Optimization method and system for optical cable transmission of data center

Abstract

The embodiment of the application provides a kind of cable transmission intelligent optimization method and system for data center, it is related to data center technical field, the light cable section granularity analysis is carried out to data center cable transmission architecture, obtains light cable section unit set;Based on initial physical attribute record, construct physical topology connectivity graph, based on initial transmission bearing state record, generate transmission health state degradation trend curve, joint transmission path elastic reconfigurability analysis is carried out to obtain elastic reconfiguration applicability identification;Accordingly, the candidate elastic reconfiguration light cable section unit subset is screened, with physical topology connectivity graph as constraint, call transmission path elastic reconfiguration strategy generation model, generate the elastic transmission path configuration scheme containing main and backup transmission path;When detecting transmission link degradation trigger signal, analyze the path switching node sequence and timing arrangement information of backup path, execute traffic piecewise migration operation.The application realizes the elastic self-healing and accurate switching of data center cable transmission network.

Description

Technical Field

[0001] This application relates to the field of data center technology, and more specifically, to a method and system for intelligent optimization of optical fiber transmission in data centers. Background Technology

[0002] In the management of optical fiber transmission networks in data centers, existing technologies typically plan transmission paths based on the static topology of the optical fiber network. When a segment of the optical fiber experiences transmission degradation or interruption, service traffic is switched to a backup path through manual configuration or preset static switching rules. In this method, static topology planning is only performed once during the initial stage of network construction. The real-time health status changes of each optical fiber segment cannot be detected during subsequent operation. Furthermore, the selection of backup paths is usually based on pre-set fixed rules, failing to consider the differentiated degradation trends of each optical fiber segment due to factors such as accumulated transmission load and fiber core aging over long-term operation. This leads to situations where, in some scenarios, the backup path itself is already under high load or nearing degradation, yet it is still selected as the switching target, causing secondary failures. Furthermore, existing technologies, when replanning transmission paths, typically treat each optical cable segment as a homogeneous node, calculating only the shortest path or minimum hop count based on the currently available bandwidth. They fail to quantitatively assess the applicability of each segment for flexible reconfiguration, i.e., they do not distinguish which segments possess the structural conditions for load redistribution without affecting overall network connectivity. They also fail to consider the constraints of spatial topology endpoint identifiers and fiber core capacity parameters on path reconfiguration feasibility, potentially rendering the replanning results infeasible in practice due to physical topology constraints. Simultaneously, existing technologies typically employ a full-scale switchover approach when performing path switching, migrating all traffic on the primary transmission path to the backup transmission path at once. This lack of segment-by-segment timing orchestration during the switchover process can easily lead to backup path congestion or even switchover failure in high-bandwidth data center scenarios due to instantaneous traffic surges. Moreover, the handling strategies generated by existing methods usually only output switchover instructions, without including a complete flexible transmission path configuration scheme for both the primary and backup transmission paths. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a method and system for intelligent optimization of optical cable transmission in data centers.

[0004] In conjunction with the first aspect of this application, a method for intelligent optimization of optical fiber transmission in data centers is provided, applied to an intelligent optimization system for optical fiber transmission in data centers, the method comprising: The optical cable transmission architecture of the data center is analyzed at the optical cable segment level to obtain the set of optical cable segment units contained in the optical cable transmission architecture of the data center. The initial physical attribute record and the initial transmission carrying status record of each optical cable segment unit are extracted. The initial physical attribute record includes the spatial topology start and end point identifiers of the optical cable segment unit and the fiber core capacity parameters of the optical cable segment unit. The initial transmission carrying status record includes the current transmission channel occupancy distribution information of the optical cable segment unit and the historical transmission interruption event markers of the optical cable segment unit. Based on the initial physical attribute records of each optical cable segment unit in the optical cable segment unit set, a physical topology connectivity diagram of the data center optical cable transmission architecture is constructed. Based on the initial transmission carrying status record of each optical cable segment unit, a transmission health status degradation trend curve of the corresponding optical cable segment unit is generated. Based on the physical topology connectivity diagram and the transmission health status degradation trend curve, a transmission path elastic reconfigurability analysis is performed on the data center optical cable transmission architecture to obtain the elastic reconfiguration applicability identifier of each optical cable segment unit in the data center optical cable transmission architecture. Based on the elastic reconfiguration applicability identifier, a subset of candidate elastic reconfiguration optical cable segment units is selected from the set of optical cable segment units, and the available transmission resource margin record of each candidate elastic reconfiguration optical cable segment unit in the subset of candidate elastic reconfiguration optical cable segment units is determined. Using the physical topology connectivity graph as the constraint boundary, a preset transmission path elastic reconfiguration strategy generation model is called to perform transmission path replanning on the subset of candidate elastic reconfiguration optical cable segment units and the available transmission resource margin record, generating an elastic transmission path configuration scheme that includes a primary transmission path and a backup transmission path. When a transmission link degradation trigger signal is detected, the path switching node sequence and path switching timing arrangement information of the backup transmission path in the elastic transmission path configuration scheme are parsed, and the transmission carrying traffic is migrated from the primary transmission path to the backup transmission path segment by segment according to the path switching node sequence and the path switching timing arrangement information.

[0005] In conjunction with the second aspect of this application, a smart optimization system for optical fiber transmission in data centers is provided. The smart optimization system for optical fiber transmission in data centers includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the smart optimization system for optical fiber transmission in data centers implements the aforementioned smart optimization method for optical fiber transmission in data centers.

[0006] In conjunction with a third aspect of this application, a computer-readable storage medium is provided, wherein computer-executable instructions are stored therein, and when the computer-executable instructions are executed, the aforementioned intelligent optimization method for optical cable transmission in data centers is implemented.

