL ≥ R × 115%, it represents that the current data transmission is a third-level anomaly;

[0079] The data anomaly magnitude coefficient SYL is calculated using the following formula:

[0080]

[0081] Where: SJY is the data anomaly coefficient SJY, and Y is the first threshold.

[0082] Furthermore, the alarm information generated by the feedback module (5) is specifically as follows:

[0083] When the current data transmission is a level 1 abnormality, a yellow alarm message will be sent to the backend maintenance personnel. The alarm message starts with DY and is set to yellow. Before the alarm is lifted, it will be sent repeatedly at an interval of 20 minutes / time. Repairs must be made within one hour. If the repair cannot be completed, data transmission will be forcibly interrupted.

[0084] When the current data transmission is a level 2 abnormality, an orange alarm message will be sent to the backend maintenance personnel. The alarm message starts with DR and is orange in color. It will be sent repeatedly at intervals of five minutes before the alarm is lifted. Repairs must be made within thirty minutes. If the repairs cannot be completed, data transmission will be forcibly interrupted.

[0085] When the current data transmission is at level three abnormality, a red alarm message will be sent to the background maintenance personnel. The alarm message starts with DS and the color is set to red. Before the alarm is lifted, it will be sent repeatedly at an interval of five seconds per time. Repairs are limited to ten minutes. If the repair cannot be completed, data transmission will be forcibly interrupted.

[0086] Embodiment 1:

[0087] See also Figure 1 , a PCIE data transmission analysis system, comprising a collection module 1, a processing module 2, a calculation module 3, an analysis module 4 and a feedback module 5;

[0088] The acquisition module 1 is used to detect and collect data during the data transmission process, thereby generating a multi-source data set;

[0089] The processing module 2 is used to preprocess the multi-source data set and organize the preprocessed multi-source data set into a first data set, a second data set and a third data set;

[0090] The calculation module 3 is used to integrate the first data set, the second data set and the third data set to calculate the data anomaly coefficient SJY;

[0091] The analysis module 4 is used to compare the data abnormality coefficient SJY with a preset first threshold value Y, thereby generating a first comparison result, and determining whether the current data transmission is abnormal according to the first comparison result;

[0092] If the first comparison result is an abnormality, the data abnormality coefficient SJY is integrated and calculated with the preset first threshold value Y to generate the data abnormality magnitude coefficient SYL, and the data abnormality magnitude coefficient SYL is compared with the preset second threshold value R to generate a second comparison result, and the abnormal magnitude of the current data transmission is determined according to the second comparison result;

[0093] The feedback module 5 is used to send the first comparison result and the second comparison result to the terminal, and generate corresponding alarm information according to the second comparison result.

[0094] In this embodiment: the acquisition module 1 is responsible for real-time detection and data acquisition of the PCIe data transmission process to generate a multi-source data set. Through high-frequency monitoring and comprehensive parameter capture, the system can obtain key data including transmission rate, delay, error rate, etc. This real-time data collection ensures the accuracy of subsequent analysis, provides a strong foundation for the system, and promptly identifies potential transmission problems and ensures data integrity.

[0095] The main task of processing module 2 is to preprocess the multi-source data sets and organize them into the first data set, the second data set, and the third data set. By cleaning and integrating data, this module eliminates redundancy and noise and improves data quality. The preprocessed data set provides a reliable input for calculation module 3, ensuring the accuracy and effectiveness of subsequent calculations, and laying a solid foundation for anomaly detection.

[0096] The calculation module 3 integrates and calculates the first data set, the second data set, and the third data set to generate the data anomaly coefficient SJY. This process evaluates the health status of data transmission through quantitative indicators, so that anomaly detection has clear standards. By calculating the anomaly coefficient, the system can identify potential problems at an early stage and provide a basis for subsequent analysis and response measures.

