An optical module and switch automatic adaptation method

Abstract

This invention relates to the field of automatic switch adaptation technology, and discloses an automatic adaptation method for optical modules and switches, comprising: collecting multi-dimensional data, constructing a failure prediction model through difference calculation, outputting adaptation risk levels, and matching main adaptation plans; constructing a fault feature library, and quickly locating the root cause by analyzing fault matching degree; adjusting key parameters using a self-healing strategy, evaluating the effectiveness of the adjustment, and performing degradation processing; collecting key operating parameters in real time after successful adaptation, and adjusting core parameters to stabilize the adaptation state; simultaneously calculating resource usage optimization amount and dynamically adjusting resource allocation; locating short-term main causes through re-failure cause matching degree; employing a rapid reconstruction and data linkage strategy, calculating reconstruction adaptation parameters and performing linkage adjustments; linking process data sources to construct a short-term adaptation data closed loop; optimizing adaptation templates and strategies by calculating iterative optimization coefficients; verifying the optimization effect and forming an iterative closed loop.

Description

Technical Field

[0001] This invention relates to the field of automatic switch adaptation technology, and specifically to a method for automatic adaptation between an optical module and a switch. Background Technology

[0002] With the rapid development of industrial internet, data centers and 5G bearer networks, the adaptation scenarios for optical modules and switches are becoming increasingly diversified, forming a pattern where two core scenarios coexist: short-term temporary services and long-term normal services.

[0003] In existing technologies, short-term temporary services, such as industrial production line debugging and emergency monitoring, are characterized by short duration, moderate environmental fluctuations, drastic traffic fluctuations, and high service priority. They require high adaptation efficiency, rapid initialization, and avoidance of debugging stagnation due to adaptation failures, while also requiring significant resource consumption. Forced collection of irrelevant data, such as equipment aging and lifecycle failures, leads to long collection cycles, data redundancy, and low adaptation initialization efficiency, failing to meet the rapid startup requirements of temporary services. Long-term, routine services, such as data center AI training, 5G bearer network fronthaul / backhaul, and routine operation in industrial wide-temperature environments, face long-term challenges such as equipment aging, cross-vendor hybrid networking, and firmware upgrade compatibility, demanding stringent requirements for adaptation stability and lifecycle adaptation capabilities. The lack of targeted collection of key data such as aging status and historical adaptation records leads to omissions in risk prediction. Factors contributing to this include: insufficient stability in adaptation; complex fault tree hierarchy in short-term temporary business scenarios, failing to meet the need for rapid repair of temporary business needs; lack of integration of aging and historical fault data in long-term routine business scenarios, and failure to differentiate the impact weight of features on different scenarios when integrating short-term dynamic features and long-term static features, resulting in low accuracy of risk prediction; fixed calibration cycles in short-term scenarios, unable to respond to rapidly fluctuating environments and traffic, and untimely parameter drift correction; lack of aging prediction integration in long-term scenarios, making stability degradation due to aging still possible after calibration; short-term scenario calibration parameters being written into long-term configurations, leading to subsequent routine business adaptation conflicts; long-term scenario calibration parameters not being updated to the template library, resulting in a lack of basis for iterative optimization; failure to design prevention and control strategies for major causes such as sudden increases in short-term environment and aggravated long-term aging, resulting in a high probability of re-failure; and reliance on repeated adaptation attempts after re-failure, leading to low reconstruction efficiency.

[0004] Therefore, there is a need to provide a method for automatic adaptation between optical modules and switches. Summary of the Invention

[0005] The purpose of this invention is to provide an automatic adaptation method for optical modules and switches. To solve the above-mentioned problems in the prior art, this invention achieves this through the following technical solution:

[0006] The first part, an embodiment of the present invention, provides an automatic adaptation method for optical modules and switches, which specifically includes the following steps:

[0007] Step 1: Collect multi-dimensional data and trigger the collection of redundant parameters as needed; build a failure prediction model through difference calculation, accurately output the adaptation risk level, and match the main adaptation plan;

[0008] Step 2: Combine multi-dimensional data to build a fault feature library, quickly locate the root cause by analyzing the fault matching degree; adopt a self-healing strategy to adjust key parameters, evaluate the effectiveness of the adjustment by the self-healing effect value, and perform downgrade processing;

[0009] Step 3: Collect key operating parameters in real time after successful adaptation, and adjust the core parameters to stabilize the adaptation state through dynamic calibration coefficients; at the same time, calculate the resource usage optimization and dynamically adjust resource allocation.

