Delay monitoring and synchronous adjusting method and system in fault injection scene

Abstract

The invention relates to the technical field of computer networks, and discloses a delay monitoring and synchronous adjustment method and system in a fault injection scene, and the system comprises a data processing module, a delay monitoring module and a synchronous adjustment module. A monitoring mark containing a sending timestamp is embedded in a data packet, delay time, average delay and a delay fluctuation value are accurately calculated, and a solid data support is provided for network condition evaluation; whether the network node is abnormal or not can be quickly judged by utilizing preset threshold comparison; when an affected network path is determined, an abnormal link is accurately positioned by means of timestamp comparison of a data packet at a node; when the adjustment strategy is evaluated, the bandwidth and load conditions are scientifically judged based on comparison of the bandwidth utilization rate and the CPU utilization rate with a threshold value; during synchronous adjustment, access lines are increased if the bandwidth is insufficient, and tasks are migrated if the load is too high; the whole system is clear in structure, all modules cooperate in a division mode, and network stability, resource utilization rate and user experience are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer networks, and particularly to a method and system for delay monitoring and synchronization adjustment in a fault injection scenario. Background Art

[0002] In an actual network environment, fault injection is an important means for testing the reliability and stability of a network system. By deliberately injecting various types of faults into the network, such as link faults, node faults, delay faults, etc., the performance of the network under abnormal conditions can be simulated, thus helping network administrators and technicians better understand the vulnerabilities of the network system and formulate corresponding optimization strategies.

[0003] However, in a fault injection scenario, the network delay problem is particularly prominent and complex. Network delay not only slows down data transmission speed and affects user experience, but may also lead to a series of serious consequences such as data loss and system errors. For example, in a financial trading system with extremely high real-time requirements, even a short delay may cause huge economic losses; in an industrial automation control system, network delay may affect the precise control of equipment and trigger production accidents.

[0004] Currently, traditional delay monitoring methods often have difficulty in comprehensively and accurately obtaining the delay conditions of network nodes. Some simple monitoring methods can only provide single delay data and cannot reflect the fluctuation characteristics of the delay, while delay fluctuations also have a significant impact on network performance. At the same time, in the face of network delay problems, existing synchronization adjustment methods lack systematicness and accuracy. Some methods are inefficient and inaccurate when determining the affected network paths, resulting in the inability to optimize the network in a timely and effective manner. In terms of adjustment strategies, key factors such as the bandwidth of the network path and node load are often not fully considered, making the adjustment effect unsatisfactory and unable to fundamentally solve the network delay problem, and it is difficult to meet the requirements for efficient and stable network performance in a modern complex network environment. Therefore, it is of great practical significance to study a more efficient and accurate method for delay monitoring and synchronization adjustment in a fault injection scenario. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for delay monitoring and synchronization adjustment in a fault injection scenario, and solve the technical problems proposed in the background art.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A method for delay monitoring and synchronization adjustment in a fault injection scenario includes the following steps: First step, delay monitoring: In a computer network system, signal marking is performed on each data to be sent, and then the data packets received on each network node are detected and their delay analysis is carried out. Subsequently, the delay time of the corresponding network node is determined. Step 2, Synchronization adjustment: Based on the delay monitoring results, delay evaluation is carried out. Subsequently, it is determined whether the relevant network nodes need synchronization adjustment. When synchronization adjustment is required for the relevant network nodes, first, the network path of the data packet is determined, and then the affected network paths are determined. Based on the bandwidth and node load of the affected network paths, adjustment strategy evaluation and execution are carried out.