[0097] The main function of the analysis module 4 is to compare the calculated data anomaly coefficient SJY with the preset first threshold value Y to generate a first comparison result, thereby determining whether the current data transmission is abnormal. If it is determined to be abnormal, the module further integrates and calculates the data anomaly magnitude coefficient SYL, and compares it with the second threshold value R to determine the severity of the anomaly. This process ensures the accuracy and reliability of anomaly detection through layer-by-layer filtering and comparison, and responds to changes in system status in a timely manner.

[0098] The feedback module 5 is responsible for sending the first and second comparison results to the terminal, and generating corresponding alarm information according to the second comparison result. Through an effective feedback mechanism, the system can promptly notify the user of the current status and abnormal conditions, so that the user can respond quickly and take necessary measures to solve the problem. This module enhances the user's sense of control and improves the usability and security of the system.

[0099] Through the collaborative work of multiple modules, this system improves the real-time, accuracy and responsiveness of data monitoring. Compared with current technical means, this system can provide more comprehensive data collection, more efficient anomaly detection and more intelligent feedback mechanism. These improvements not only optimize the stability and reliability of data transmission, but also reduce the risk of system failure and improve overall operational efficiency.

[0100] Embodiment 2:

[0101] See also Figure 1 , the acquisition module 1 includes a first acquisition unit 11, a second acquisition unit 12 and a third acquisition unit 13;

[0102] The first acquisition unit 11 is used to acquire bus rate, effective data rate, channel width and peak throughput;

[0103] The second collection unit 12 is used to collect link delay, data packet response time, queue depth and response waiting time;

[0104] The third collection unit 13 is used to collect data packet error rate, retransmission rate, CRC check error rate and packet loss rate;

[0105] The bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate, and packet loss rate constitute the multi-source data set.

[0106] In this embodiment: the acquisition module 1 divides data acquisition into three units, covering multiple key indicators. By refining the acquisition indicators, the system can fully reflect the performance and health status of PCIe transmission and enhance the depth and accuracy of data analysis.

[0107] The first acquisition unit 11, the second acquisition unit 12 and the third acquisition unit 13 are respectively responsible for data monitoring in different dimensions. This multi-dimensional monitoring method enables the system to analyze data transmission performance from multiple angles, discover potential problems in a timely manner, and reduce the occurrence rate of failures. Each acquisition unit can obtain and update data in real time, such as link delay and data packet response time. Real-time evaluation can quickly identify bottlenecks in the transmission process, ensure the efficiency and stability of data transmission, and improve user experience.

[0108] Embodiment three:

[0109] See also Figure 1 , the processing module 2 includes a pre-processing unit 21 and a finishing unit 22;

[0110] The preprocessing unit 21 is used to preprocess the multi-source data set and make it dimensionless, and the sorting unit 22 is used to sort the processed multi-source data set to generate a first data set, a second data set, and a third data set;

[0111] The first data set includes bus rate, effective data rate, channel width, and peak throughput;

[0112] The bus rates are denoted as A1, A2, A3, ..., An according to the timestamps;

[0113] The effective data rates are respectively marked as B1, B2, B3, ..., Bn according to the time stamp;

[0114] The channel widths are marked as C1, C2, C3, ..., Cn according to the timestamps;

[0115] The peak throughputs are denoted as D1, D2, D3, ..., Dn according to the timestamps;

[0116] The second data set includes link latency, packet response time, queue depth, and response latency;

[0117] The link delays are denoted as E1, E2, E3, ..., En according to the timestamps;

[0118] The data packet response times are marked as F1, F2, F3, ..., Fn according to the timestamps;

[0119] The queue depths are denoted by timestamp as G1, G2, G3, ..., Gn;

[0120] The response waiting times are marked as H1, H2, H3, ..., Hn according to the timestamps;

[0121] The third data set includes data packet error rate, retransmission rate, CRC check error rate and packet loss rate;

[0122] The packet error rates are denoted as I1, I2, I3, ..., In according to the timestamps;

[0123] The retransmission rates are denoted as J1, J2, J3, ..., Jn according to the timestamps;

[0124] The CRC check error rates are respectively denoted as K1, K2, K3, ..., Kn according to the timestamp;

[0125] The packet loss rates are recorded as L1, L2, L3, ..., Ln according to the timestamps.

