Network delay reduction method and device, terminal and medium

Abstract

The invention discloses a network delay reduction method and device, a terminal and a medium, and relates to the technical field of data processing, and the method comprises the steps: carrying out the preprocessing of department business data corresponding to each department in a company and gateway state data corresponding to each department, and obtaining a structured data set; dividing each department into at least two different department groups according to the structured data set through a preset neural network model; based on the structured data set and the department group, calculating to obtain a bandwidth allocation dynamic threshold value corresponding to each department and a delay tolerance dynamic threshold value corresponding to each department; and obtaining a target network delay reduction scheme based on the bandwidth allocation dynamic threshold, the delay tolerance dynamic threshold and the structured data set. The invention aims to realize company-level network delay minimization and ensure that the network service quality of each department reaches the required level at the same time.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, terminal and medium for reducing network latency. Background Technology

[0002] Modern enterprise networks face complex business scenarios, including remote work, cloud computing applications, and real-time data transmission (such as video conferencing, ERP system interaction, and IoT device communication). These scenarios place extremely high demands on network latency and stability.

[0003] However, current technologies for reducing latency in corporate networks have the following shortcomings: the manual network configuration strategies employed are static, extensive, and manual, making it difficult to cope with the dynamic, complex, and differentiated needs of modern enterprise networks; at the same time, the lack of detailed consideration of differences in latency tolerance and bandwidth requirements leads to resource misallocation, such as non-critical services consuming too many resources while critical services experience latency due to insufficient resources. If the optimization direction is based on a single objective (such as minimizing average latency), some departments may sacrifice their own network service quality for overall optimization, thereby failing to guarantee the overall network service quality. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, terminal, and medium for reducing network latency, which aims to achieve precise and efficient management of network services across multiple departments within a company, thereby minimizing network latency at the company level while ensuring that the network service quality of each department reaches the required level.

[0005] To achieve the above objectives, this application provides a method for reducing network latency, the method comprising: Preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset; By using a pre-set neural network model, based on the structured dataset, each department is divided into at least two different department groups; Based on the structured dataset and the department group, the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department are calculated. Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, a target network latency reduction scheme is obtained.

[0006] Specifically, the preset neural network model includes a first model and a second model, and the structured dataset includes department feature vectors and gateway status feature vectors. The first model is used to predict the network demand status of each department, and the second model is used to classify each department based on the network demand status of each department. The method involves using a pre-defined neural network model to divide each department into at least two different department groups based on the structured dataset, including: Using the first model, based on the department feature vector and the gateway status feature vector, the department network demand feature vector corresponding to each department is obtained; The second model is used to classify each department based on the department network demand feature vector to obtain the department group.

[0007] Specifically, the first model includes a first input layer and a hidden layer; The step of obtaining the departmental network demand feature vector for each department through the first model, based on the departmental feature vector and the gateway state feature vector, includes: The first input vector is obtained through the first input layer based on the department feature vector and the gateway status feature vector; The hidden layer is used to obtain the departmental network demand feature vector based on the first input vector.

[0008] Specifically, the second model includes a second input layer and a competition layer; The second model classifies each department based on the departmental network demand feature vector to obtain the departmental group, including: The second input layer obtains the second input vector based on the departmental network demand feature vector. Through the competition layer, the department group is obtained based on the second input vector, wherein the department group is used to characterize the department categories divided based on different latency tolerances and different bandwidth requirements.

[0009] Specifically, the calculation of the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department, based on the structured dataset and the department group, includes: Set the initial bandwidth threshold, the initial latency tolerance threshold, and the priority weight for each department group; Based on the real-time traffic and historical average traffic of each department in the structured dataset, the traffic fluctuation factor for each department is calculated. Based on the number of departmental devices in each department and the average number of departmental devices in each department group in the structured dataset, the device density factor for each department is calculated. The dynamic threshold for bandwidth allocation is calculated based on the initial bandwidth threshold, the priority weight, the traffic fluctuation factor, the device density factor, and the current gateway load rate corresponding to each department in the structured dataset. Based on the gateway signal strength of each department in the structured dataset and the baseline gateway signal strength of each department in the structured dataset, the signal quality factor of each department is calculated. The dynamic latency tolerance threshold is calculated based on the initial latency tolerance threshold, the signal quality factor, and the historical average latency of each department in the structured dataset.

[0010] Specifically, the step of obtaining the target network latency reduction scheme based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset includes: Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, an objective function and corresponding constraints are constructed with the goal of minimizing overall network latency and maximizing the departmental transmission quality compliance rate. Based on the objective function and the constraints, the objective decision variables are obtained, and based on the objective decision variables, the objective network latency reduction scheme is obtained.