[0007] Combining any of the above aspects, by performing fine-grained analysis of the data center optical cable transmission architecture at the optical cable segment level, each optical cable segment is treated as an independent unit, extracting its spatial topology start and end point identifiers, fiber core capacity parameters, transmission channel occupancy distribution information, and historical transmission interruption event markers. Instead of using static topology as the sole planning basis, a joint analysis of the physical topology connectivity diagram and the transmission health degradation trend curves of each optical cable segment unit is conducted to perform transmission path elastic reconfigurability analysis on the entire data center optical cable transmission architecture. This assigns an elastic reconfigurability identifier to each optical cable segment unit, distinguishing which optical cable segments possess the structural conditions and health margin to undertake load balancing functions during network reconfiguration. This ensures that the subsequent selection of candidate elastic reconfigurable optical cable segment unit subsets is no longer a blind bandwidth comparison, but rather based on elasticity... The system selects the appropriate reconfiguration strategy based on the physical topology connectivity graph as a rigid constraint boundary. This ensures that the generated reconfigurable transmission path configuration scheme, which includes the primary and backup transmission paths, is fully feasible at the physical topology level, while making full use of the available transmission resources of each candidate optical cable segment. When the transmission link degradation trigger signal arrives, the system performs a segment-by-segment migration operation by parsing the path switching node sequence and path switching timing arrangement information of the backup transmission path. This ensures that the migration process of traffic from the primary to the backup transmission path is precisely arranged in the time dimension, avoiding the instantaneous congestion risk caused by full switching. This significantly improves the self-healing capability and service continuity guarantee level of the data center optical cable transmission network in the face of link degradation. Attached Figure Description

[0008] Figure 1 This application provides a flowchart illustrating the intelligent optimization method for optical cable transmission in data centers. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0010] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0011] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0012] Figure 1 This illustration shows a flowchart of an intelligent optimization method for optical fiber transmission in data centers, provided in an embodiment of this application. The method operates in a computing environment with data communication connections to the network element management system, performance monitoring platform, historical fault ticket database, and resource management database of the data center's optical fiber transmission architecture. Each step of the method is automatically implemented by processing units within the computing environment through the execution of pre-compiled program instructions. All data acquisition, transmission, computation, and instruction issuance processes are completed through network interfaces. Before performing any data acquisition operation, the computing environment has obtained explicit authorization from the data center operator. All acquisition activities are limited to the data necessary for the operation and maintenance of the transmission network. Privacy-related data is processed using a differential privacy mechanism with added Laplace noise. Data transmission is encrypted using a transport layer security protocol, and storage is encrypted using an advanced encryption standard algorithm. Access control follows a role-based access control policy, ensuring the legality and compliance of the entire lifecycle of data acquisition, transmission, storage, and use.

[0013] Step S110: Perform fiber optic segment granularity analysis on the data center fiber optic transmission architecture to obtain the set of fiber optic segment units contained in the data center fiber optic transmission architecture, and extract the initial physical attribute record and initial transmission bearer status record of each fiber optic segment unit. The initial physical attribute record includes the spatial topology start and end point identifiers of the fiber optic segment unit and the fiber core capacity parameters of the fiber optic segment unit. The initial transmission bearer status record includes the current transmission channel occupancy distribution information of the fiber optic segment unit and the historical transmission interruption event markers of the fiber optic segment unit.

[0014] Step S111: Parse the fiber optic patch panel port mapping table and fiber optic cable routing diagram information of the data center fiber optic transmission architecture, extract the fiber optic patch panel port identifiers at both ends of the fiber optic cable segment as spatial topology start and end point identifiers, and extract the total number of fiber cores contained in the fiber optic cable segment as the fiber core capacity parameter.

[0015] In this embodiment, the computing environment first accesses the digital engineering document repository of the data center optical cable transmission architecture, and reads the fiber optic distribution frame port mapping table and optical cable routing drawing information. The fiber optic distribution frame port mapping table is stored in the form of a structured data table, with each record containing fields for fiber optic distribution frame identifier, port number, port type, and port number. The optical cable routing drawing information is stored in the form of a vector graphics file, containing layers such as optical cable path, fiber optic distribution frame location, duct route, and splice nodes. The computing environment performs a relational algebraic projection operation on the fiber optic distribution frame port mapping table to extract the fiber optic distribution frame port number field, and simultaneously parses the optical cable layer objects in the optical cable routing drawing information to extract the coordinates of the two endpoints of the optical cable segment objects. The computing environment performs spatial proximity matching between the coordinates of the two endpoints of the optical cable segment and the coordinates of the fiber optic distribution frame. The matching radius is set to a preset spatial tolerance parameter. The port numbers of the fiber optic distribution frames that meet the spatial proximity condition are identified as the port identifiers of the two ends of the optical cable segment, forming the spatial topology start and end point identifiers. This identifier is a string array containing two elements: the first element is the port identifier of the starting fiber optic distribution frame, and the second element is the port identifier of the ending fiber optic distribution frame. Next, the computing environment reads the total number of fiber cores from the attribute field of the optical cable segment object as the core capacity parameter, which is a positive integer scalar. The output of this step is an associated data tuple for each optical cable segment, containing the spatial topology start and end point identifiers and the core capacity parameter.

[0016] Step S112: Aggregate fiber bundles with the same spatial topology start and end point identifiers into a fiber segment unit, and traverse all fiber bundles in the data center fiber cable transmission architecture to form a set of fiber segment units.

[0017] Next, the computing environment performs a hash grouping operation on the spatial topology start and end point identifiers of all optical cable segments extracted in step S111. Using the concatenated string value of the spatial topology start and end point identifier as the hash key, fiber bundles with the same hash key are aggregated into the same hash bucket. Each hash bucket corresponds to one optical cable segment unit. The computing environment assigns a unique optical cable segment unit identifier to each optical cable segment unit, in string format. The computing environment iterates through all hash buckets, summing the fiber core capacity parameters of all fiber bundles within each hash bucket to obtain the aggregated fiber core capacity parameters of that optical cable segment unit, while retaining the spatial topology start and end point identifiers unchanged. All optical cable segment units form an optical cable segment unit set. The data structure of the optical cable segment unit set is an associative array with the optical cable segment unit identifier as the key and a structure containing the spatial topology start and end point identifiers and the aggregated fiber core capacity parameters as the values.

[0018] Step S113: Access the network element management system of the data center optical cable transmission architecture, extract the active transmission channel identifier currently carried by each optical cable segment unit and the fiber core number occupied by the corresponding transmission channel, and summarize them to form transmission channel occupancy distribution information.

[0019] Then, the computing environment initiates a session connection through the northbound interface provided by the network element management system, using the Simple Network Management Protocol (SMMP) or the Representational State Transmission (RSST) application programming interface to query the network element management system for the transmission channel configuration information of each network element. The query request carries the fiber optic segment unit identifier as a filtering condition. The response message returned by the network element management system contains a list of active transmission channel identifiers, each of which is associated with a list of occupied fiber core numbers. The list of occupied fiber core numbers is an integer array, where each integer represents the physical arrangement number of the fiber core occupied by that transmission channel within the fiber optic segment unit in a multi-fiber optical cable. The computing environment organizes the returned data according to the fiber optic segment unit identifier to form transmission channel occupancy distribution information. The data structure of the transmission channel occupancy distribution information is a nested mapping structure, where the key of the outer mapping is the fiber optic segment unit identifier, the key of the inner mapping is the active transmission channel identifier, and the value is the list of occupied fiber core numbers.