[0126] In this embodiment: the preprocessing unit 21 performs dimensionless processing on the multi-source data set to eliminate the influence of different data units. This processing method enhances the comparability and consistency of the data, facilitates subsequent calculation and analysis, and improves the accuracy of the results.

[0127] The sorting unit 22 is responsible for sorting the processed multi-source data sets into three structured data sets. The structured data sets provide a clear framework for subsequent calculations and analysis, making the data processing process more efficient and orderly. The introduction of timestamps enables the data to reflect dynamic changes, which helps to analyze performance trends and identify time-related anomalies, thereby enhancing the timeliness of the data. This comprehensive coverage of indicators provides a rich data foundation for subsequent calculations, ensuring that system performance can be evaluated from multiple dimensions and potential problems can be discovered in a timely manner.

[0128] Embodiment 4:

[0129] See also Figure 1 , the calculation module 3 includes a first calculation unit 31 and a second calculation unit 32;

[0130] The first calculation unit 31 is used to calculate the first data set, the second data set and the third data set into a transmission rate coefficient CSX, a delay analysis coefficient YCX and a validation coefficient XYX respectively;

[0131] The second calculation unit 32 is used to calculate the transmission rate coefficient CSX, the delay analysis coefficient YCX and the validation coefficient XYX into a data anomaly coefficient SJY.

[0132] In this embodiment: the calculation module 3 is divided into a first calculation unit 31 and a second calculation unit 32 to implement hierarchical calculation. This structure enables different types of data to be processed in a targeted manner, optimizes the calculation process, and enhances the flexibility and scalability of the module.

[0133] The first calculation unit 31 calculates the transmission rate coefficient, delay analysis coefficient and verification coefficient for the first data set, the second data set and the third data set respectively. By calculating specific performance indicators, the system can more accurately evaluate the overall performance and health status of data transmission, and provide a clear basis for subsequent anomaly detection.

[0134] The second calculation unit 32 integrates and calculates various performance indicators into a data anomaly coefficient. The unified anomaly detection standard enables the system to quickly identify potential problems in data transmission through the quantified data anomaly coefficient, thereby improving the accuracy and efficiency of anomaly detection.

[0135] Embodiment five:

[0136] See also Figure 1The transmission rate coefficient CSX, delay analysis coefficient YCX and validation coefficient XYX are calculated by the following formulas:

[0137]

[0138] Where: An, Bn, Cn, Dn, En, Fn, Gn, Hn, In, Jn, Kn and Ln are the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate and packet loss rate at the nth timestamp respectively;

[0139] PA is the bus rate intervention reference coefficient, PB is the effective data rate intervention reference coefficient, PC is the channel width intervention reference coefficient, PD is the peak throughput intervention reference coefficient, PE is the link delay intervention reference coefficient, PF is the packet response time intervention reference coefficient, PG is the queue depth intervention reference coefficient, PH is the response waiting time intervention reference coefficient, PI is the packet error rate intervention reference coefficient, PJ is the retransmission rate intervention reference coefficient, PK is the check error rate intervention reference coefficient, and PL is the packet loss rate intervention reference coefficient;

[0140] a1, a2, a3 and a4 are weight values, and the values ​​of a1, a2, a3 and a4 are adjusted and set by the user, and ln is a logarithmic function.

[0141] The bus rate intervention reference coefficient PA, the effective data rate intervention reference coefficient PB, the channel width intervention reference coefficient PC, the peak throughput intervention reference coefficient PD, the link delay intervention reference coefficient PE, the packet response time intervention reference coefficient PF, the queue depth intervention reference coefficient PG, the response waiting time intervention reference coefficient PH, the packet error rate intervention reference coefficient PI, the retransmission rate intervention reference coefficient PJ, the check error rate intervention reference coefficient PK, and the packet loss rate intervention reference coefficient PL are calculated and obtained by the following formulas:

[0142]

[0143]

[0144] Where: Ai, Bi, Ci, Di, Ei, Fi, Gi, Hi, Ii, Ji, Ki and Li are the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate and packet loss rate at the i-th timestamp respectively;

[0145] An, Bn, Cn, Dn, En, Fn, Gn, Hn, In, Jn, Kn and Ln are the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate and packet loss rate at the nth timestamp respectively;

[0146] m is the size of the window, which means there are m data points in each window. m data points are taken continuously to calculate the mean, and ln is the logarithmic function.