[0007] As a further solution of the present invention, the delay monitoring method is as follows: Step A1, Signal marking: In a computer network system, a predetermined delay monitoring mark is embedded in the data packets to be sent by the relevant network nodes. Among them, the delay monitoring mark includes the sending timestamp of the data packet. Step A2, Reception record: When the data packet M i with the delay monitoring mark arrives at the receiving node, the receiving node extracts the sending timestamp TF i from it. The receiving node records the arrival timestamp of the relevant data packet and marks it as TD i . Step A3, Delay analysis: Based on the difference between the sending timestamp TF i in the relevant data packet M i and the arrival timestamp TD i , the delay time of the relevant data packet M i during network transmission is determined and marked as D i . Subsequently, the delay times Di of n data packets are extracted respectively, and then the average value of the corresponding delay times Di of the n data packets is calculated and marked as the average delay DP. At the same time, based on the delay times Di of the n data packets and their average value, the standard deviation of the corresponding delay times Di of the n data packets is calculated and marked as the delay fluctuation value DB.

[0008] As a further solution of the present invention, the delay fluctuation value is used to measure the stability of the delay.

[0009] As a further solution of the present invention, the synchronization adjustment method is as follows: Step B1, Delay evaluation: Extract the preset delay threshold DPy and fluctuation threshold DBy, and then compare them with the average delay DP and delay fluctuation value DB corresponding to the relevant network node respectively: If at least one of DP > DPy and DB > DBy holds, it is determined that the relevant network node has an abnormal delay situation, and then synchronous adjustment is performed; If both DP ≤ DPy and DB ≤ DBy hold, it is determined that the relevant network node has no abnormal delay situation, and no synchronous adjustment is performed; Step B2, Network path determination: After it is determined that synchronous adjustment is required, the affected network path is determined; Step B3, Evaluate adjustment strategies: After the affected network path is determined, different adjustment strategies are evaluated according to the bandwidth and node load of the network path; As a further solution of the present invention, the CPU utilization threshold LCy is the upper limit of the CPU load preset according to the design requirements of the network node.

[0010] As a further solution of the present invention, the signal marking method is: Extract each data packet to be sent and mark it as Pi, i = 1, 2,... n, where n represents the number of data packets to be sent; Embed the transmission timestamp into the data packet before sending the data packet, and then obtain the data packet M with a delay monitoring mark added i ; Among them, Mi = {P i , TF i}, M i represents the data packet with a delay monitoring mark added, P i is the data packet before sending, and TF i is the transmission timestamp of the corresponding data packet.

[0011] As a further solution of the present invention, the calculation formula for the delay time corresponding to the relevant data packet is: D i = TD i − TF i ; The calculation formula for the average delay is: ; The calculation formula for the delay fluctuation value is: .

[0012] As a further solution of the present invention, the network path determination method is: When a data packet is transmitted in a network, several network nodes passed by it are extracted, and the arrival timestamp and transmission timestamp of the relevant data packet recorded at each network node are extracted, that is, the time when the data packet arrives at the relevant node and the time when the data packet leaves the relevant node; Subsequently, the transmission time between two adjacent network nodes during transmission is extracted, and then the transmission time is compared with a preset transmission time threshold: Among them, the transmission time threshold is set by taking 95% of the average value corresponding to the transmission time between two adjacent network nodes in historical transmission as the relevant threshold; When the transmission time exceeds the transmission time threshold, it is determined that there is an abnormality in the link between the two adjacent network nodes during transmission, that is, it is an affected network path.

[0013] As a further solution of the present invention, the evaluation and adjustment strategy is as follows: Step B3.1, Bandwidth evaluation and adjustment: On the affected network path, its total bandwidth is obtained and marked as B0; At the same time, the used bandwidth at multiple time nodes within a specified period is obtained and marked as B j , j = 1, 2,... m, where m represents the number of time nodes within the specified period; Subsequently, through: , the bandwidth utilization rate BL within the specified period is calculated; The bandwidth utilization rate BL within the specified period is compared with a preset bandwidth utilization rate threshold BLy: When BL > BLy, it is determined that the bandwidth on this network path is insufficient, and then the bandwidth of this network path is increased; Step B3.2, Load evaluation and adjustment: At the relevant network node, the duration of its corresponding CPU in the running state within a specified period is obtained and marked as CT; Subsequently, through: , the CPU utilization rate LC of the relevant network node within the specified period is calculated; in the formula, T0 is the duration of the specified period; The CPU utilization rate LC is compared with a preset CPU utilization rate threshold LCy: When LC > LCy, it is determined that the load of this network node is too high, and then load balancing adjustment is performed.