[0020] Step S114: Access the historical fault work order database of the data center optical cable transmission architecture, extract the timestamp and duration of optical cable interruption events that occurred in each optical cable segment unit within the preset backtracking time window, and statistically form historical transmission interruption event markers.

[0021] Next, the computing environment connects to the historical fault work order database and executes a structured query language query. The query conditions specify the fault type as fiber optic cable interruption, the affected fiber optic cable segment unit identifiers as various identifiers in the fiber optic cable segment unit set, and the fault occurrence timestamp within a preset backtracking time window. The preset backtracking time window is a time interval with the current system time as the endpoint and a preset backtracking duration as the span. The query result set contains the fault event identifier, fiber optic cable segment unit identifier, fault occurrence timestamp, and fault duration fields. The computing environment performs grouping and aggregation operations on the result set according to the fiber optic cable segment unit identifier, sorting all fault events corresponding to each fiber optic cable segment unit in ascending order by fault occurrence timestamp, forming a historical transmission interruption event marker for that fiber optic cable segment unit. The data structure of the historical transmission interruption event marker is an array of timestamps and a corresponding array of durations, both arrays being of equal length. Each timestamp and the duration at that index position constitute an interruption event record.

[0022] Step S115: Combine the spatial topology start and end point identifiers and fiber core capacity parameters into an initial physical attribute record, and combine the transmission channel occupancy distribution information and historical transmission interruption event markers into an initial transmission bearer status record.

[0023] Finally, the computing environment constructs an initial physical attribute record and an initial transmission bearer status record for each optical cable segment unit in the optical cable segment unit set. The initial physical attribute record is a structure containing a spatial topology start / endpoint identifier field and a fiber core capacity parameter field. The spatial topology start / endpoint identifier field is an array containing two string elements, and the fiber core capacity parameter field is a positive integer scalar. The initial transmission bearer status record is a structure containing a transmission channel occupancy distribution information field and a historical transmission interruption event flag field. The transmission channel occupancy distribution information field is a nested mapping structure, and the historical transmission interruption event flag field is a structure containing a timestamp array and a duration array. The computing environment associates the initial physical attribute record and the initial transmission bearer status record with the corresponding optical cable segment unit identifier and stores them in a main memory data structure for subsequent steps to retrieve.

[0024] Step S120: Construct a physical topology connectivity diagram of the data center optical cable transmission architecture based on the initial physical attribute records of each optical cable segment unit in the optical cable segment unit set, and generate a transmission health status degradation trend curve of the corresponding optical cable segment unit based on the initial transmission bearer status record of each optical cable segment unit. Based on the physical topology connectivity diagram and the transmission health status degradation trend curve, perform a transmission path elastic reconfigurability analysis on the data center optical cable transmission architecture to obtain the elastic reconfiguration applicability identifier of each optical cable segment unit in the data center optical cable transmission architecture.

[0025] Step S121: Abstract the spatial topology start and end point identifiers of each optical cable segment unit in the optical cable segment unit set into graph nodes, abstract the optical cable segment units into graph edges connecting the corresponding graph nodes, and assign the graph edges with the fiber core capacity parameter as edge weights to construct a physical topology connectivity graph.

[0026] In this embodiment, the computing environment initializes an empty undirected graph data structure. Each optical cable segment unit in the set of optical cable segment units is traversed, and the spatial topology start and end point identifiers in its initial physical attribute record are extracted. The start-end fiber distribution frame port identifier in the spatial topology start and end point identifiers is used as the first graph node identifier, and the end-end fiber distribution frame port identifier is used as the second graph node identifier. It is checked whether the first and second graph node identifiers already exist in the node set of the graph; if not, a new node is created and added to the node set. Then, an undirected edge is created connecting the first and second graph nodes, and the fiber core capacity parameter of the optical cable segment unit is extracted and assigned to the edge weight attribute of the edge. After traversal, a physical topology connectivity graph is obtained. The physical topology connectivity graph is an undirected weighted graph. The graph node set is a deduplicated set of all fiber distribution frame port identifiers in the data center optical cable transmission architecture, and the graph edge set is the undirected edges corresponding to all optical cable segment units, with each edge having a fiber core capacity parameter as its edge weight. The physical topology connectivity graph is stored in an adjacency list data structure. Each entry in the adjacency list corresponds to a graph node, storing the list of adjacent graph nodes of that graph node and the edge weights of the connecting edges.

[0027] Step S122: Extract the timestamp and duration of the optical cable interruption event from the historical transmission interruption event markers of each optical cable segment unit, arrange them in time sequence to form an interruption event sequence, fit the degradation trend of the interruption event sequence in the time dimension, and generate a transmission health status degradation trend curve.

[0028] Next, the computing environment extracts historical transmission interruption event markers from the initial transmission bearer status record of each optical cable segment unit in the optical cable segment unit set. These historical transmission interruption event markers contain an array of timestamps and an array of durations. The computing environment converts each timestamp in the timestamp array into a relative time value calculated from a preset reference time point, and uses the duration of each interruption event as the interruption severity value at that time point. The computing environment uses the relative time value as the independent variable and the interruption severity value as the dependent variable to construct an interruption event sequence, which is a two-dimensional point sequence. The computing environment performs local weighted regression fitting on the interruption event sequence. The kernel function for local weighted regression is a trigonometric kernel function, and the bandwidth parameter is the reciprocal of the square root of the number of points in the interruption event sequence. During the fitting process, at each independent variable value, a low-order polynomial is fitted using weighted least squares, with the weights determined by the kernel function based on the distance between the data point and the independent variable value. The fitted curve is the transmission health status degradation trend curve, which is a continuous function defined on the time axis. The function value is the trend value of the predicted severity of the interruption at that time point.

[0029] Step S123: Identify the inflection point of the transmission health status degradation trend curve, extract the critical time point of the degradation rate change in the transmission health status degradation trend curve, and divide the transmission health status degradation trend curve into multiple degradation stage intervals according to the critical time point.