[0147] In this embodiment: by incorporating multiple key performance indicators into the calculation formula, the transmission rate coefficient CSX, the delay analysis coefficient YCX and the verification coefficient XYX are calculated to achieve a comprehensive evaluation of the data transmission status. This comprehensive evaluation method enables the system to accurately reflect the performance of data transmission in multiple dimensions, providing a more solid foundation for subsequent anomaly detection.

[0148] Flexible weight settings enable the system to adapt to different scenarios and needs, enhance the system's customizability and adaptability, and ensure its effectiveness in various working environments. The logarithmic function ln is used in the calculation formula to effectively handle fluctuations in a large range of data. The introduction of the logarithmic function can suppress the impact of extreme values ​​on the calculation results, improve the stability and reliability of coefficient calculation, and make anomaly detection more accurate.

[0149] By using the intervention reference coefficient of each indicator, smoothing of data changes is achieved. The use of the intervention reference coefficient of each indicator reduces the impact of instantaneous fluctuations on the results, ensures the long-term stability of the calculation, and provides a more reliable basis for anomaly detection.

[0150] Embodiment six:

[0151] See also Figure 1 , the data anomaly coefficient SJY is calculated by the following formula:

[0152] SJY=b1×ln(CSX)+b2×ln(YCX)+b3×ln(XYX);

[0153] Wherein: b1, b2 and b3 are weight values, and the values ​​of b1, b2 and b3 are adjusted and set by the user, CSX is the transmission rate coefficient, YCX is the delay analysis coefficient, and XYX is the validation coefficient.

[0154] In this embodiment, the data anomaly coefficient SJY is obtained by comprehensively calculating the transmission rate coefficient CSX, the delay analysis coefficient YCX, and the verification coefficient XYX, enabling the data anomaly coefficient SJY to comprehensively reflect potential anomalies in the data transmission process, enhancing the accuracy of anomaly detection, and enabling the system to promptly identify possible problems.

[0155] The flexible weight setting enables the system to adapt to different application scenarios and performance requirements, enhancing the personalization and adaptability of the system and ensuring effectiveness in various working environments.

[0156] When calculating the data anomaly coefficient SJY, the logarithmic function ln is used to process coefficients of different magnitudes. The introduction of the logarithmic function helps reduce the impact of data fluctuations on the result, thereby improving the stability and reliability of the anomaly coefficient calculation and ensuring that the system can still effectively monitor anomalies under changing conditions.

[0157] Embodiment Seven:

[0158] Please refer to Figure 1 , the analysis module 4 includes a first analysis unit 41 and a second analysis unit 42. The first analysis unit 41 is used to generate a first comparison result, and the second analysis unit 42 is used to generate a second comparison result;

[0159] The first comparison result is as follows:

[0160] When SJY < Y, it represents that the current data transmission is normal;

[0161] When SJY ≥ Y, it represents that the current data transmission has an anomaly;

[0162] The second comparison result is as follows:

[0163] When SYL < R, it represents that the current data transmission is a first-level anomaly;

[0164] When R ≤ SYL < R × 115%, it represents that the current data transmission is a second-level anomaly;

[0165] When SYL ≥ R × 115%, it represents that the current data transmission is a third-level anomaly;

[0166] The data anomaly magnitude coefficient SYL is obtained by calculating with the following formula:

[0167]

[0168] In the formula: SJY is the data anomaly coefficient SJY, and Y is the first threshold.