[0014] As a further solution of the present invention, if the adjustment strategy is to increase the bandwidth of this network path, then on the basis of the current network access line, one or more new access lines are added; If the adjustment strategy is load balancing adjustment, calculate the CPU utilization rate LC of other network nodes, and then select the network nodes corresponding to LC < (LCy / 2) as the target nodes for task migration of the network nodes with excessive load, and then migrate the relevant tasks of the network nodes with excessive load to the target nodes.

[0015] A delay monitoring and synchronization adjustment system in a fault injection scenario, each network node in the system includes: A data processing module, which is used to mark the signal of each data to be sent in the computer network system; it is also used to perform delay evaluation according to the result of the delay monitoring module, and then determine whether the relevant network nodes need synchronization adjustment; A delay monitoring module, which is used to detect the data packets received on each network node, perform delay analysis on them, and then determine the delay time of the corresponding network node; A synchronization adjustment module, which is used to perform synchronization adjustment on the relevant network nodes according to the determination result of the data processing module. The synchronization adjustment method is to first determine the network path of the data packet, then determine the affected network path, and perform adjustment strategy evaluation and execution according to the bandwidth and node load of the affected network path.

[0016] As a further solution of the present invention, the data processing module, the delay monitoring module and the synchronization adjustment module are communicatively connected.

[0017] The beneficial effects of the present invention: In the present invention, by embedding a delay monitoring mark containing a sending timestamp in the data packet to be sent, and the receiving node records the arrival timestamp, the delay time of each data packet is accurately calculated, and then the average delay and the delay fluctuation value are calculated, which can comprehensively and accurately monitor the delay situation in network transmission, and provide a reliable data basis for subsequent evaluation and adjustment.

[0018] In the present invention, a delay threshold and a fluctuation threshold are set, and the calculated average delay and delay fluctuation value are compared with them, which can quickly determine whether there is an abnormal delay situation in the network node, so as to take synchronization adjustment measures in time, ensure the normal operation of the network, and improve the stability and reliability of the network.

[0019] In the present invention, when determining the affected network path, by extracting the arrival and sending timestamps of the network nodes passed by the data packet and comparing the transmission time between adjacent nodes with a preset threshold, the link with anomalies, that is, the affected network path, can be accurately found, providing a clear direction for targeted network adjustment.

[0020] In the present invention, when evaluating and adjusting the strategy, for bandwidth evaluation, by calculating the bandwidth utilization rate within a specified period and comparing it with a preset threshold, it is possible to accurately determine whether the bandwidth on the network path is insufficient, thereby deciding whether to increase the bandwidth; for load evaluation, by calculating the CPU utilization rate of network nodes within a specified period and comparing it with a preset threshold, it can accurately determine whether the node load is too high, and then perform load balancing adjustment. This data-based evaluation method makes the adjustment strategy more scientific and effective, and improves the utilization efficiency of network resources.

[0021] In the present invention, during synchronous adjustment, different strategies are adopted according to the evaluation results. If the bandwidth is insufficient, the access line is increased to improve the overall bandwidth; if the load is too high, the task is migrated to a node with a low CPU utilization rate to achieve load balancing. This flexible adjustment method can adapt to different network conditions, optimize network performance, and improve the user experience.

[0022] In the present invention, in the delay monitoring and synchronous adjustment system in the fault injection scenario, the data processing module, the delay monitoring module, and the synchronous adjustment module have clear divisions of labor and cooperate with each other, ensuring the efficient operation of the entire system, enabling the delay monitoring and synchronous adjustment work to proceed in an orderly manner, and improving the maintainability and scalability of the system. Brief Description of the Drawings

[0023] The present invention will be further described below in conjunction with the accompanying drawings.