[0030] Then, the computing environment calculates the second derivative function of the transmission health status degradation trend curve. Within the domain of the transmission health status degradation trend curve, the computing environment generates a sequence of discrete sampling points at a preset sampling interval, which is one-twentieth of the preset backtracking time window span. For each sampling point, the second derivative value of the transmission health status degradation trend curve at that point is calculated. The computing environment scans the second derivative value sequence, detecting consecutive sampling point pairs where the sign of the second derivative value changes. Within each sign change interval, linear interpolation is used to determine the critical time point where the second derivative value is exactly zero. All detected critical time points are arranged in ascending chronological order, dividing the domain of the transmission health status degradation trend curve into multiple degradation stage intervals. Each degradation stage interval corresponds to a degradation trend characteristic type, including accelerated degradation intervals, decelerated degradation intervals, and stable degradation intervals.

[0031] Step S124: Assign a stage degradation coefficient to each degradation stage interval, and use the stage degradation coefficient of the degradation stage interval at the current time point as the instantaneous degradation degree value of the corresponding optical cable segment unit.

[0032] Next, the computing environment performs feature analysis on each degradation stage interval. For each degradation stage interval, it extracts the sampling point sequence of the transmission health status degradation trend curve within that interval, and calculates the mean of the first derivative of the sampling point sequence. The mean of the first derivative is the stage degradation coefficient for that degradation stage interval. The stage degradation coefficient is a positive real number; a positive value indicates an increasing degradation trend, a negative value indicates a decreasing degradation trend, and zero indicates a stable degradation trend. The computing environment obtains the current system time, determines the degradation stage interval in which the current system time falls, and extracts the stage degradation coefficient for that degradation stage interval as the instantaneous degradation degree value for that optical cable segment unit. The instantaneous degradation degree value is a real number, and its range includes positive, negative, and zero.

[0033] Step S125: Using the instantaneous degradation degree value as the failure probability factor, perform failure simulation on the graph edge corresponding to each optical cable segment unit in the physical topology connectivity graph, and calculate the maximum change in the size of the remaining connected subgraph of the physical topology connectivity graph after removing the graph edge.

[0034] Then, the computing environment inputs the instantaneous degradation value into a preset failure probability mapping function, mapping the instantaneous degradation value to a failure probability. The failure probability mapping function is a logistic function, expressed as the failure probability equal to the sum of the unit value divided by the unit value and the result of an exponential function, where the exponent is the negative instantaneous degradation value multiplied by a scaling parameter. The scaling parameter is a preset positive real number. For each edge in the physical topology connectivity graph, the computing environment generates a random number uniformly distributed between zero and one. If this random number is less than the failure probability mapping result corresponding to that edge, the edge is marked as failed and temporarily removed from the physical topology connectivity graph. After simulating the failure of all edges, the computing environment performs a breadth-first search on the physical topology connectivity graph after removing failed edges, calculating the node size of all connected subgraphs, and taking the largest number of nodes in the largest connected subgraph as the maximum size of the connected subgraph after failure. The computing environment also calculates the maximum size of the largest connected subgraph in the physical topology connectivity graph without removing any edges, i.e., the total number of graph nodes. The change in the size of the maximum connected subgraph is equal to the difference between the size of the maximum connected subgraph before edge removal and the size of the maximum connected subgraph after failure. Each calculation performs an independent failure simulation for one graph edge, while the remaining graph edges remain in the unremoved state.

[0035] Step S126: Compare the change in the size of the largest connected subgraph with the preset connectivity impact threshold. When the change in the size of the largest connected subgraph exceeds the preset connectivity impact threshold, set the elastic reconfiguration applicability flag of the corresponding optical cable segment unit to the priority participation flag; otherwise, set it to the regular participation flag.

[0036] Finally, the computing environment obtains a preset connectivity impact threshold, which is a positive integer representing the maximum allowable connectivity loss tolerance. The computing environment compares the maximum change in the size of the connected subgraph corresponding to each graph edge with the preset connectivity impact threshold. If the change in the size of the maximum connected subgraph is greater than the preset connectivity impact threshold, the elastic reconfiguration suitability flag of the optical cable segment unit corresponding to that edge is set as a priority participation flag, which is a preset enumeration constant. If the change in the size of the maximum connected subgraph is less than or equal to the preset connectivity impact threshold, the elastic reconfiguration suitability flag of the optical cable segment unit is set as a regular participation flag, which is another preset enumeration constant. The computing environment stores the elastic reconfiguration suitability flag in association with the optical cable segment unit flag.

[0037] Step S130: Based on the elastic reconfiguration applicability identifier, select a subset of candidate elastic reconfiguration optical cable segment units from the set of optical cable segment units, and determine the available transmission resource reserve record for each candidate elastic reconfiguration optical cable segment unit in the subset of candidate elastic reconfiguration optical cable segment units. Using the physical topology connectivity graph as the constraint boundary, call the preset transmission path elastic reconfiguration strategy generation model to perform transmission path replanning on the subset of candidate elastic reconfiguration optical cable segment units and the available transmission resource reserve record, and generate an elastic transmission path configuration scheme including the primary transmission path and the backup transmission path.

[0038] Step S131: Extract optical cable segment units with the elastic reconfiguration applicability identifier as the priority participation identifier, and form a subset of candidate elastic reconfiguration optical cable segment units.

[0039] In this embodiment, the computing environment iterates through the resilient reconfiguration suitability identifiers of all optical cable segment units in the optical cable segment unit set, and selects optical cable segment units whose resilient reconfiguration suitability identifiers are equal to the identifier enumeration constant. The selected optical cable segment units are added to the candidate resilient reconfiguration optical cable segment unit subset. The candidate resilient reconfiguration optical cable segment unit subset is a set of optical cable segment unit identifiers, and the data structure is a list or set type. The optical cable segment units included in this subset are those that have a significant impact on network connectivity in the failure simulation and are given priority in subsequent transmission path resilient reconfiguration planning.

[0040] Step S132: For each candidate resilient reconfiguration optical cable segment unit in the candidate resilient reconfiguration optical cable segment unit subset, extract the difference between its fiber core capacity parameter and the total number of occupied fiber core numbers in the transmission channel occupancy distribution information as the available transmission resource reserve record.