[0010] Step 4: Combining historical data and real-time operational data, locate the main short-term causes by matching the re-failure causes; adopt a rapid reconstruction and data linkage strategy to calculate reconstruction adaptation parameters and make linkage adjustments.

[0011] Step 5: Link the data sources of the process to build a short-term adaptive data closed loop; optimize the adaptation template and strategy by calculating the iterative optimization coefficients; verify the optimization effect and form an iterative closed loop.

[0012] The second part, an embodiment of the present invention, provides an automatic adaptation system for optical modules and switches, specifically including the following modules:

[0013] Risk grading module: Collects multi-dimensional data and triggers the collection of redundant parameters as needed; constructs a failure prediction model through difference calculation, accurately outputs the appropriate risk level, and matches the main appropriate contingency plan;

[0014] Assessment and Adjustment Module: Combines multi-dimensional data to build a fault feature library, quickly locates the root cause by analyzing the fault matching degree; adopts a self-healing strategy to adjust key parameters, evaluates the effectiveness of the adjustment by the self-healing effect value, and performs downgrade processing;

[0015] The optimization allocation module collects key operating parameters in real time after successful adaptation, adjusts core parameters to stabilize the adaptation state through dynamic calibration coefficients, and calculates resource usage optimization to dynamically adjust resource allocation.

[0016] Reconstruction and Adjustment Module: Combining historical data and real-time operational data, it identifies the main short-term causes by matching the re-failure triggers; it adopts a rapid reconstruction and data linkage strategy to calculate reconstruction adaptation parameters and perform linkage adjustments.

[0017] Optimize the adaptation module: Link the process data source to build a short-term adaptation data closed loop; optimize the adaptation template and strategy by calculating the iterative optimization coefficient; verify the optimization effect to form an iterative closed loop.

[0018] The beneficial effects of this invention are:

[0019] 1. To address the critical pain points of short-term adaptation scenarios, a comprehensive system has been constructed, encompassing multi-dimensional data collection, risk prediction, fault self-healing, dynamic calibration, and data closed-loop optimization. This system employs a short-process data collection and verification mechanism and a differentiated difference calculation model, combined with a composite algorithm of fixed weights and dynamic correction factors. This enables precise quantification of adaptation risks and intelligent triggering of redundant parameter collection, simplifying the complexity of dynamic weight calculation. A database of key fault characteristics and self-healing strategies for short-term scenarios has been built. Root causes are accurately located using feature matching formulas, and targeted parameter adjustments and self-healing effect evaluations are combined to avoid the drawbacks of blind debugging in traditional adaptation. A dynamic calibration and resource optimization linkage mechanism, along with a full-process data closed-loop iteration system, has been constructed to achieve the dual goals of stable adaptation status and efficient resource utilization, overcoming the static limitations of traditional adaptation which terminates after a single completion.

[0020] 2. By matching risk prediction with major contingency plans, the accuracy of adaptation is greatly improved, and the consumption of resources for ineffective adaptation is reduced; the fault self-healing and degradation handling mechanism shortens the time for adaptation fault repair and ensures the continuous transmission of core services; the dynamic calibration and resource optimization strategy balances the stability of adaptation and the efficiency of resource utilization; the full-process data closed loop and iterative optimization mechanism enables the adaptation templates and strategies to be continuously upgraded, improves the adaptability and generalization ability of the solution to short-term scenarios, and reduces operation and maintenance costs and debugging downtime losses. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the steps of an automatic adaptation method between an optical module and a switch provided in Embodiment 1 of the present invention;

[0023] Figure 2 This is a schematic diagram of the structure of an automatic adaptation system for optical modules and switches provided in Embodiment 2 of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0025] Example 1: As Figure 1 As shown in the figure, an automatic adaptation method for optical modules and switches provided by an embodiment of the present invention specifically includes the following steps:

[0026] Step 1: Collect multi-dimensional data and trigger the collection of redundant parameters as needed; build a failure prediction model through difference calculation, accurately output the adaptation risk level, and match the main adaptation plan;

[0027] In a specific embodiment, the multi-dimensional data includes: optical module basic parameters, port configuration parameters, and core link environment parameters; the optical module basic parameters are collected in real time through the parameter acquisition unit built into the optical module, and the optical module basic parameters include: optical module model identifier, rated operating rate, rated transmit optical power, and received optical power threshold.