[0011] Specifically, the target decision variables include a first target decision variable and a second target decision variable, and the objective function and the constraints are represented by the following expressions;

[0012]

[0013]

[0014]

[0015]

[0016]

[0017]

[0018] in, This represents the objective of the objective function. This indicates the preset weighting coefficient. This represents the average network latency data within the company's departments. This indicates the rate of substandard network transmission quality within the company. Indicates the number of departments within the company. This indicates the number of intranet gateways within the company. Indicates department via gateway network latency, This represents the first objective decision variable. Indicates department By gateway Serve, Indicates department Not by gateway Serve, Indicates an indicator function with attached conditions. The condition for the indicated function is a department. via gateway Network latency greater than department The corresponding delay tolerance dynamic threshold, when the condition of the indicator function is met, When the condition of the indicator function is not met, , Indicates department The corresponding delay tolerance dynamic threshold, This represents the second objective decision variable, which is used to characterize the gateway. Assigned to department bandwidth data, Indicates gateway Total bandwidth capacity Indicates department The corresponding dynamic threshold for bandwidth allocation, Indicates department via gateway Actual network traffic This means that any department within a department must be served by at least one gateway.

[0019] To achieve the above objectives, this application also provides a network latency reduction device, the device comprising: The first unit is used to preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset. The second unit is used to divide each department into at least two different department groups based on the structured dataset using a preset neural network model. The third unit is used to calculate the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department based on the structured dataset and the department group. The fourth unit is used to obtain a target network latency reduction scheme based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset.

[0020] To achieve the above objectives, this application also provides a terminal, including a memory storing multiple instructions; the processor loads instructions from the memory to execute the steps in any of the methods provided in this application.

[0021] To achieve the above objectives, this application also provides a medium storing a plurality of instructions adapted for loading by a processor to execute the steps in any of the methods provided in this application.

[0022] This application provides a network latency reduction method, apparatus, terminal, and medium. First, it preprocesses departmental business data and gateway status data corresponding to each department within a company to obtain a structured dataset. Then, using a pre-defined neural network model, it divides each department into at least two different department groups based on the structured dataset. Next, based on the structured dataset and the department groups, it calculates dynamic thresholds for bandwidth allocation and latency tolerance for each department. Finally, based on the dynamic thresholds for bandwidth allocation, latency tolerance, and the structured dataset, it obtains a target network latency reduction scheme that can accurately and efficiently manage network services across multiple departments within the company. This minimizes network latency at the company level while ensuring that the network service quality of each department reaches the required level. Attached Figure Description

[0023] Figure 1 This is a schematic diagram illustrating an application scenario of the FTTR-B network latency reduction system provided in this application embodiment; Figure 2 A flowchart illustrating the method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the application process of the first model provided in the embodiments of this application; Figure 4 This is a schematic diagram of the application process of the second model provided in the embodiments of this application; Figure 5 This is a schematic diagram of the network latency reduction device provided in the embodiments of this application; Figure 6 This is a schematic diagram of the terminal structure provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Current corporate network latency reduction technologies have the following shortcomings: the manual network configuration strategies employed are static, extensive, and manual, making it difficult to cope with the dynamic, complex, and differentiated needs of modern enterprise networks; at the same time, the lack of detailed consideration of latency tolerance and bandwidth requirements leads to resource misallocation, such as non-critical services consuming too many resources while critical services experience latency due to insufficient resources. If the optimization direction is based on a single objective (such as minimizing average latency), some departments may sacrifice their own network service quality for overall optimization, thus failing to guarantee the overall network service quality.

[0026] Therefore, embodiments of this application provide a method, apparatus, terminal, and medium for reducing network latency to solve practical technical problems.

[0027] In some embodiments, the device may be integrated into an electronic device, such as a terminal or server.

[0028] In some embodiments, the server may also be implemented as a terminal.

[0029] The server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0030] The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. The terminal and the server can be connected directly or indirectly through wired or wireless communication, which is not limited herein.

[0031] The following sections provide detailed descriptions of each example. It should be noted that the sequence numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0032] This application provides a method for reducing network latency, which can minimize network latency at the company level while ensuring that the network service quality of each department reaches the required level.

[0033] A network latency reduction method is applied to an FTTR-B network latency reduction system, which is used to manage the network status of multiple departments within a company. The FTTR-B network latency reduction system includes a master gateway and multiple slave gateways.