[0169] The alarm information generated by the feedback module 5 is specifically as follows:

[0170] When the current data transmission is a level 1 abnormality, a yellow alarm message will be sent to the backend maintenance personnel. The alarm message starts with DY and is set to yellow. Before the alarm is lifted, it will be sent repeatedly at an interval of 20 minutes / time. Repairs must be made within one hour. If the repair cannot be completed, data transmission will be forcibly interrupted.

[0171] When the current data transmission is a level 2 abnormality, an orange alarm message will be sent to the backend maintenance personnel. The alarm message starts with DR and is orange in color. It will be sent repeatedly at intervals of five minutes before the alarm is lifted. Repairs must be made within thirty minutes. If the repairs cannot be completed, data transmission will be forcibly interrupted.

[0172] When the current data transmission is at level three abnormality, a red alarm message will be sent to the background maintenance personnel. The alarm message starts with DS and the color is set to red. Before the alarm is lifted, it will be sent repeatedly at an interval of five seconds per time. Repairs are limited to ten minutes. If the repair cannot be completed, data transmission will be forcibly interrupted.

[0173] In this embodiment: through the first analysis unit 41 and the second analysis unit 42, the system can generate layered abnormality comparison results. This layered detection mechanism enables abnormalities of different degrees to be accurately identified, facilitates the adoption of corresponding measures, and enhances the flexibility and responsiveness of the system.

[0174] Through specific thresholds and formula calculations, the system can clearly define abnormal states of data transmission. Clear standardized detection reduces ambiguity, ensures a quick and accurate response to abnormal situations, and improves system reliability.

[0175] According to the abnormality level, the feedback module 5 will send alarm information of different colors and frequencies. This dynamic alarm mechanism can promptly notify maintenance personnel, enhance the monitoring and management of abnormal situations, help to quickly take repair measures, and reduce system risks. Different levels of abnormalities have corresponding repair time limits and subsequent processing measures. This hierarchical processing strategy ensures that appropriate response measures are taken in different abnormal situations, avoids potential serious consequences, and improves the overall safety of the system.

[0176] Compared with the prior art, the present invention has the following beneficial effects:

[0177] 1. This system improves the real-time, accuracy and responsiveness of data monitoring through the collaborative work of multiple modules. Compared with current technical means, this system can provide more comprehensive data collection, more efficient anomaly detection and more intelligent feedback mechanism. These improvements not only optimize the stability and reliability of data transmission, but also reduce the risk of system failure and improve overall operational efficiency.

[0178] 2. By incorporating multiple key performance indicators into the calculation formula, the transmission rate coefficient CSX, delay analysis coefficient YCX and verification coefficient XYX are calculated to achieve a comprehensive evaluation of the data transmission status. This comprehensive evaluation method enables the system to accurately reflect the performance of data transmission in multiple dimensions, providing a more solid foundation for subsequent anomaly detection. By using the intervention reference coefficients of each indicator, smoothing of data changes is achieved. The use of intervention reference coefficients of each indicator reduces the impact of instantaneous fluctuations on the results, ensures the long-term stability of the calculation, and provides a more reliable basis for anomaly detection.

[0179] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A PCIE data transmission analysis system, characterized in that: The system comprises: a collection module (1), a processing module (2), a calculation module (3), an analysis module (4) and a feedback module (5); The acquisition module (1) is used to detect and acquire data during the data transmission process, thereby generating a multi-source data set; The processing module (2) is used to preprocess the multi-source data set and organize the preprocessed multi-source data set into a first data set, a second data set and a third data set; The calculation module (3) is used to integrate the first data set, the second data set and the third data set to calculate the data anomaly coefficient SJY; The analysis module (4) is used to compare the data abnormality coefficient SJY with a preset first threshold value Y, thereby generating a first comparison result, and determining whether the current data transmission is abnormal according to the first comparison result; If the first comparison result is an abnormality, the data abnormality coefficient SJY is integrated and calculated with the preset first threshold value Y to generate the data abnormality magnitude coefficient SYL, and the data abnormality magnitude coefficient SYL is compared with the preset second threshold value R to generate a second comparison result, and the abnormal magnitude of the current data transmission is determined according to the second comparison result; The feedback module (5) is used to send the first comparison result and the second comparison result to the terminal, and generate corresponding alarm information according to the second comparison result.