[0028] Port configuration parameters are collected through monitoring units integrated into the switch ports. These parameters include: port speed configuration, port communication protocol version, and maximum supported power. Core link environmental parameters are collected through miniature environmental sensors deployed in the link. These parameters include: link transmission distance and current link bit error rate.

[0029] During the data collection process, a short-process data collection and verification mechanism is adopted. After each multi-dimensional data collection is completed, the format is immediately verified. If the verification passes, the next multi-dimensional data collection will proceed. If the verification fails, the repeated collection of the corresponding multi-dimensional data will be terminated directly, and the collection anomaly flag will be recorded to avoid invalid loop collection.

[0030] The difference between the collected optical module basic parameter values ​​and the midpoint value of the preset optical module basic parameter range of the corresponding switch port is calculated, and the absolute value is then compared with the midpoint value of the optical module basic parameter range to obtain the module basic parameter difference degree.

[0031] The difference between the obtained port configuration parameter collection value and the midpoint value of the corresponding optical module's supported configuration parameter range is calculated, and the absolute value is then compared with the midpoint value of the configuration parameter range to obtain the port configuration parameter difference degree.

[0032] Obtain the difference in bit error rate fluctuation among the core link environment parameters in the last three times. Calculate the difference by taking the absolute value of the difference between the current bit error rate and the average of the previous two bit error rates.

[0033] A risk prediction model is established by combining the differences in basic parameters, port configuration parameters, and bit error rate fluctuations, using the formula:

[0034]

[0035] Analysis yields appropriate risk levels ,in, This is a rounding function to ensure that the risk level is an integer. To determine the differences in basic module parameters, This is the weighting coefficient for the module's basic parameters. In short-term adaptation scenarios, the module's basic parameters have a significant impact on adaptation. The preset value is 0.6, which is the optimal fixed value verified through a large number of short-term adaptation scenarios, avoiding the complex calculation of dynamic weights. Configure the port parameter difference. Configure the port parameter weight coefficient. The difference correction factor is a dynamically adjusted parameter with a value range of [0.8, 1.5]. It is dynamically corrected based on the adaptation failure rate of short-term adaptation scenarios in historical adaptation data, while also controlling the trigger threshold for redundant parameter collection. This represents the difference in bit error rate fluctuation. This is the fluctuation trend weighting coefficient, which adjusts the impact of link fluctuations on adaptation failures in short-term adaptation scenarios. The preset value is 0.3.

[0036] After analyzing and determining the appropriate risk level, the main appropriate contingency plan is matched.

[0037] Specifically, Level 0 risk: A rapid adaptation plan is adopted, directly entering the basic adaptation process without redundant parameter verification; Level 1 risk: Redundant parameter collection is triggered, and after supplementary verification, the regular adaptation process is entered, with the monitoring of two key adaptation nodes added; Level 2 risk: A parameter pre-adjustment and adaptation plan is adopted, and parameters with large differences are initially fine-tuned based on the calculation results of differences in basic module parameters and port configuration parameters, before entering the adaptation process; Level 3 risk and above: The adaptation suspension and early warning plan is directly activated to avoid ineffective adaptation consuming resources, while prompting maintenance personnel to confirm parameter compatibility;

[0038] After the core parameters are collected, if the risk prediction model initially determines that there is a potential compatibility risk, the collection of redundant parameters is triggered. The redundant parameters include: the operating temperature threshold of the optical module and the buffer capacity of the switch port. The collection frequency is reduced to 5Hz to further reduce resource consumption.

[0039] If the core parameters are collected without any abnormalities and the initial assessment indicates no risk, then skip the collection of redundant parameters and proceed directly to the adaptation plan matching stage.

[0040] Step 2: Combine multi-dimensional data to build a fault feature library, quickly locate the root cause by analyzing the fault matching degree; adopt a self-healing strategy to adjust key parameters, evaluate the effectiveness of the adjustment by the self-healing effect value, and perform downgrade processing;

[0041] In a specific embodiment, a fault feature library is constructed based on common adaptation failure types in short-term scenarios. The fault feature library contains 5 types of feature parameters: parameter incompatibility feature value F1, link bit error rate exceeding standard feature value F2, insufficient optical power feature value F3, protocol version mismatch feature value F4, and insufficient port buffer feature value F5. The feature parameters are standardized, and a clear parameter abnormality range is set for each fault feature in combination with the collected standard parameter range.