[0034] like Figure 1 In some embodiments, the FTTR-B network latency reduction system can be a network latency reduction system based on FTTR-B (Fiber to the Room-Business), a solution designed specifically for enterprise-level network environments. Built on Fiber to the Room (FTTR) technology, it aims to reduce network latency in various departments within a company and improve overall network performance through refined network resource management and intelligent scheduling.

[0035] Specifically, the FTTR-B network latency reduction system can be applied in the financial securities industry. Since highly sensitive trading departments require microsecond-level latency, the FTTR-B network latency reduction system ensures trading stability through dedicated channels and dynamic bandwidth. The FTTR-B network latency reduction system can also be applied in workshops with dense Industrial Internet of Things (IIoT) devices. By optimizing bandwidth allocation through device density factors, it avoids production interruptions caused by data acquisition delays. When applied in multinational corporations, the master gateway in the FTTR-B network latency reduction system coordinates the load balancing of slave gateways in global branches, reducing cross-regional data transmission latency and improving the quality of real-time services such as video conferencing and cloud collaboration.

[0036] In some embodiments, the main gateway acts as the core hub of the system, responsible for global network traffic scheduling, policy distribution, and load balancing management.

[0037] Specifically, the main gateway aggregates the status data of all slave gateways (such as load rate and available bandwidth); schedules traffic allocation to slave gateways based on global policies; monitors network latency, transmission quality and other indicators across the company; and coordinates load balancing among slave gateways to avoid single-point overload.

[0038] In some embodiments, the gateways are distributed across departments or regions, directly connecting to terminal devices (such as computers, servers, IoT devices, etc.) to perform specific network control tasks such as bandwidth allocation and channel management.

[0039] Specifically, the gateway allocates dedicated network channels to its respective departments (based on instructions from the main gateway); adjusts departmental bandwidth allocation in real time and executes dynamic threshold strategies; collects departmental network data (such as real-time traffic, number of devices, and signal strength); and feeds back the local network status to the main gateway.

[0040] In some embodiments, the FTTR-B network latency reduction system may further include a data acquisition module, which can collect business data from various departments (e.g., peak traffic, transmission protocol type); monitor gateway status data (e.g., load rate, signal strength, available bandwidth); and record network transmission quality data (e.g., latency, packet loss rate).

[0041] In some embodiments, assuming the company has four departments: R&D (D001), Finance (D002), Marketing (D003), and Administration (D004), one main gateway and four secondary gateways (G01-G04) are deployed. The FTTR-B network latency reduction system needs to ensure real-time code synchronization in the R&D department, transaction data transmission in the Finance department, video conferencing in the Marketing department, and daily office work in the Administration department, while also reducing the company's overall network latency.

[0042] like Figure 2 The specific process of the method can be as follows: S110. Preprocess the departmental business data and gateway status data corresponding to each department within the company to obtain a structured dataset.

[0043] In some embodiments, the departmental business data may include the following: R&D Department (D001): Real-time traffic 200Mbps (code transmission), 50 devices, historical average latency 45ms; Finance Department (D002): Real-time traffic 150Mbps (transaction data), number of devices 30, historical average latency 40ms; Marketing Department (D003): Real-time traffic 80Mbps (video conferencing), 20 devices, historical average latency 70ms; Administration Department (D004): Real-time traffic 30Mbps (document transfer), 10 devices, historical average latency 110ms.

[0044] Specifically, real-time traffic can be the average inbound and outbound traffic within the current 5 minutes (collected in real time from the gateway); the number of devices can be the total number of intranet terminal devices in the department (e.g., computers, servers, testing equipment, etc.); and the historical average latency can be the average round-trip latency (RTT) from the department to the main gateway over the past 24 hours.

[0045] In some embodiments, gateway status data may include the following: From gateway G01 (Service R&D Department): Load rate 60%, signal strength -55dBm, link bandwidth 1000Mbps; From gateway G02 (Service Finance Department): Load rate 55%, signal strength -58dBm, link bandwidth 1000Mbps; From gateway G03 (Service Marketing Department): Load rate 40%, signal strength -65dBm, link bandwidth 1000Mbps; From gateway G04 (Service Administration Department): Load rate 30%, signal strength -70dBm, link bandwidth 1000Mbps.

[0046] Specifically, the load rate can be the ratio between the current processing traffic of the gateway and the link bandwidth of the gateway; the signal strength can be the Wi-Fi signal strength of the core area of ​​the department, where the smaller the negative value, the stronger the signal strength, and the reference signal strength can be set to -50dBm; the link bandwidth can be the total bandwidth of the fiber optic link from the gateway to the main gateway, wherein the link bandwidth in the FTTR-B network latency reduction system can be uniformly set to 1000Mbps.