2. The PCIE data transmission analysis system according to claim 1, characterized in that: The acquisition module (1) comprises a first acquisition unit (11), a second acquisition unit (12) and a third acquisition unit (13); The first acquisition unit (11) is used to acquire bus rate, effective data rate, channel width and peak throughput; The second collection unit (12) is used to collect link delay, data packet response time, queue depth and response waiting time; The third collection unit (13) is used to collect data packet error rate, retransmission rate, CRC check error rate and packet loss rate; The bus rate, effective data rate, channel width, peak throughput, link delay, data packet response time, queue depth, response waiting time, data packet error rate, retransmission rate, CRC check error rate and packet loss rate constitute a multi-source data set.

3. The PCIE data transmission analysis system according to claim 2, characterized in that: The processing module (2) comprises: a pre-processing unit (21) and a finishing unit (22); The preprocessing unit (21) is used to preprocess and dimensionlessly transform the multi-source data set; The collating unit (22) is used to sort the processed multi-source data sets, thereby generating a first data set, a second data set and a third data set.

4. The PCIE data transmission analysis system according to claim 3, characterized in that: The first data set includes bus rate, effective data rate, channel width, and peak throughput; The bus rates are denoted as A1, A2, A3, ..., An according to the timestamps; The effective data rates are respectively marked as B1, B2, B3, ..., Bn according to the time stamp; The channel widths are marked as C1, C2, C3, ..., Cn according to the timestamps; The peak throughputs are denoted as D1, D2, D3, ..., Dn according to the timestamps; The second data set includes link latency, packet response time, queue depth, and response latency; The link delays are denoted as E1, E2, E3, ..., En according to the timestamps; The data packet response times are marked as F1, F2, F3, ..., Fn according to the timestamps; The queue depths are denoted by timestamp as G1, G2, G3, ..., Gn; The response waiting times are marked as H1, H2, H3, ..., Hn according to the timestamps; The third data set includes data packet error rate, retransmission rate, CRC check error rate and packet loss rate; The packet error rates are denoted as I1, I2, I3, ..., In according to the timestamps; The retransmission rates are denoted as J1, J2, J3, ..., Jn according to the timestamps; The CRC check error rates are respectively denoted as K1, K2, K3, ..., Kn according to the timestamp; The packet loss rates are recorded as L1, L2, L3, ..., Ln according to the timestamps.

5. The PCIE data transmission analysis system according to claim 4, characterized in that: The calculation module (3) comprises a first calculation unit (31) and a second calculation unit (32); The first calculation unit (31) is used to calculate the first data set, the second data set and the third data set into a transmission rate coefficient CSX, a delay analysis coefficient YCX and a validation coefficient XYX respectively; The second calculation unit (32) is used to calculate the transmission rate coefficient CSX, the delay analysis coefficient YCX and the validation coefficient XYX into a data anomaly coefficient SJY.

6. The PCIE data transmission analysis system as claimed in claim 5, characterized in that ,, the transmission rate coefficient CSX, delay analysis coefficient YCX and validation coefficient XYX are calculated and obtained by the following formulas respectively: Where: An, Bn, Cn, Dn, En, Fn, Gn, Hn, In, Jn, Kn and Ln are the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC check error rate and packet loss rate at the nth timestamp respectively; PA is the bus rate intervention reference coefficient, PB is the effective data rate intervention reference coefficient, PC is the channel width intervention reference coefficient, PD is the peak throughput intervention reference coefficient, PE is the link delay intervention reference coefficient, PF is the packet response time intervention reference coefficient, PG is the queue depth intervention reference coefficient, PH is the response waiting time intervention reference coefficient, PI is the packet error rate intervention reference coefficient, PJ is the retransmission rate intervention reference coefficient, PK is the check error rate intervention reference coefficient, and PL is the packet loss rate intervention reference coefficient; a1, a2, a3 and a4 are weight values, and the values ​​of a1, a2, a3 and a4 are adjusted and set by the user, and ln is a logarithmic function.