[0042] Root cause localization employs a feature matching and deviation calculation method: Real-time parameter data is extracted when adaptation fails. This data includes the current values ​​of module basic parameters, port configuration parameters, and link environment parameters. Preliminary matching of this data with feature values ​​in the fault feature database is performed to filter out candidate fault types with a matching degree greater than or equal to 60%. The root cause is then precisely located using the fault matching degree calculation, based on the formula:

[0043]

[0044] The analysis yields the fault matching degree M, where, The number of feature parameters corresponding to the candidate fault types. These are real-time parameter values, i.e., the actual values ​​of the parameters corresponding to the fault characteristics collected when adaptation fails. These are the standard values ​​for fault characteristics, that is, the abnormal standard values ​​of parameters corresponding to the fault type in the fault characteristic database. This is the feature weight correction coefficient, with a value range of [0.9, 1.2]. It is dynamically determined in conjunction with the risk level R; the higher the risk level, the larger the feature weight correction coefficient. The maximum value of the characteristic parameter. The minimum value of the characteristic parameter. The index of the feature parameters corresponding to the candidate fault type;

[0045] The candidate fault type with the largest fault matching degree M is taken as the root cause of the adaptation failure. If there are multiple faults with the same matching degree, further screening is carried out in combination with the results of redundant parameter collection, and the fault type related to the abnormality of redundant parameters is given priority.

[0046] For different root causes, a self-healing strategy is adopted, adjusting the self-healing adjustment amount to avoid complex multi-parameter linkage adjustments and shorten the recovery time, through the formula: Analysis yielded self-healing adjustment amount ,in, The value is the midpoint of the standard parameter range. These are real-time parameter values. This is the self-healing adjustment coefficient, with a value range of [0.6, 1.0].

[0047] Specific self-healing adjustment methods for different root causes:

[0048] Parameter incompatibility: Adjust the switch port speed configuration parameters; the adjustment amount should be a self-healing adjustment amount. After adjustment, the rated operating rate of the optical module remains unchanged to ensure core rate matching;

[0049] Link bit error rate exceeds standard: Adjust the optical module's transmit optical power by a self-healing adjustment amount. At the same time, it reduces bandwidth for non-core services and improves transmission stability;

[0050] Insufficient optical power: Adjust the emitted optical power of the optical module by a self-healing adjustment amount. This ensures that the emitted optical power reaches more than 80% of the rated value;

[0051] Protocol version mismatch: Adjust the communication protocol version of the switch port to the lowest version compatible with the optical module;

[0052] Insufficient port cache: Adjust the switch port cache capacity allocation, temporarily allocating cache resources from non-core services to the currently adapted services. The adjustment amount is a self-healing adjustment amount. Ensure that the cache capacity meets the current business transmission requirements;

[0053] After self-healing adjustment, the effectiveness of the adjustment is verified using a self-healing effect evaluation formula. The formula is as follows:

[0054]

[0055] The analysis yields a self-healing effect value E, which ranges from [0,1]. If the self-healing effect value E is greater than or equal to 0.8, the self-healing is considered successful, and the adaptation retry process begins. If E is less than 0.8, the self-healing is considered to have failed, and a downgrade process is executed.

[0056] in, To adjust the parameter values, The correction coefficient for effect evaluation is preset to 0.9 to avoid complex adjustments to dynamic coefficients;

[0057] The degradation process adopts a core business priority protection strategy: if self-healing fails, the currently adapted business is immediately switched to a preset temporary transmission channel. The channel guarantees the transmission of core business data, such as critical monitoring data and instruction data, while suspending the transmission of non-core business data. At the same time, the degradation status is recorded and early warning information is sent to the operation and maintenance personnel to ensure that the temporary core business is not interrupted and to reduce the losses caused by debugging stagnation.

[0058] Step 3: Collect key operating parameters in real time after successful adaptation, and adjust the core parameters to stabilize the adaptation state through dynamic calibration coefficients; at the same time, calculate the resource usage optimization and dynamically adjust resource allocation.

[0059] In a specific embodiment, after successful adaptation, a collection strategy of high-frequency acquisition of core operating parameters and low-frequency acquisition of auxiliary operating parameters is adopted to avoid excessive resource consumption by full high-frequency acquisition.