[0047] In some embodiments, the network transmission quality data corresponding to each department may also be preprocessed, wherein the network transmission quality data may include the following: R&D Department (D001): Actual latency (ms): 42; Packet loss rate (%): 0.5; Bandwidth utilization (%): 83; Finance Department (D002): Actual latency (ms): 45; Packet loss rate (%): 0.3; Bandwidth utilization (%): 75; Marketing Department (D003): Actual latency (ms): 70; Packet loss rate (%): 1.0; Bandwidth utilization (%): 60; Administration Department (D004): Actual latency (ms): 110; Packet loss rate (%): 2.0; Bandwidth utilization (%): 30.

[0048] Specifically, the actual latency can be the real-time round-trip latency from the current department's terminal to the main gateway (collected through ICMP ping test); the packet loss rate can be the proportion of network layer packet loss in the past minute (reflecting link stability); bandwidth utilization = actual bandwidth used / link bandwidth × 100%, with the R&D department having the highest utilization rate due to dynamic bandwidth allocation.

[0049] In some embodiments, outlier handling can be performed first on the departmental business data and gateway status data corresponding to each department within the company. For example, if the traffic of the R&D department suddenly increases to 800Mbps (exceeding 3 times the standard deviation of the mean) at a certain moment, it is considered an anomaly and filled with the average of 200Mbps from the previous 10 minutes. Then, the departmental business data and the gateway status data are normalized. For example, numerical data such as real-time traffic and the number of devices are scaled to [0,1] using Min-Max. Finally, the structured dataset is generated, which may include: Department feature vector (taking R&D department as an example): [Business type (R&D=3), real-time traffic (0.8), number of devices (1.0), historical average latency (0.45)]; Gateway state vector (G01 as an example): [load rate (0.6), signal strength (0.45), link bandwidth (1.0)].

[0050] S120. Using a preset neural network model, based on the structured dataset, divide each department into at least two different department groups.

[0051] In some embodiments, the preset neural network model includes a first model and a second model, and the structured dataset includes department feature vectors and gateway status feature vectors. The first model is used to predict the network demand status of each department, and the second model is used to classify each department based on the network demand status of each department.

[0052] Specifically, the step of dividing each department into at least two different department groups based on the structured dataset using a preset neural network model includes the steps S121 to S122 as shown below: S121, such as Figure 3 The first model is used to obtain the department network demand feature vector corresponding to each department based on the department feature vector and the gateway status feature vector.

[0053] In some embodiments, the first model includes a first input layer and a hidden layer; Specifically, the step of obtaining the departmental network demand feature vector corresponding to each department through the first model based on the departmental feature vector and the gateway state feature vector includes the following steps S1211 to S1212: S1211. The first input vector is obtained through the first input layer based on the department feature vector and the gateway state feature vector.

[0054] In some embodiments, the first input layer can be used to concatenate the 6-dimensional department feature vector and the 3-dimensional gateway state feature vector into a 9-dimensional first input vector.

[0055] S1212. The departmental network demand feature vector is obtained through the hidden layer based on the first input vector.

[0056] In some embodiments, core features are extracted from the first input vector through a fully connected layer in the hidden layer, and a 4-dimensional departmental network demand feature vector is output: [bandwidth demand intensity, latency sensitivity coefficient, device density index, signal dependence].

[0057] S122, such as Figure 4 The second model is used to classify each department according to the department network demand feature vector to obtain the department group.

[0058] In some embodiments, the second model includes a second input layer and a competition layer.

[0059] Specifically, the step of classifying each department according to the department network demand feature vector through the second model to obtain the department group includes the following steps S1221 to S1222: S1221. The second input vector is obtained through the second input layer based on the department network demand feature vector.

[0060] In some embodiments, a 4-dimensional departmental network demand feature vector is input into the second input layer to obtain the second input vector.

[0061] S1222. Through the competition layer, the department group is obtained according to the second input vector, wherein the department group is used to characterize the department categories divided based on different latency tolerances and different bandwidth requirements.

[0062] In some embodiments, a self-organizing map (SOM) or K-means clustering algorithm corresponding to the competition layer is used to divide each department into at least two of the following categories based on latency tolerance and bandwidth requirements: Department Group A (Highly Sensitive Department Group): R&D Department (D001), Finance Department (D002), latency tolerance ≤50ms, minimum bandwidth ≥100Mbps; Department Group B (Medium / Low Sensitivity Department Group): Marketing Department (D003) and Administration Department (D004), with latency tolerances of ≤80ms and ≤120ms respectively, and guaranteed bandwidth of 50Mbps and 20Mbps.