[0041] Next, for each candidate resilient reconfigurable optical cable segment unit in the subset of candidate reconfigurable optical cable segment units, the computing environment extracts the fiber core capacity parameter from the initial physical attribute record of the optical cable segment unit and extracts the transmission channel occupancy distribution information from the initial transmission carrying status record of the optical cable segment unit. The computing environment traverses all active transmission channel identifiers in the transmission channel occupancy distribution information. For each active transmission channel identifier, it obtains its occupied fiber core sequence number list, and takes the size of the occupied fiber core sequence number list as the number of fiber cores occupied by that transmission channel. The computing environment sums up the number of fiber cores occupied by all active transmission channels to obtain the total number of occupied fiber core sequence numbers. Then, the computing environment subtracts the total number of occupied fiber core sequence numbers from the fiber core capacity parameter; the difference is the available transmission resource reserve. The available transmission resource reserve represents the number of idle fiber cores in the optical cable segment unit that are not yet occupied by any transmission channel. The computing environment associates the available transmission resource reserve with the candidate resilient reconfigurable optical cable segment unit identifier to form an available transmission resource reserve record. The available transmission resource reserve record is a mapping entry with the optical cable segment unit identifier as the key and the available transmission resource reserve as the value.

[0042] Step S133: Obtain the sequence of optical cable segment units traversed by the current primary transmission path in the data center optical cable transmission architecture and the start and end node identifiers of the transmission service to be optimized, and map the start and end node identifiers to the corresponding start and end graph nodes in the physical topology connectivity graph.

[0043] Then, the computing environment accesses the network element management system of the data center's optical fiber transmission architecture to query the path information of the current primary transmission path. The query request carries the service identifier of the transmission service to be optimized as a filtering condition. The network element management system returns a list of optical fiber segment units traversed by the current primary transmission path, arranged in order from the start to the end of the path, forming a sequence of optical fiber segment units for the current primary transmission path. Simultaneously, the computing environment extracts the start and end node identifiers of the transmission service to be optimized from the service configuration database. The start and end node identifiers are the start-end fiber optic distribution frame port identifier and the end-end fiber optic distribution frame port identifier, respectively. The computing environment retrieves the graph node corresponding to the start-end fiber optic distribution frame port identifier from the graph node set of the physical topology connectivity graph as the start graph node, and retrieves the graph node corresponding to the end-end fiber optic distribution frame port identifier as the end graph node. If no corresponding graph node is found, an error is reported and the process exits.

[0044] Step S134: Input the candidate flexible reconfigurable optical cable segment unit subset, the available transmission resource margin record of each candidate flexible reconfigurable optical cable segment unit, the starting graph node and the ending graph node into the transmission path flexible reconfiguration strategy generation model. The transmission path flexible reconfiguration strategy generation model performs path search under the constraints of the physical topology connectivity graph and outputs a transmission path set composed of candidate flexible reconfigurable optical cable segment units.

[0045] Next, the computing environment invokes the transmission path resilient reconstruction strategy generation model. This model is a path planning model based on deep reinforcement learning, with an architecture consisting of a neural network containing a graph attention network encoder and a policy decoder. The graph attention network encoder is composed of multiple stacked graph attention layers. Each layer calculates the attention weights for each graph node in the physical topology connectivity graph, focusing on its proximity to neighboring nodes. The attention weights are calculated by first linearly transforming the feature vectors of the central node and its neighboring nodes, then concatenating the transformed feature vectors and performing a dot product with the attention parameter vector. After activation by a linear unit with leakage correction and passing through a normalized exponential function, the attention weights are obtained. Finally, the attention weights are used to weight and sum the linearly transformed feature vectors of the neighboring nodes, and the result is used as the updated feature vector for that node. The input to the graph attention network encoder is the adjacency list of the physical topology connectivity graph and the initial feature vectors of each graph node. The initial feature vectors of each node are composed of the encoded port type and port level of the fiber optic patch panel port corresponding to that node, concatenated together.

[0046] The policy decoder is a state decoding network based on gated recurrent units. At each path search step, the policy decoder receives the feature vector output from the graph attention network encoder of the current graph node and the encoded vector of the historical path as input, and outputs a transition probability distribution from the current graph node to its neighboring graph nodes. The encoded vector of the historical path is obtained by the gated recurrent unit encoding the sequence of visited graph nodes. Each component of the transition probability distribution corresponds to a neighboring graph node of the current graph node, and the value of the component represents the probability of selecting that neighboring graph node as the next step.

[0047] The transmission path resilient reconfiguration strategy generation model employs a near-end policy optimization algorithm for parameter updates during the training phase. Training data consists of historical transmission path configuration records of the data center fiber optic transmission architecture and simulated link failure scenarios. The reward function during training is set as follows: a positive reward is given for successfully finding a reachable path; an additional positive reward is given if the path contains candidate resilient reconfigurable fiber optic segment units; the fewer graph nodes the path traverses, the higher the positive reward; and a negative penalty is given when selecting a fiber optic segment unit with zero available transmission resource margin.

[0048] During the inference phase, the computing environment inputs the adjacency list structure of the physical topology connectivity graph, the graph edge identifier list corresponding to the candidate flexible reconfigurable optical cable segment unit subsets, the available transmission resource margin record for each candidate flexible reconfigurable optical cable segment unit, the starting graph node identifier, and the ending graph node identifier into the transmission path flexible reconfiguration strategy generation model. Starting from the starting graph node, the model samples the next graph node at each step based on the transition probability distribution output by the policy decoder. If the selected graph edge does not belong to a candidate flexible reconfigurable optical cable segment unit subset, it is allowed to pass with a preset low probability. If the available transmission resource margin of the selected graph edge is zero, it backtracks one step and resamples. When the ending graph node is reached, the path search is complete. The model generates multiple transmission paths through multiple Monte Carlo samplings, forming a transmission path set. Each transmission path in the transmission path set is a sequence of graph nodes, and adjacent graph nodes in the sequence are connected by graph edges in the physical topology connectivity graph.

[0049] Step S135: Select the transmission path with the fewest shared optical cable segment units with the current primary transmission path from the transmission path set as the backup transmission path, and combine the current primary transmission path and the backup transmission path into a flexible transmission path configuration scheme.

[0050] Finally, the computing environment converts the current primary transmission path's optical cable segment unit sequence into a corresponding graph edge set. For each transmission path in the transmission path set, the intersection size of the graph edge set of that transmission path and the graph edge set of the current primary transmission path is calculated; this intersection size represents the number of shared optical cable segment units. The computing environment selects the transmission path with the smallest number of shared optical cable segment units as the backup transmission path. If multiple transmission paths have the same smallest number of shared optical cable segment units, the transmission path with the largest sum of total fiber core capacity parameters is selected as the backup transmission path. The computing environment combines the current primary transmission path and the selected backup transmission path into a flexible transmission path configuration scheme. The flexible transmission path configuration scheme is a structure containing a primary transmission path field and a backup path field, with each path field storing the optical cable segment unit sequence.