[0060] It should be noted that the core operating parameters include: the actual transmit optical power of the optical module, the actual receive optical power, the actual operating rate of the switch port, and the real-time bit error rate of the link. The sampling frequency is set to 8Hz to ensure timely capture of parameter fluctuations. The auxiliary operating parameters include: the operating temperature of the optical module and the resource utilization rate of the switch port. The sampling frequency is set to 3Hz for resource optimization evaluation.

[0061] During the data collection process, a fluctuation-triggered storage mechanism is adopted. If the fluctuation value of the core operating parameter exceeds the preset threshold, the historical data of the corresponding core operating parameter is stored; otherwise, it is not stored, thereby reducing the data storage resource consumption.

[0062] By combining the fluctuation trends of core operating parameters with the dynamic calibration coefficients based on historical adaptation data, the problems of unstable adaptation and excessive resource consumption are simultaneously addressed through dynamic adjustment of the dynamic calibration coefficients, using the formula:

[0063]

[0064] Analysis yields dynamic calibration coefficients ,in, This represents the average fluctuation value of the core operating parameters over the last 5 times. The standard fluctuation threshold for the core operating parameters, The fluctuation correction weight is preset to 0.3. This represents the current resource utilization rate of the switch port. This represents the maximum allowed resource utilization rate for the switch port. The default value for resource optimization weight is 0.2;

[0065] The target parameter value of the core operating parameters, i.e. the initial parameter value when the adaptation is successful, is multiplied by the dynamic calibration coefficient to obtain the adjusted target adjustment parameter value, which is then adjusted in real time through the parameter adjustment interface between the optical module and the switch.

[0066] For example, the target value of the actual transmitted optical power of the optical module is P0, and the adjusted target value is P0×K. If the average fluctuation value of the core operating parameters in the last 5 times is greater than the preset fluctuation threshold, then K is greater than 1, and the target parameter value is appropriately increased to enhance transmission stability; if the current resource occupancy rate of the switch port is greater than the preset occupancy threshold, then K is less than 1, and the target parameter value is appropriately decreased to reduce resource consumption.

[0067] Combining the resource occupancy rate data in the auxiliary operation parameters, the difference between the current resource occupancy rate of the switch port and the target value of the switch port resource occupancy is calculated, and then multiplied with the preset dynamic resource optimization coefficient to obtain the resource occupancy optimization amount. The resource optimization coefficient has a value range of [0.5, 0.8], which is determined by the dynamic calibration coefficient to prioritize the adaptation stability.

[0068] Specific resource optimization measures: Dynamically adjust the resource allocation of adapted services based on the resource usage optimization amount, such as: reducing the transmission priority of non-core data and reducing the amount of cached resources allocated; if the resource usage optimization amount is greater than 10%, temporarily restrict the non-core functions of adapted services to ensure that resource usage is reduced to below the target value;

[0069] Step 4: Combining historical data and real-time operational data, locate the main short-term causes by matching the re-failure causes; adopt a rapid reconstruction and data linkage strategy to calculate reconstruction adaptation parameters and make linkage adjustments.

[0070] In a specific embodiment, a linkage acquisition mechanism for historical data and real-time data is constructed: the core operating parameters at the time of re-failure are collected in real time, and at the same time, the adaptive risk level data, self-healing adjustment amount and dynamic calibration coefficient are called to form a re-failure data matrix.

[0071] Based on the main characteristics of short-term re-failure scenarios, a short-term re-failure cause library is constructed, which includes four main short-term causes: short-term link fluctuation exacerbation cause value G1, temporary service traffic surge cause value G2, optical module short-term temperature drift cause value G3, and switch port temporary excessive load cause value G4. Each cause corresponds to a clear combination of historical data characteristics and real-time data characteristics.

[0072] By combining the linked data matrix, the triggering factors can be accurately located using the formula:

[0073]

[0074] Analysis yielded the re-failure cause matching degree The value range is [0,1]. If the match degree of the failure cause is greater than or equal to 0.7, it is determined that the cause has been accurately located. Historical data matching degree, which is the ratio of the number of matched historical data features to the total number of corresponding historical data features. The weights are historical data, with values ​​ranging from [0.4, 0.6]. Real-time data matching degree, which is the ratio of the number of matched real-time data features to the total number of corresponding real-time data features. This is the linkage correction coefficient, with a preset value of 1.1;

[0075] If there are multiple triggers with a relapse trigger matching degree greater than or equal to 0.7, then calculate the absolute value of the difference between the historical data matching degree and the real-time data matching degree of each trigger to obtain the linkage deviation. The one with the smallest linkage deviation is the main trigger. The smaller the deviation, the more consistent the historical data characteristics are with the real-time data characteristics, and the higher the credibility of the trigger.