[0063] In some embodiments, the departments can also be divided into the following three categories based on latency tolerance and bandwidth requirements: Department Group A (Highly Sensitive Department Group): R&D Department (D001), Finance Department (D002), latency tolerance ≤50ms, minimum bandwidth ≥100Mbps; Department Group B (Medium Sensitive Department Group): Marketing Department (D003), latency tolerance ≤80ms, minimum bandwidth ≥50Mbps; Department Group C (Low Sensitive Department Group): Administration Department (D004), latency tolerance ≤120ms, minimum bandwidth ≥20Mbps.

[0064] S130. Based on the structured dataset and the department group, calculate the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department.

[0065] In some embodiments, the step of calculating the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department based on the structured dataset and the department group includes the steps S131 to S136 as shown below: S131. Set the initial bandwidth threshold, the initial latency tolerance threshold, and the priority weight for each department group.

[0066] In some embodiments, it is assumed that there are three department groups: Department Group A (Highly Sensitive Department Group): R&D Department (D001), Finance Department (D002), latency tolerance ≤50ms, minimum bandwidth ≥100Mbps; Department Group B (Medium Sensitive Department Group): Marketing Department (D003), latency tolerance ≤80ms, minimum bandwidth ≥50Mbps; Department Group C (Low Sensitive Department Group): Administration Department (D004), latency tolerance ≤120ms, minimum bandwidth ≥20Mbps.

[0067] Specifically, the initial bandwidth threshold, the initial latency tolerance threshold, and the priority weight for department group A are set to 100Mbps, 50ms, and 1.2, respectively. Set the initial bandwidth threshold, initial latency tolerance threshold, and priority weight for department group B to 50Mbps, 80ms, and 1.0, respectively. The initial bandwidth threshold, initial latency tolerance threshold, and priority weight for department group C are set to 20Mbps, 120ms, and 0.8, respectively.

[0068] S132. Based on the real-time traffic and historical average traffic of each department in the structured dataset, calculate the traffic fluctuation factor for each department.

[0069] In some embodiments, the flow fluctuation factor can be calculated using the following formula: Traffic fluctuation factor = Real-time traffic / Historical average traffic For example, for the R&D department, if the real-time traffic of the R&D department is 200Mbps and the historical average traffic of the R&D department is 180Mbps, then the traffic fluctuation factor corresponding to the R&D department is 200 / 180=1.11.

[0070] S133. Based on the number of departmental devices corresponding to each department in the structured dataset and the average number of departmental devices corresponding to each department group, calculate the device density factor corresponding to each department.

[0071] In some embodiments, the device density factor can be calculated using the following formula: Equipment density factor = Number of devices in a department / Average number of devices in the department group For example, for the R&D department, if the department has 50 pieces of equipment and the average number of pieces of equipment in department group A to which the R&D department belongs is 40, then the equipment density factor corresponding to the R&D department is 50 / 40=1.25.

[0072] S134. The bandwidth allocation dynamic threshold is calculated based on the initial bandwidth threshold, the priority weight, the traffic fluctuation factor, the device density factor, and the current load rate of the gateway corresponding to each department in the structured dataset.

[0073] In some embodiments, the dynamic threshold for bandwidth allocation can be calculated using the following formula: Dynamic bandwidth allocation threshold = Initial bandwidth threshold × Priority weight × (Traffic fluctuation factor × Device density factor) / (1 + Gateway current load rate) For example, for the R&D department, the priority weight of the department group corresponding to the R&D department is 1.2, and the current load rate of the gateway corresponding to the R&D department is 60%. Then the dynamic threshold for bandwidth allocation corresponding to the R&D department is 100×1.2×(1.11×1.25) / (1+60%)=104.04Mbps.

[0074] S135. Based on the gateway signal strength of each department in the structured dataset and the baseline gateway signal strength of each department in the structured dataset, calculate the signal quality factor for each department.

[0075] In some embodiments, the signal quality factor can be calculated using the following formula: Signal quality factor = Gateway signal strength / Reference gateway signal strength For example, for the R&D department, if the gateway signal strength of the R&D department is -55dBm and the baseline gateway signal strength is preset to -50dBm, then the signal quality factor of the R&D department is 55 / 50=1.1.

[0076] S136. Based on the initial delay tolerance threshold, the signal quality factor, and the historical average delay of each department in the structured dataset, the dynamic delay tolerance threshold is calculated.