[0051] Step S140: When a transmission link degradation trigger signal is detected, the path switching node sequence and path switching timing arrangement information of the backup transmission path in the elastic transmission path configuration scheme are parsed, and the transmission carrying traffic is migrated from the primary transmission path to the backup transmission path segment by segment according to the path switching node sequence and path switching timing arrangement information.

[0052] Step S141: Receive the transmission link degradation trigger signal sent by the performance monitoring platform of the data center optical cable transmission architecture. The transmission link degradation trigger signal carries the identifier of the target optical cable segment unit where degradation has occurred and the current bit error rate parameter of the target optical cable segment unit.

[0053] In this embodiment, the computing environment listens to the event message queue from the performance monitoring platform. The performance monitoring platform continuously monitors the transmission performance indicators of each optical cable segment unit. When the performance monitoring platform detects that the transmission performance degradation of a certain optical cable segment unit reaches the trigger condition, the performance monitoring platform generates a transmission link degradation trigger signal and pushes it to the computing environment through the message queue. The transmission link degradation trigger signal is a structured message object, and the message body contains a trigger type field, a target optical cable segment unit identifier field, and a current bit error rate parameter field. The trigger type field has a value of "link degradation", the target optical cable segment unit identifier field is a string indicating the optical cable segment unit where performance degradation has occurred, and the current bit error rate parameter field is a positive real number representing the bit error rate measured by the optical cable segment unit in the most recent monitoring period.

[0054] Step S142: Determine the primary transmission path and backup transmission path according to the elastic transmission path configuration scheme, and extract the starting fork node where the primary transmission path and backup transmission path begin to fork and the ending rejoining node where they rejoin.

[0055] Next, the computing environment reads the primary transmission path field and the backup transmission path field from the flexible transmission path configuration scheme. The computing environment converts the optical cable segment unit sequence of the primary transmission path into the corresponding graph node sequence, and the optical cable segment unit sequence of the backup transmission path into the corresponding graph node sequence. Starting from the beginning of the two graph node sequences, the computing environment compares the graph node identifiers hop-by-hop, finding the first graph node pair with a different identifier; the preceding graph node of this pair is the starting fork node. Starting from the end of the two graph node sequences, the computing environment compares the graph node identifiers hop-by-hop in reverse, finding the first graph node pair with a different identifier from the end; the following graph node of this pair is the ending merge node. Both the starting fork node and the ending merge node are fiber optic patch panel port identifiers.

[0056] Step S143: Traverse the backup transmission path hop by hop from the starting fork node to the ending rendezvous node, extract the port identifier of each optical cable segment unit traversed, and arrange them in the traversal order to form a path switching node sequence.

[0057] Then, the computational environment traverses the graph node sequence along the alternative transmission path, starting from the next graph node after the initial fork node and moving hop-by-hop towards the final merge node. For each traversed graph node, its corresponding fiber optic patch panel port identifier is extracted and added to the path switching node sequence. When the final merge node is reached, it is not added to the sequence because it is the re-merging point of two paths and does not require a switching operation. After the traversal is complete, a path switching node sequence is formed, which is an ordered list of fiber optic patch panel port identifiers, with the order in the list corresponding to the execution order of the switching operations.

[0058] Step S144: Configure a switching trigger time for each path switching node in the path switching node sequence. The switching trigger time is the system time after the previous path switching node completes traffic migration plus the preset fiber link stabilization waiting time, forming path switching timing arrangement information.

[0059] Next, the computational environment orchestrates the handover trigger time for each path switching node in the path switching node sequence. The path switching timing orchestration information is an ordered list of the same length as the path switching node sequence, with each element representing a time offset. The handover trigger time for the first path switching node is set to the current system time plus a preset handover preparation delay. Starting from the second path switching node, the handover trigger time for each path switching node is calculated as follows: the handover trigger time of this path switching node equals the handover trigger time of the previous path switching node plus a preset single-hop handover operation time plus a preset fiber link stabilization waiting time. The preset single-hop handover operation time is a positive real number, representing the average time required to complete a one-hop traffic migration operation. The preset fiber link stabilization waiting time is a positive real number, representing the time margin required for optical power stabilization after fiber link handover.

[0060] Step S145: Starting from the initial fork node, according to the path switching node sequence and path switching timing arrangement information, sequentially switch the transmission channel carried by the corresponding optical cable segment unit on the primary transmission path to the idle fiber core indicated by the available transmission resource reserve record of the corresponding optical cable segment unit on the backup transmission path at each path switching node.

[0061] Finally, the computing environment initiates the handover execution process. The computing environment sets a timer based on the first time offset in the path handover timing arrangement information. When the timer expires, the computing environment sends a handover command to the network element corresponding to the first path handover node in the path handover node sequence. The handover command includes the source transmission channel identifier, the destination optical cable segment unit identifier, and the destination fiber core number. The source transmission channel identifier comes from the transmission channel carried by that optical cable segment unit on the primary transmission path, and the destination optical cable segment unit identifier and destination fiber core number come from the available transmission resource reserve record of the corresponding optical cable segment unit on the backup transmission path, indicating the available fiber core. After receiving the handover command, the network element performs a fiber cross-connection reconfiguration operation on the transmission channel, switching the transmission channel from the primary path fiber core to the backup path fiber core. After the handover operation is completed, the network element returns a handover completion confirmation message to the computing environment. Upon receiving the confirmation message, the computing environment sets the next timer according to the path handover timing arrangement information and continues to execute the handover operation of the next path handover node. Once all path switching nodes have completed their switching operations, the transmission traffic is completely migrated from the primary transmission path to the backup transmission path, and the migration operation is completed segment by segment.

[0062] Step S150: This method also includes an additional step of generating a transmission link degradation trigger signal.

[0063] Step S210: Extract optical power attenuation sampling data and receiver bit error rate sampling data of each optical cable segment unit within a preset monitoring period from the transmission performance log of the data center optical cable transmission architecture.

[0064] In this embodiment, the computing environment periodically pulls data from the transmission performance logs of the performance monitoring platform at preset monitoring intervals. The transmission performance logs are structured log files, with each log record containing a timestamp field, an optical cable segment unit identifier field, an optical power attenuation sample value field, and a receiver bit error rate (BER) sample value field. The optical power attenuation sample value field is a positive real number representing the difference between the transmitting and receiving optical power, expressed in decibels and milliwatts. The receiver BER sample value field is a positive real number representing the ratio of the number of erroneous bits detected by the receiver to the total number of received bits. The computing environment pulls all log records within the preset monitoring period, groups them by optical cable segment unit identifier, sorts the log records corresponding to each optical cable segment unit in ascending order by timestamp, and extracts the optical power attenuation sample value sequence and the receiver BER sample value sequence, respectively.