[0024] Figure 1 It is a system block diagram of the delay monitoring and synchronous adjustment system in the fault injection scenario of the present invention.

[0025] Figure 2 It is a flowchart of the delay monitoring method in the fault injection scenario of the present invention.

[0026] Figure 3 It is a flowchart of the synchronous adjustment method in the fault injection scenario of the present invention. Detailed Embodiments

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] Embodiment 1 Please refer to Figure 1 、 Figure 2 and Figure 3 As shown, the present invention provides a delay monitoring method in the fault injection scenario, including the following steps: Step A1, Signal Marking: In a computer network system, embed a predetermined delay monitoring mark into the data packets to be sent by relevant network nodes; Among them, the delay monitoring mark includes the transmission timestamp of the data packet; The method is as follows: Extract each data packet to be sent and mark it as P i , i = 1, 2,... n, where n represents the number of data packets to be sent; Embed the transmission timestamp into the data packet before sending the data packet, and then obtain the data packet M with the delay monitoring mark added i ; Among them, Mi = {P i , TF i}, M i represents the data packet with the delay monitoring mark added, P i is the data packet before sending, and TF i is the transmission timestamp of the corresponding data packet; Step A2, Receiving and Recording: When the data packet M with the delay monitoring mark i arrives at the receiving node, the receiving node extracts the transmission timestamp TF i in it, the receiving node records the arrival timestamp of the relevant data packet and marks it as TD i ; Step A3, Delay Analysis: Based on the difference between the transmission timestamp TF i and the arrival timestamp TD i in the relevant data packet M i , determine the delay time of the relevant data packet M i during network transmission and mark it as D i ; The calculation formula is: D i = TD i − TF i ; Then extract the respective delay times Di of the n data packets, and then calculate the average value of the corresponding delay times Di of the n data packets and record this average value as the average delay DP; The calculation formula is: ; At the same time, calculate the standard deviation of the corresponding delay times Di of the n data packets based on the delay times Di of the n data packets and their average value and mark it as the delay fluctuation value DB; The calculation formula is: .

[0029] The delay monitoring method in the fault injection scenario provided by Embodiment 1 embeds a transmission timestamp before sending a data packet as a delay monitoring mark. The receiving node records the arrival timestamp, and accurately calculates the delay time of a single data packet. On this basis, the average value and standard deviation of the delay times of multiple data packets are further calculated, which can comprehensively and meticulously reflect the average level and fluctuation status of network transmission delay, providing strong data support for in-depth understanding of network delay characteristics, helping network operation and maintenance personnel clearly master the network delay status, and laying a solid foundation for subsequent analysis and optimization of network performance.