[0077] In some embodiments, the delay tolerance dynamic threshold can be calculated using the following formula: Delay tolerance dynamic threshold = Initial delay tolerance threshold × Signal quality factor × Departmental historical average delay / Preset baseline delay coefficient For example, for the R&D department, the initial latency tolerance threshold is 50ms, the department's historical average latency can be 45ms, and the corresponding preset baseline latency coefficient can be set to 50. Then, the dynamic latency tolerance threshold of the R&D department is 50×1.1×45 / 50=49.5ms.

[0078] S140. Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, a target network latency reduction scheme is obtained.

[0079] In some embodiments, obtaining the target network latency reduction scheme based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset includes the following steps S141 to S142: S141. Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, construct an objective function and corresponding constraints for the objective function, with the goal of minimizing overall network latency and maximizing the departmental transmission quality compliance rate.

[0080] S142. Based on the objective function and the constraints, the objective decision variables are obtained, and based on the objective decision variables, the objective network latency reduction scheme is obtained.

[0081] In some embodiments, the target decision variables include a first target decision variable and a second target decision variable, and the objective function and the constraints are represented by the following expressions;

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088] in, This represents the objective of the objective function. This indicates the preset weighting coefficient. This represents the average network latency data within the company's departments. This indicates the rate of substandard network transmission quality within the company. Indicates the number of departments within the company. This indicates the number of intranet gateways within the company. Indicates department via gateway network latency, This represents the first objective decision variable. Indicates department By gateway Serve, Indicates department Not by gateway Serve, Indicates an indicator function with attached conditions. The condition for the indicated function is a department. via gateway Network latency greater than department The corresponding delay tolerance dynamic threshold, when the condition of the indicator function is met, When the condition of the indicator function is not met, , Indicates department The corresponding delay tolerance dynamic threshold, This represents the second objective decision variable, which is used to characterize the gateway. Assigned to department bandwidth data, Indicates gateway Total bandwidth capacity Indicates department The corresponding dynamic threshold for bandwidth allocation, Indicates department via gateway Actual network traffic This means that any department within a department must be served by at least one gateway.

[0089] In some embodiments, a preset weighting coefficient can be set. =0.7. If the proportion of highly sensitive business in the department is higher, a preset weighting coefficient can be assigned. A higher value.

[0090] In some embodiments, a mixed-integer programming solver (such as Gurobi) can be used, taking the objective function and the constraints as input, to obtain the objective decision variables as shown below: Research and Development Department: Service gateway: G01; Allocated bandwidth: 160Mbps; Estimated latency: 48.2ms; Finance Department: Service gateway: G02; Allocated bandwidth: 140Mbps; Predicted latency: 45.1ms; Marketing Department: Service gateway: G03; Allocated bandwidth: 50Mbps; Estimated latency: 85.7ms; Administration Department: Service gateway: G04; allocated bandwidth: 20Mbps; predicted latency: 145.2ms.

[0091] In some embodiments, based on the target decision variables described above, the following target network latency reduction scheme is obtained: The R&D and Finance departments are guaranteed through independent QoS channels, with allocated bandwidths of 160Mbps and 140Mbps respectively (exceeding the dynamic threshold, prioritizing highly sensitive services); the Marketing department's original gateway G03 had a load rate of 40%, and some traffic was reallocated to G01 (the load rate decreased from 60% to 55%) to reduce overall latency; the Administration department's latency tolerance threshold is 154ms, and the actual latency is 145.3ms, meeting the constraints, and is allocated the lowest priority bandwidth of 20Mbps.

[0092] The above-mentioned target network latency reduction scheme can achieve the following technical effects: Overall network latency decreased from an average of 65ms before optimization to 49.8ms, a reduction of 23%. Compliance rate: All 4 departments met the standards, achieving a 100% compliance rate (exceeding the target value by 90%). Resource utilization: The average gateway load rate decreased from 51.25% to 48.75%, and bandwidth allocation became more balanced.

[0093] In summary, this application provides a network latency reduction method that can implement a precise and efficient network latency reduction scheme for network services in multiple departments within a company, thereby minimizing network latency at the company level while ensuring that the network service quality of each department reaches the required level.

[0094] To better implement the above methods, this application also provides a network latency reduction device, which can be integrated into an electronic device, such as a terminal or server. The terminal can be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, or personal computer; the server can be a single server or a server cluster composed of multiple servers.

[0095] For example, in this embodiment, the method of this application embodiment will be described in detail by taking the network latency reduction device specifically integrated into the terminal as an example.