[0076] For different main causes, a strategy of rapid reconstruction of adaptation parameters and linkage adjustment with historical data is adopted to adjust the linkage adjustment amount, avoiding a complex re-adaptation process;

[0077] After re-adaptation, a rapid verification process is initiated to verify the adaptation status of core business functions. If the verification is successful, business transmission is restored.

[0078] Step 5: Link the data sources of the process to build a short-term adaptive data closed loop; optimize the adaptation template and strategy by calculating the iterative optimization coefficients; verify the optimization effect and form an iterative closed loop;

[0079] In a specific embodiment, a full-process data linkage and integration model is constructed. The integrated process data sources include: multi-dimensional data, risk level matching data, contingency plan matching data, root cause data, self-healing adjustment data, degradation processing data, dynamic calibration data, resource optimization data, re-failure trigger data, reconstruction adjustment data, and rapid verification data.

[0080] During the integration process, a unified mapping mechanism for key fields is adopted to map and associate each core parameter with a unified term, forming a short-term adaptation data matrix for the entire process. The data in the short-term adaptation data matrix is ​​sorted by timestamp for easy traceability and analysis.

[0081] Establish a data quality assessment mechanism to eliminate invalid data. The assessment indicators are data integrity and data consistency to ensure the reliability of linked data. The data integrity is calculated as the ratio of the number of complete data entries to the total number of data entries. The data consistency is calculated as the ratio of the number of consistent data entries in the process data source to the total number of data entries for the corresponding parameter.

[0082] Based on the integrated full-process data matrix, an iterative optimization coefficient calculation method is designed. By dynamically adjusting the iterative optimization coefficient, the adaptation template and adaptation strategy are optimized simultaneously. The success rate of the current short-term adaptation, the efficiency index of the current short-term adaptation, and the failure frequency of the short-term adaptation are weighted and fused to obtain the iterative optimization coefficient.

[0083] Specific measures for iterative optimization:

[0084] Adaptation template optimization: Adjust the parameter range and preset threshold in the adaptation template according to the value of the iterative optimization coefficient. If the iterative optimization coefficient is greater than 1.1, the prediction threshold of the adaptation risk level is reduced by 10%, and the fault feature library is updated to add the current re-failure cause features. If the iterative optimization coefficient is greater than 1 and less than or equal to 1.1, the value range of the dynamic calibration coefficient K is reduced by 0.02 to improve calibration accuracy. If the iterative optimization coefficient is less than or equal to 1.0, the current adaptation template is retained, and the current adaptation data is recorded as historical samples.

[0085] Adaptation strategy optimization: Link up data across the entire process and optimize strategy parameters, such as: adjusting the trigger threshold for redundant parameter collection based on the validity data of the current redundant parameter collection, and adjusting the range of historical data weight η based on the accuracy data of the current cause location, to ensure that the optimized strategy is adapted to the characteristics of short-term setup scenarios.

[0086] After optimization, a closed-loop verification process is initiated: by simulating short-term scenarios and reusing historical data, three sets of historical short-term adaptation data with the same adaptation type as the current adaptation are selected, and the optimized adaptation template and strategy are applied to simulate adaptation to verify whether the three core indicators of adaptation success rate, adaptation efficiency and failure frequency have all improved by more than 10%.

[0087] If all conditions are met, the optimized adaptation template and strategy will be stored in the adaptation strategy library for subsequent short-term adaptation; if not, the iterative optimization coefficients will be recalculated and the optimization measures will be adjusted until the verification is successful.

[0088] Example 2: As Figure 2 As shown in the figure, an automatic adaptation system for optical modules and switches provided in this embodiment of the invention specifically includes the following modules:

[0089] Risk grading module: Collects multi-dimensional data and triggers the collection of redundant parameters as needed; constructs a failure prediction model through difference calculation, accurately outputs the appropriate risk level, and matches the main appropriate contingency plan;

[0090] Assessment and Adjustment Module: Combines multi-dimensional data to build a fault feature library, quickly locates the root cause by analyzing the fault matching degree; adopts a self-healing strategy to adjust key parameters, evaluates the effectiveness of the adjustment by the self-healing effect value, and performs downgrade processing;

[0091] The optimization allocation module collects key operating parameters in real time after successful adaptation, adjusts core parameters to stabilize the adaptation state through dynamic calibration coefficients, and calculates resource usage optimization to dynamically adjust resource allocation.