[0065] Step S220: Fit the optical power attenuation sampling data to an attenuation trend function of optical power attenuation over time, and discretize the bit error rate sampling data at the receiving end into a time series to generate a discrete time series of bit error rate.

[0066] Next, the computing environment performs polynomial regression fitting on the optical power attenuation sample value sequence for each optical cable segment unit. A quadratic polynomial is used as the fitting model, with time as the independent variable and the optical power attenuation sample value as the dependent variable. The computing environment solves for the polynomial coefficients using the least squares method to minimize the sum of squares of the fitting residuals. The resulting quadratic polynomial is the attenuation trend function, which is a smooth, continuous function over time. Simultaneously, the computing environment groups each sample value in the receiver's bit error rate (BER) sample value sequence and its corresponding timestamp into a data point. All data points are arranged in ascending time order to form a discrete-time BER series. The BER discrete-time series is a time series data structure, where each element is a timestamp and a BER value pair.

[0067] Step S230: Analyze the slope of the decay trend function within a unit time interval. When the slope of the decay change exceeds the preset decay acceleration threshold for multiple consecutive unit time intervals, generate a pre-deterioration warning mark.

[0068] Then, the computing environment calculates the first derivative of the attenuation trend function to obtain the attenuation rate function. The value of the attenuation rate function at any time point represents the instantaneous rate of change of optical power attenuation at that time point. The computing environment discretely samples the attenuation rate function with a unit time interval as the step size, calculating the attenuation rate function value at each sampling point. The unit time interval is one-tenth of the preset monitoring period. The computing environment scans the sequence of attenuation rate function values ​​and counts the number of unit time intervals that consecutively exceed the preset attenuation acceleration threshold. The preset attenuation acceleration threshold is a positive real number threshold. When the number of unit time intervals that consecutively exceed the preset attenuation acceleration threshold reaches the preset consecutive exceedance threshold, the computing environment generates a pre-deterioration warning mark for that optical cable segment unit. The pre-deterioration warning mark is a structured record containing the optical cable segment unit identifier, warning type, and warning generation timestamp.

[0069] Step S240: Analyze the bit error rate jump amplitude between adjacent time points in the discrete time series of bit error rate. When the bit error rate jump amplitude exceeds the preset jump tolerance limit, generate a burst degradation alarm flag.

[0070] Next, the computing environment iterates through each element in the discrete-time series of bit error rates (BER), calculating the BER jump magnitude for the k-th element and its preceding element. The BER jump magnitude is equal to the BER value of the k-th element minus the absolute value of the BER value of the element k minus one. The computing environment compares the BER jump magnitude with a preset jump tolerance upper limit. The preset jump tolerance upper limit is a positive real number threshold. If the BER jump magnitude exceeds the preset jump tolerance upper limit, the computing environment generates a burst degradation alarm flag for that optical cable segment unit. The burst degradation alarm flag is a structured record containing the optical cable segment unit identifier, alarm type, and alarm generation timestamp.

[0071] Step S250: When both pre-degradation warning flag and sudden degradation alarm flag exist in the same optical cable segment unit, a transmission link degradation trigger signal is triggered for that optical cable segment unit. The transmission link degradation trigger signal carries the optical cable segment unit identifier and the current bit error rate parameter of that optical cable segment unit.

[0072] Finally, the computing environment checks the marking status of each optical cable segment unit. If an optical cable segment unit generates both a pre-degradation warning mark and a sudden degradation alarm mark within the same preset monitoring period, the computing environment comprehensively determines that a link degradation event has occurred in that optical cable segment unit. The computing environment generates a transmission link degradation trigger signal, which includes a trigger type field, a target optical cable segment unit identifier field, and a current bit error rate parameter field. The target optical cable segment unit identifier field is taken from the optical cable segment unit identifier of that optical cable segment unit, and the current bit error rate parameter field is taken from the latest bit error rate sample value in the discrete time series of bit error rates. The computing environment pushes the transmission link degradation trigger signal to the event message queue monitored in step S141.

[0073] Step S160: This method also includes a common cause failure risk assessment and reselection step for the resilient transmission path configuration scheme.

[0074] Step S310: Extract the primary optical cable segment unit sequence corresponding to the primary transmission path and the backup optical cable segment unit sequence corresponding to the backup transmission path in the flexible transmission path configuration scheme.

[0075] In this embodiment, after the flexible transmission path configuration scheme is generated, or at a preset periodic evaluation point, the computing environment extracts the primary transmission path field and the backup transmission path field from the flexible transmission path configuration scheme. The primary transmission path field is a list of optical cable segment unit identifiers, and the backup transmission path field is a list of optical cable segment unit identifiers. The computing environment stores them as a primary optical cable segment unit sequence and a backup optical cable segment unit sequence, respectively.

[0076] Step S320: Identify shared optical cable segments in the primary optical cable segment unit sequence and the backup optical cable segment unit sequence, extract optical cable segment units that appear in both sequences as shared risk optical cable segment units, and treat optical cable segment units that appear in only one sequence as independent optical cable segment units.

[0077] Next, the computing environment converts the primary optical cable segment unit sequence into a first set and the spare optical cable segment unit sequence into a second set. The computing environment calculates the intersection of the first and second sets; the optical cable segment units in the intersection result are the shared risk optical cable segment units. The computing environment calculates the union of the first and second sets; the optical cable segment units obtained by subtracting the intersection result from the union result are the independent optical cable segment units.

[0078] Step S330: Extract the historical transmission interruption event markers of the shared risk optical cable segment unit, and calculate the number of common cause failure events in the shared risk optical cable segment unit that simultaneously cause the primary transmission path and the backup transmission path to become unusable in the historical transmission interruption events.

[0079] Then, for each shared-risk optical cable segment unit, the computing environment extracts historical transmission interruption event markers from the initial transmission bearer state record of that segment unit. These historical transmission interruption event markers contain a timestamp array. The computing environment iterates through each timestamp in the timestamp array, checking whether the primary transmission path and the backup transmission path are simultaneously unavailable at the corresponding time point. The determination is based on the following: if the shared-risk optical cable segment unit experiences an interruption at that timestamp, and there are no other redundant paths besides that shared-risk optical cable segment unit, then both the primary and backup transmission paths are determined to be simultaneously unavailable. The computing environment counts the number of events that cause the primary and backup transmission paths to be simultaneously unavailable across all historical interruption event timestamps for all shared-risk optical cable segment units, summing them to obtain the total number of common-cause failure events.