[0096] For example, such as Figure 5 As shown, the network latency reduction device 500 may include a first unit 501, a second unit 502, a third unit 503, and a fourth unit 504. The device includes: Unit 501 is used to preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset. The second unit 502 is used to divide each department into at least two different department groups based on the structured dataset using a preset neural network model. The third unit 503 is used to calculate the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department based on the structured dataset and the department group. Unit 504 is used to obtain a target network latency reduction scheme based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset.

[0097] In practice, each of the above units can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units, please refer to the previous method embodiments, which will not be repeated here.

[0098] As can be seen from the above, the embodiments of this application can minimize network latency at the company level while ensuring that the network service quality of each department reaches the required level.

[0099] This application also provides an electronic device, which can be a terminal, a server, or other similar device. The terminal can be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers, etc.

[0100] In some embodiments, the product processing device may also be integrated into multiple electronic devices, such as multiple servers, with the network latency reduction method of this application being implemented by the multiple servers.

[0101] In this embodiment, the electronic device will be described in detail as a terminal, for example, such as... Figure 6 As shown, it illustrates a structural schematic diagram of the terminal 600 involved in an embodiment of this application. Specifically: The terminal 600 may include components such as a processor 601 with one or more processing cores, a memory 602 with one or more media, a power supply 603, an input module 604, and a communication module 605. Those skilled in the art will understand that... Figure 6 The terminal 600 structure shown does not constitute a limitation on the terminal 600, and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 601 is the control center of the terminal 600. It connects various parts of the terminal 600 via various interfaces and lines, and performs various functions and processes data by running or executing software programs and / or modules stored in the memory 602, and by calling data stored in the memory 602, thereby providing overall monitoring of the terminal 600. In some embodiments, the processor 601 may include one or more processing cores; in some embodiments, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 601.

[0102] The memory 602 can be used to store software programs and modules. The processor 601 executes various functional applications and data processing by running the software programs and modules stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the terminal 600, etc. In addition, the memory 602 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 602 may also include a memory controller to provide the processor 601 with access to the memory 602.

[0103] The terminal 600 also includes a power supply 603 that supplies power to the various components. In some embodiments, the power supply 603 can be logically connected to the processor 601 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 603 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0104] The terminal 600 may also include an input module 604, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0105] The terminal 600 may also include a communication module 605. In some embodiments, the communication module 605 may include a wireless module. The terminal 600 can perform short-range wireless transmission through the wireless module of the communication module 605, thereby providing users with wireless broadband Internet access. For example, the communication module 605 can be used to help users send and receive emails, browse web pages, and access streaming media.

[0106] Although not shown, terminal 600 may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, processor 601 in terminal 600 loads the executable files corresponding to the processes of one or more applications into memory 602 according to the following instructions, and processor 601 runs the applications stored in memory 602 to realize various functions, as follows: Preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset; By using a pre-set neural network model, based on the structured dataset, each department is divided into at least two different department groups; Based on the structured dataset and the department group, the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department are calculated. Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, a target network latency reduction scheme is obtained.

[0107] As can be seen from the above, the embodiments of this application can achieve precise and efficient management of network services of multiple departments within a company, thereby minimizing network latency at the company level while ensuring that the network service quality of each department reaches the required level.

[0108] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be accomplished by instructions, or by instructions controlling related hardware. These instructions can be stored in a medium and loaded and executed by a processor.

[0109] To this end, embodiments of this application provide a medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the network latency reduction methods provided in embodiments of this application. For example, the instructions can execute the following steps: Preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset; By using a pre-set neural network model, based on the structured dataset, each department is divided into at least two different department groups; Based on the structured dataset and the department group, the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department are calculated. Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, a target network latency reduction scheme is obtained.

[0110] The medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0111] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a medium. A processor of a computer device reads the computer instructions from the medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations of the above embodiments.

[0112] Since the instructions stored in the medium can execute the steps of any of the network latency reduction methods provided in the embodiments of this application, the beneficial effects that any of the network latency reduction methods provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0113] The foregoing has provided a detailed description of a network latency reduction method, apparatus, terminal, and medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for reducing network latency, characterized in that, The method includes: Preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset; By using a pre-set neural network model, based on the structured dataset, each department is divided into at least two different department groups; Based on the structured dataset and the department group, the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department are calculated. Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, a target network latency reduction scheme is obtained.