[0030] Embodiment 2 Please refer to Figure 1 、 Figure 2 and Figure 3 As shown in, as Embodiment 2 of the present invention, in the specific implementation of this application, compared with Embodiment 1, the technical solution of this embodiment is only different from that of Embodiment 1 in that in this embodiment, the present invention also provides a synchronization adjustment method in the fault injection scenario, including the following steps: Step B1, Delay evaluation: Extract the pre-set delay threshold DPy and fluctuation threshold DBy, and then compare them with the average delay DP and delay fluctuation value DB corresponding to the relevant network nodes respectively: If at least one of DP > DPy and DB > DBy holds, it is determined that the relevant network node has an abnormal delay situation, and then synchronization adjustment is performed; If both DP ≤ DPy and DB ≤ DBy hold, it is determined that the relevant network node does not have an abnormal delay situation, and no synchronization adjustment is performed; Step B2, Network path determination: When it is determined that synchronization adjustment is required, then determine the affected network path; The method is as follows: When a data packet is transmitted in the network, extract several network nodes it passes through, and extract the arrival timestamp and transmission timestamp of the relevant data packet recorded at each network node, that is, the time when the data packet arrives at the relevant node and the time when the data packet leaves the relevant node; Then extract the transmission time between two adjacent network nodes from them, and then compare the transmission time with the pre-set transmission time threshold: When the transmission time exceeds the transmission time threshold, it is determined that there is an abnormality in the link between the two adjacent network nodes, that is, it is the affected network path; In this embodiment, the transmission time threshold is set by taking 95% of the average value corresponding to the transmission times between multiple adjacent network nodes in historical data as the relevant threshold; Illustrate with an example: Suppose the two network nodes for adjacent transmission are network node A and node B respectively; then collect the transmission times of 100 data packets between network node A and node B within 30 days, then calculate the average value of the transmission times corresponding to these 100 data packets, and use the result obtained by multiplying this average value by 95% as the transmission time threshold; Step B3, evaluate adjustment strategies: After determining the affected network path, evaluate different adjustment strategies according to the bandwidth and node load of this network path; Step B3.1, bandwidth evaluation and adjustment: On the affected network path, obtain its total bandwidth and mark it as B0; At the same time, obtain the used bandwidth at multiple time nodes within a specified period and mark it as B j , j = 1, 2, …… m, where m represents the number of time nodes within the specified period; Then through: , calculate the bandwidth utilization rate BL within the specified period; Compare the bandwidth utilization rate BL within the specified period with the preset bandwidth utilization rate threshold BLy: When BL > BLy, it is determined that the bandwidth on this network path is insufficient, and then increase the bandwidth of this network path; Step B3.2, load evaluation and adjustment: On the relevant network node, obtain the duration of its corresponding CPU in the running state within a specified period and mark it as CT; Then through: , calculate the CPU utilization rate LC of the relevant network node within the specified period; in the formula, T0 is the duration of the specified period; Compare the CPU utilization rate LC with the preset CPU utilization rate threshold LCy: When LC > LCy, it is determined that the load of this network node is too high, and then perform load balancing adjustment; Step B4, synchronous execution: If the adjustment strategy is to increase the bandwidth of this network path, then on the basis of the current network access line, add one or more new access lines to improve the overall bandwidth; If the adjustment strategy is load balancing adjustment, then calculate the CPU utilization rate LC of other network nodes, then select the network node corresponding to LC < (LCy / 2) as the target node for task migration of the network node with too high load, and then migrate the relevant tasks of the network node with too high load to the target node.

[0031] Based on Embodiment 1, Embodiment 2 adds a synchronization adjustment method in the fault injection scenario. First, by setting a delay threshold and a fluctuation threshold and comparing them with the average delay and delay fluctuation value, it is possible to accurately determine whether there is abnormal delay in the network node and discover problems in a timely manner. Second, using the arrival and sending timestamps of data packets at the network node and combining with the transmission time threshold set based on historical data, it is possible to accurately locate the affected network path. Finally, according to the bandwidth and node load of the affected network path, bandwidth evaluation adjustment and load evaluation adjustment strategies are respectively adopted, and corresponding operations are executed to effectively solve the network delay problem and significantly improve the stability and performance of network operation.

[0032] Embodiment 3 Please refer to Figure 1 、 Figure 2 and Figure 3 As shown in, as Embodiment 3 of the present invention, in the specific implementation of the present application, compared with Embodiment 1 and Embodiment 2, the technical solution of this embodiment is to combine the solutions of the above-mentioned Embodiment 1 and Embodiment 2 for implementation.

[0033] Embodiment 3 organically combines the delay monitoring method of Embodiment 1 with the synchronization adjustment method of Embodiment 2 to form a complete delay monitoring and synchronization adjustment system in the fault injection scenario.