[0092] Reconstruction and Adjustment Module: Combining historical data and real-time operational data, it identifies the main short-term causes by matching the re-failure triggers; it adopts a rapid reconstruction and data linkage strategy to calculate reconstruction adaptation parameters and perform linkage adjustments.

[0093] Optimize the adaptation module: Link the process data source to build a short-term adaptation data closed loop; optimize the adaptation template and strategy by calculating the iterative optimization coefficient; verify the optimization effect to form an iterative closed loop.

[0094] The above provides a detailed description of one embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. The above formulas are all dimensionless numerical calculations, and the formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world situation. The preset parameters in the formulas are set by those skilled in the art based on actual conditions and historical experience, and can be adjusted according to actual conditions. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. All equivalent changes and improvements made in accordance with the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A method for automatic adaptation between optical modules and switches, characterized in that, Includes the following steps: Collect multi-dimensional data and trigger the collection of redundant parameters as needed; build a failure prediction model through difference calculation, accurately output the adaptation risk level of optical modules and switches, and match the main adaptation plans; By combining multi-dimensional data, a fault feature library is constructed, and the root cause is quickly located by analyzing the fault matching degree; a self-healing strategy is adopted to adjust key parameters, the effectiveness of the adjustment is evaluated by the self-healing effect value, and a downgrade process is performed. After successful adaptation, key operating parameters of the optical module and switch are collected in real time, and the core parameters are adjusted to stabilize the adaptation state through dynamic calibration coefficients; at the same time, the resource consumption optimization of the optical module and switch is calculated, and resource allocation is dynamically adjusted. By combining historical and real-time operational data of optical modules and switches, the main short-term causes are located through the re-failure cause matching degree; a rapid reconstruction and data linkage strategy is adopted to calculate reconstruction adaptation parameters and make linkage adjustments. Link data sources in the process to build a short-term adaptation data closed loop; optimize the adaptation templates and strategies for optical modules and switches by calculating iterative optimization coefficients; Verify the optimization effect and form an iterative closed loop.

2. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for constructing the failure prediction model is as follows: The difference between the collected optical module basic parameter values ​​and the midpoint value of the preset optical module basic parameter range of the corresponding switch port is calculated, and the absolute value is then compared with the midpoint value of the optical module basic parameter range to obtain the module basic parameter difference degree. The difference between the obtained port configuration parameter collection value and the midpoint value of the corresponding optical module's supported configuration parameter range is calculated, and the absolute value is then compared with the midpoint value of the configuration parameter range to obtain the port configuration parameter difference degree. Obtain the bit error rate fluctuation difference in the core link environment parameters of the last 3 times. Calculate the difference by taking the absolute value of the difference between the current bit error rate and the average of the previous two bit error rates.

3. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for matching and adapting the contingency plan is as follows: By combining the differences in basic parameters, port configuration parameters, and bit error rate fluctuations, a risk prediction model is established to analyze and determine the adaptation risk level. After analyzing and determining the appropriate risk level, the main appropriate contingency plan is matched. If the risk prediction model initially determines that there is a potential compatibility risk, it will trigger the collection of redundant parameters, including: the operating temperature threshold of the optical module and the buffer capacity of the switch port. If the core parameters are collected without any abnormalities and the initial assessment indicates no risk, then skip the collection of redundant parameters and proceed directly to the adaptation plan matching stage.

4. The automatic adaptation method for optical modules and switches according to claim 1, characterized in that, The method for quickly locating the root cause is as follows: The feature matching and deviation calculation method is adopted: real-time parameter data when the adaptation fails is extracted, and the real-time parameter data is initially matched with each feature value in the fault feature library to filter out candidate fault types. The root cause is accurately located by calculating the fault matching degree, and the fault matching degree is obtained by analysis. The candidate fault type with the highest fault matching degree is selected as the root cause of the adaptation failure. If there are multiple faults with the same matching degree, further screening is carried out in combination with the results of redundant parameter collection, and the fault type related to the abnormality of redundant parameters is selected first.

5. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for assessing the effectiveness of the adjustment is as follows: For different root causes, a self-healing strategy is adopted, adjusting the self-healing adjustment amount to avoid complex multi-parameter linkage adjustments and shorten the recovery time, through the formula: Analysis yielded self-healing adjustment amount ,in, The value is the midpoint of the standard parameter range. These are real-time parameter values. This is the self-healing adjustment coefficient, with a value range of [0.6, 1.0]. After self-healing adjustment, the effectiveness of the adjustment is verified by the self-healing effect evaluation formula, and the self-healing effect value is obtained by analysis. The degradation process adopts a core business priority protection strategy: if self-healing fails, the currently adapted business will be immediately switched to a preset temporary transmission channel.

6. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for achieving a stable adaptation state is as follows: A fluctuation-triggered storage mechanism is adopted. If the fluctuation value of the core operating parameter exceeds the preset threshold, the historical data of the corresponding core operating parameter is stored; otherwise, it is not stored, thereby reducing the consumption of data storage resources. By combining the fluctuation trend of core operating parameters with the dynamic calibration coefficient of historical adaptation data, the dynamic calibration coefficient is obtained through dynamic adjustment. The target parameter value of the core operating parameters, i.e. the initial parameter value when the adaptation is successful, is multiplied by the dynamic calibration coefficient to obtain the adjusted target parameter value, which is then adjusted in real time through the parameter adjustment interface between the optical module and the switch.

7. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for dynamically adjusting resource allocation is as follows: Combining the resource occupancy rate data in the auxiliary operation parameters, the difference between the current resource occupancy rate of the switch port and the target value of the switch port resource occupancy is calculated, and then multiplied with the preset dynamic resource optimization coefficient to obtain the resource occupancy optimization amount. The resource optimization coefficient has a value range of [0.5, 0.8], which is determined by the dynamic calibration coefficient to prioritize the adaptation stability. Specific resource optimization measures: Dynamically adjust resource allocation to adapt to the business based on the amount of resource usage optimization.

8. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for identifying the primary short-term triggers is as follows: Construct a linkage acquisition mechanism for historical and real-time data: collect core operating parameters during re-failure in real time, and simultaneously call up adaptive risk level data, self-healing adjustment amount and dynamic calibration coefficient to form a re-failure data matrix; Based on the main characteristics of short-term re-failure, a short-term re-failure cause library is constructed, with each cause corresponding to a clear combination of historical and real-time data features; By combining the linked data matrix, the triggering factors can be accurately located using the formula: Analysis yielded the re-failure cause matching degree The value range is [0,1]. If the match degree of the failure cause is greater than or equal to 0.7, it is determined that the cause has been accurately located. Historical data matching degree, which is the ratio of the number of matched historical data features to the total number of corresponding historical data features. The weights are historical data, with values ​​ranging from [0.4, 0.6]. Real-time data matching degree, which is the ratio of the number of matched real-time data features to the total number of corresponding real-time data features. This is the linkage correction coefficient, with a preset value of 1.1; If there are multiple triggers with a relapse trigger matching degree greater than or equal to 0.7, then calculate the absolute value of the difference between the historical data matching degree and the real-time data matching degree of each trigger to obtain the linkage deviation. The trigger with the smallest linkage deviation is the main trigger.

9. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for performing the linkage adjustment is as follows: For different main causes, a strategy of rapid reconstruction of adaptive parameters and linkage adjustment with historical data is adopted to adjust the linkage adjustment amount; After re-adaptation, a rapid verification process is initiated to verify the adaptation status of core business functions. If the verification is successful, business transmission is restored.

10. The method for automatic adaptation of optical modules and switches according to claim 1, characterized in that, The method for verifying the optimization effect is as follows: A full-process data linkage and integration model is constructed, and a unified mapping mechanism for key fields is adopted to map and associate each core parameter according to a unified term, forming a short-term adaptive data matrix for the entire process. The data in the short-term adaptive data matrix is ​​sorted by timestamp. Establish a data quality assessment mechanism to eliminate invalid data, with assessment indicators including data integrity and data consistency; Based on the integrated full-process data matrix, an iterative optimization coefficient calculation method is designed. By dynamically adjusting the iterative optimization coefficient, the adaptation template and adaptation strategy are optimized simultaneously. The success rate of the current short-term adaptation, the efficiency index of the current short-term adaptation, and the failure frequency of the short-term adaptation are weighted and fused to obtain the iterative optimization coefficient. After optimization, a closed-loop verification process is initiated: by simulating short-term scenarios and reusing historical data, three sets of historical short-term adaptation data with the same type as the current adaptation are selected, and the optimized adaptation template and strategy are applied to simulate adaptation.

Images