2. The method as described in claim 1, characterized in that, The preset neural network model includes a first model and a second model. The structured dataset includes department feature vectors and gateway status feature vectors. The first model is used to predict the network demand status of each department, and the second model is used to classify each department based on the network demand status of each department. The method involves using a pre-defined neural network model to divide each department into at least two different department groups based on the structured dataset, including: Using the first model, based on the department feature vector and the gateway status feature vector, the department network demand feature vector corresponding to each department is obtained; The second model is used to classify each department based on the department network demand feature vector to obtain the department group.

3. The method as described in claim 2, characterized in that, The first model includes a first input layer and a hidden layer; The step of obtaining the departmental network demand feature vector for each department through the first model, based on the departmental feature vector and the gateway state feature vector, includes: The first input vector is obtained through the first input layer based on the department feature vector and the gateway status feature vector; The hidden layer is used to obtain the departmental network demand feature vector based on the first input vector.

4. The method as described in claim 2, characterized in that, The second model includes a second input layer and a competition layer; The second model classifies each department based on the departmental network demand feature vector to obtain the departmental group, including: The second input layer obtains the second input vector based on the departmental network demand feature vector. Through the competition layer, the department group is obtained based on the second input vector, wherein the department group is used to characterize the department categories divided based on different latency tolerances and different bandwidth requirements.

5. The method as described in claim 1, characterized in that, The calculation of dynamic bandwidth allocation thresholds and dynamic latency tolerance thresholds for each department, based on the structured dataset and the department groups, includes: Set the initial bandwidth threshold, the initial latency tolerance threshold, and the priority weight for each department group; Based on the real-time traffic and historical average traffic of each department in the structured dataset, the traffic fluctuation factor for each department is calculated. Based on the number of departmental devices in each department and the average number of departmental devices in each department group in the structured dataset, the device density factor for each department is calculated. The dynamic threshold for bandwidth allocation is calculated based on the initial bandwidth threshold, the priority weight, the traffic fluctuation factor, the device density factor, and the current gateway load rate corresponding to each department in the structured dataset. Based on the gateway signal strength of each department in the structured dataset and the baseline gateway signal strength of each department in the structured dataset, the signal quality factor of each department is calculated. The dynamic latency tolerance threshold is calculated based on the initial latency tolerance threshold, the signal quality factor, and the historical average latency of each department in the structured dataset.

6. The method as described in claim 1, characterized in that, The method for obtaining a target network latency reduction scheme based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset includes: Based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset, an objective function and corresponding constraints are constructed with the goal of minimizing overall network latency and maximizing the departmental transmission quality compliance rate. Based on the objective function and the constraints, the objective decision variables are obtained, and based on the objective decision variables, the objective network latency reduction scheme is obtained.

7. The method as described in claim 6, characterized in that, The target decision variables include a first target decision variable and a second target decision variable, and the objective function and the constraints are represented by the following expressions; in, This represents the objective of the objective function. This indicates the preset weighting coefficient. This represents the average network latency data within the company's departments. This indicates the rate of substandard network transmission quality within the company. Indicates the number of departments within the company. This indicates the number of intranet gateways within the company. Indicates department via gateway network latency, This represents the first objective decision variable. Indicates department By gateway Serve, Indicates department Not by gateway Serve, Indicates an indicator function with attached conditions. The condition for the indicated function is a department. via gateway Network latency greater than department The corresponding delay tolerance dynamic threshold, when the condition of the indicator function is met, When the condition of the indicator function is not met, , Indicates department The corresponding delay tolerance dynamic threshold, This represents the second objective decision variable, which is used to characterize the gateway. Assigned to department bandwidth data, Indicates gateway Total bandwidth capacity Indicates department The corresponding dynamic threshold for bandwidth allocation, Indicates department via gateway Actual network traffic This means that any department within a department must be served by at least one gateway.

8. A network latency reduction device, characterized in that, The device includes: The first unit is used to preprocess the departmental business data and gateway status data of each department within the company to obtain a structured dataset. The second unit is used to divide each department into at least two different department groups based on the structured dataset using a preset neural network model. The third unit is used to calculate the dynamic threshold for bandwidth allocation and the dynamic threshold for latency tolerance for each department based on the structured dataset and the department group. The fourth unit is used to obtain a target network latency reduction scheme based on the bandwidth allocation dynamic threshold, the latency tolerance dynamic threshold, and the structured dataset.

9. A terminal, characterized in that, The method includes a processor and a memory, the memory storing multiple instructions; the processor loads instructions from the memory to perform the steps of the method as described in any one of claims 1 to 7.

10. A medium, characterized in that, The medium stores a plurality of instructions adapted for loading by a processor to execute the steps of the method according to any one of claims 1 to 7.

Images