[0034] The present invention also provides a delay monitoring and synchronization adjustment system in the fault injection scenario. Each network node in this system includes: A data processing module for signal marking of each data to be sent in the computer network system; A delay monitoring module for detecting data packets received on each network node, performing delay analysis on them, and then determining the delay time of the corresponding network node; The data processing module is also used to perform delay evaluation according to the results of the delay monitoring module, and then determine whether the relevant network node needs synchronization adjustment; A synchronization adjustment module for synchronously adjusting the relevant network node according to the determination result of the data processing module. The synchronization adjustment method is to first determine the network path of the data packet, then determine the affected network path, and perform adjustment strategy evaluation and execution according to the bandwidth and node load of the affected network path.

[0035] The data processing module in the system is responsible for signal marking and delay evaluation. The delay monitoring module focuses on data packet delay analysis. The synchronization adjustment module executes the adjustment strategy according to the evaluation result. Each module has a clear division of labor and collaborates with each other, realizing the full-process automated management from delay monitoring to synchronization adjustment, greatly improving the efficiency and accuracy of the system in dealing with network delay problems, enhancing the reliability and stability of the network, and meeting the requirements for network performance optimization in complex network environments.

[0036] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0037] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claimed rights.

Claims

1. A method for delay monitoring and synchronization adjustment in a fault injection scenario, characterized in that It includes the following steps: The first step, delay monitoring: In a computer network system, signal marking is performed on each data packet to be sent, then the data packets received on each network node are detected, and delay analysis is performed on them, and then the delay time of the corresponding network node is determined; The second step, synchronization adjustment: Based on the delay monitoring results, delay evaluation is performed, and then it is determined whether the relevant network nodes need synchronization adjustment. When synchronization adjustment is required for the relevant network nodes, first determine the network path of the data packet, then determine the affected network path, and evaluate and execute the adjustment strategy according to the bandwidth and node load of the affected network path.

2. The delay monitoring and synchronization adjustment method in the fault injection scenario according to claim 1, characterized in that The delay monitoring method is as follows: Step A1, signal marking: In a computer network system, a predetermined delay monitoring mark is embedded in the data packets to be sent by the relevant network nodes; Among them, the delay monitoring mark contains the sending timestamp of the data packet; Step A2, reception record: When the data packet with the delay monitoring mark arrives at the receiving node, the receiving node extracts the sending timestamp therein, and the receiving node records the arrival timestamp of the relevant data packet; Step A3, delay analysis: Based on the difference between the sending timestamp and the arrival timestamp in the relevant data packet, determine the delay time of the relevant data packet during network transmission; Then extract the delay time of each of multiple data packets, and then calculate the average value of the corresponding delay times of the multiple data packets, and record this average value as the average delay; At the same time, calculate the standard deviation of the corresponding delay times of the multiple data packets based on the delay times and their average values of the multiple data packets, and mark it as the delay fluctuation value.

3. The delay monitoring and synchronization adjustment method in the fault injection scenario according to claim 2, characterized in that, The synchronization adjustment method is as follows: Step B1, delay evaluation: Extract the preset delay threshold and fluctuation threshold, and then compare them with the average delay and delay fluctuation value corresponding to the relevant network node respectively, and based on the comparison results, determine whether the relevant network node needs synchronization adjustment; Step B2, network path determination: After it is determined that synchronization adjustment is required, then determine the affected network path; Step B3, evaluation of adjustment strategy: After determining the affected network path, evaluate different adjustment strategies according to the bandwidth and node load of this network path, and perform synchronous execution.

4. The method for delay monitoring and synchronization adjustment in a fault injection scenario according to claim 3, characterized in that In step B1: If at least one of the average delay > delay threshold and delay fluctuation value > fluctuation threshold holds, it is determined that the relevant network node has an abnormal delay situation, and then synchronization adjustment is performed; If both the average delay ≤ delay threshold and delay fluctuation value ≤ fluctuation threshold hold, it is determined that the relevant network node does not have an abnormal delay situation, and no synchronization adjustment is performed.