7. The PCIE data transmission analysis system according to claim 6, characterized in that: The bus rate intervention reference coefficient PA, the effective data rate intervention reference coefficient PB, the channel width intervention reference coefficient PC, the peak throughput intervention reference coefficient PD, the link delay intervention reference coefficient PE, the packet response time intervention reference coefficient PF, the queue depth intervention reference coefficient PG, the response waiting time intervention reference coefficient PH, the packet error rate intervention reference coefficient PI, the retransmission rate intervention reference coefficient PJ, and the CRC error rate intervention reference coefficient PK, and the packet loss rate intervention reference coefficient PL are respectively calculated and obtained through the following formulas: Where: Ai, Bi, Ci, Di, Ei, Fi, Gi, Hi, Ii, Ji, Ki, and Li are respectively the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC error rate, and packet loss rate at the i-th timestamp; An, Bn, Cn, Dn, En, Fn, Gn, Hn, In, Jn, Kn, and Ln are respectively the bus rate, effective data rate, channel width, peak throughput, link delay, packet response time, queue depth, response waiting time, packet error rate, retransmission rate, CRC error rate, and packet loss rate at the n-th timestamp; m is the size of the window, indicating that there are m data points in each window. Continuously take m data points to calculate the mean value, and ln is the logarithmic function.

8. The PCIE data transmission analysis system according to claim 6 or 7, characterized in that: The data anomaly coefficient SJY is calculated and obtained through the following formula: SJY = b1×ln(CSX) + b2×ln(YCX) + b3×ln(XYX); Where: b1, b2, and b3 are weight values, and the values of b1, b2, and b3 are adjusted and set by the user. CSX is the transmission rate coefficient, YCX is the delay analysis coefficient, and XYX is the CRC coefficient.

9. The PCIE data transmission analysis system according to claim 8, characterized in that: The analysis module (4) includes a first analysis unit (41) and a second analysis unit (42); the first analysis unit (41) is used to generate a first comparison result, and the second analysis unit (42) is used to generate a second comparison result; The first comparison result is as follows: When SJY < Y, it represents that the current data transmission is normal; When SJY ≥ Y, it represents that the current data transmission has an anomaly; The second comparison result is as follows: When SYL < R, it represents that the current data transmission is a first-level anomaly; When R ≤ SYL < R×115%, it represents that the current data transmission is a second-level anomaly; When SYL ≥ R×115%, it represents that the current data transmission is a third-level anomaly; The data anomaly level coefficient SYL is calculated and obtained through the following formula: Where: SJY is the data anomaly coefficient SJY, and Y is the first threshold.

10. The PCIE data transmission analysis system according to claim 9, characterized in that: The alarm information generated by the feedback module (5) is specifically as follows: When the current data transmission is a first-level anomaly, a yellow alarm message is sent to the background maintenance personnel. The alarm message starts with DY, and the color is set to yellow. Before the alarm is解除, it is repeated at intervals of twenty minutes / time, and the repair is restricted within one hour. If the repair cannot be completed, the data transmission is forced to be interrupted; When the current data transmission is a level 2 abnormality, an orange alarm message will be sent to the backend maintenance personnel. The alarm message starts with DR and is orange in color. It will be sent repeatedly at intervals of five minutes before the alarm is lifted. Repairs must be made within thirty minutes. If the repairs cannot be completed, data transmission will be forcibly interrupted. When the current data transmission is at level three abnormality, a red alarm message will be sent to the background maintenance personnel. The alarm message starts with DS and the color is set to red. Before the alarm is lifted, it will be sent repeatedly at an interval of five seconds per time. Repairs are limited to ten minutes. If the repair cannot be completed, data transmission will be forcibly interrupted.

Images