5. The delay monitoring and synchronization adjustment method in the fault injection scenario according to claim 2, wherein The signal marking method is: Extract each data packet to be sent and mark it as P i , where i = 1, 2, …… n, and n represents the number of data packets to be sent; Embed the sending timestamp into the data packet before sending the data packet, and then obtain the data packet Mi={P i , TF i} with the delay monitoring mark added; Among them, M i represents the data packet with a delay monitoring mark added, P i is the data packet before sending, TF i is the sending timestamp of the corresponding data packet.

6. The delay monitoring and synchronization adjustment method in the fault injection scenario according to claim 3, characterized in that The network path determination method is: When the data packet is transmitted in the network, extract several network nodes it passes through, and extract the arrival timestamp and sending timestamp of the relevant data packet recorded at each network node, that is, the time when the data packet arrives at the relevant node and the time when the data packet leaves the relevant node; Then extract the transmission time between two adjacent network nodes during transmission, and then compare this transmission time with the preset transmission time threshold: When the transmission time exceeds the transmission time threshold, it is determined that there is an abnormality in the link between the two neighboring network nodes for transmission, that is, it is an affected network path.

7. The method for delay monitoring and synchronization adjustment in a fault injection scenario according to claim 3, wherein The method for evaluating and adjusting the strategy is as follows: Step B3.1, Bandwidth evaluation and adjustment: On the affected network path, obtain its total bandwidth and label it as B0; Obtain the used bandwidth at multiple time nodes within a specified period simultaneously and mark it as B j , where j = 1, 2,..., m, and m represents the number of time nodes within the specified period; Followed by: , calculate the bandwidth utilization rate BL within the specified period; Compare the bandwidth utilization rate BL within a specified period with the preset bandwidth utilization rate threshold BLy: When BL > BLy, it is determined that the bandwidth on this network path is insufficient, and then increase the bandwidth of this network path; Step B3.2, Load evaluation and adjustment: At the relevant network node, obtain the duration of the corresponding CPU in the running state within a specified period and label it as CT; Followed by: , calculate the CPU utilization rate LC of relevant network nodes within a specified period; where T0 is the duration of the specified period; Compare the CPU utilization rate LC with the preset CPU utilization rate threshold LCy: When LC > LCy, it is determined that the load of this network node is too high, and then perform load balancing adjustment.

8. The delay monitoring and synchronization adjustment method in the fault injection scenario according to claim 7, characterized in that If the corresponding adjustment strategy to be executed synchronously is to increase the bandwidth of this network path, then on the basis of the current network access line, add a new access line.

9. The delay monitoring and synchronization adjustment method in the fault injection scenario according to claim 7, wherein If the corresponding adjustment strategy to be executed synchronously is load balancing adjustment, calculate the CPU utilization rate LC of other network nodes, then select the network node corresponding to LC < (LCy / 2) as the target node for task migration of the network node with too high load, and then migrate the relevant tasks of the network node with too high load to the target node.

10. A delay monitoring and synchronization adjustment system in a fault injection scenario, which is implemented by the delay monitoring and synchronization adjustment method in a fault injection scenario according to any one of claims 1-9, characterized in that Each network node in this system includes a data processing module, a delay monitoring module, and a synchronization adjustment module, and they are communicatively connected to each other: The data processing module is used to mark signals for each data to be sent in the computer network system; it is also used to perform delay evaluation based on the results of the delay monitoring module, and then determine whether the relevant network nodes need synchronization adjustment; The delay monitoring module is used to detect the data packets received on each network node, perform delay analysis on them, and then determine the delay time of the corresponding network node; The synchronization adjustment module is used to perform synchronization adjustment on the relevant network nodes according to the determination results of the data processing module. The synchronization adjustment method is to first determine the network path of the data packet, then determine the affected network path, and evaluate and execute the adjustment strategy according to the bandwidth and node load of the affected network path.

Images