[0080] Step S340: Calculate the proportion of common cause failure events in the total number of historical transmission interruption events. When the proportion of common cause failure events exceeds the preset risk tolerance limit, reselect the path for the elastic transmission path configuration scheme.

[0081] Next, the computing environment counts the total number of historical transmission interruption events for all shared-risk optical cable segment units, i.e., the total number of historical transmission interruption events. The computing environment divides the number of common-cause failure events by the total number of historical transmission interruption events to obtain the common-cause failure percentage. The common-cause failure percentage is a real number between zero and one. The computing environment compares the common-cause failure percentage with a preset risk tolerance upper limit, which is a preset real-valued threshold. If the common-cause failure percentage exceeds the preset risk tolerance upper limit, the computing environment determines that the common-cause failure risk of the current flexible transmission path configuration scheme is too high, and path reselection is required for the flexible transmission path configuration scheme.

[0082] Step S350: During the path reselection process, with the goal of minimizing the number of shared optical cable segment units between the new backup transmission path and the primary transmission path, the preset transmission path elastic reconstruction strategy is invoked again to generate the updated backup transmission path, and the updated backup transmission path is used to replace the backup transmission path in the original elastic transmission path configuration scheme.

[0083] Finally, the computing environment re-invokes the transmission path resilient reconfiguration strategy to generate the model. During this revocation, the penalty weight for the number of shared fiber optic segment units in the model's reward function is increased by a preset multiple, making the model more inclined to select transmission paths that share fewer fiber optic segment units with the primary transmission path during path search. The computing environment inputs the same physical topology connectivity graph, candidate resilient reconfiguration fiber optic segment unit subsets, available transmission resource reserves, start graph nodes, and end graph nodes into the model, and the model outputs a new alternative transmission path. The computing environment replaces the alternative transmission path field in the original resilient transmission path configuration scheme with the newly generated alternative transmission path, completing the update of the resilient transmission path configuration scheme.

[0084] Step S170: This method also includes a post-migration update step for the transmission channel occupancy distribution.

[0085] Step S410: After the segmented migration operation of the transmission carrying traffic from the primary transmission path to the backup transmission path is completed, the transmission channel occupancy distribution information of each optical cable segment unit is parsed and updated to form the transmission channel occupancy distribution information after migration.

[0086] Step S411: Extract the released optical cable segment units and released fiber core numbers that no longer carry transmission traffic in the primary transmission path, and mark the released fiber core numbers as idle transmission resources.

[0087] In this embodiment, after the segment-by-segment migration operation in step S145 is completed, the computing environment traverses each optical cable segment unit in the primary transmission path optical cable segment unit sequence. For each optical cable segment unit, the list of transmission channel identifiers carried by that optical cable segment unit is searched from the transmission channel occupancy distribution information before migration. For each transmission channel identifier, it is checked whether it is still allocated to that optical cable segment unit. If the transmission channel has been migrated to the backup transmission path, the fiber core number originally occupied by the transmission channel on that optical cable segment unit is the released fiber core number. The computing environment marks the released fiber core number as an idle transmission resource. The idle transmission resource is a record structure containing the optical cable segment unit identifier and the list of idle fiber core numbers.

[0088] Step S412: Extract the occupied optical cable segment unit and occupied fiber core number of the newly added transmission traffic in the backup transmission path, and mark the occupied fiber core number as an active transmission resource.

[0089] Next, the computing environment traverses each optical cable segment unit in the sequence of backup transmission path optical cable segment units. For each optical cable segment unit, it queries the list of transmission channel identifiers carried by that optical cable segment unit from the migrated network element management system. For each transmission channel identifier, it obtains the list of fiber core numbers it occupies. Fiber core numbers that did not appear in the transmission channel occupancy distribution information of that optical cable segment unit before migration are extracted as occupied fiber core numbers. The computing environment marks the occupied fiber core numbers as active transmission resources. An active transmission resource is a record structure containing the optical cable segment unit identifier and the list of active fiber core numbers.

[0090] Step S413: Summarize idle and active transmission resources, update the transmission channel occupancy distribution information of the corresponding optical cable segment unit in the data center optical cable transmission architecture, and form the transmission channel occupancy distribution information after migration.

[0091] Then, the computing environment updates the transmission channel occupancy distribution information for each affected optical cable segment unit based on the records of idle and active transmission resources. The update operation is as follows: in the transmission channel occupancy distribution information of the optical cable segment unit, the transmission channel occupancy record entry corresponding to the released fiber core sequence number is removed, and the transmission channel occupancy record entry corresponding to the occupied fiber core sequence number is added. After the update is completed, the migrated transmission channel occupancy distribution information is obtained. The data structure of the migrated transmission channel occupancy distribution information is the same as that of the transmission channel occupancy distribution information in step S113.

[0092] Step S414: Synchronize the migration transmission channel occupancy distribution information to the resource management database of the data center optical cable transmission architecture, and replace the original transmission channel occupancy distribution information of the corresponding optical cable segment unit.

[0093] Finally, the computing environment establishes a connection with the resource management database of the data center fiber optic transmission architecture and initiates a data update transaction for each affected fiber optic segment unit. Within the transaction, the computing environment updates the transmission channel occupancy distribution information table in the resource management database, writing the corresponding fields of the migrated transmission channel occupancy distribution information to the database record and deleting the original transmission channel occupancy distribution information record. After the transaction is committed, the data in the resource management database remains consistent with the actual network state.

[0094] Step S180: This method also includes a step of adjusting the overall network transmission resource balance.

[0095] Step S510: Periodically collect the transmission channel occupancy distribution information of all optical cable segment units in the data center optical cable transmission architecture to form a snapshot of the network transmission resource occupancy status.

[0096] In this embodiment, the computing environment sets a scheduled task that is triggered at preset network-wide data collection intervals. When the scheduled task is triggered, the computing environment sends batch query requests for transmission channel occupancy distribution information to the network element management system. It iterates through all optical cable segment unit identifiers in the optical cable segment unit set to obtain the transmission channel occupancy distribution information for each optical cable segment unit. The computing environment summarizes the transmission channel occupancy distribution information of all optical cable segment units to form a snapshot of the network-wide transmission resource occupancy status. The snapshot of the network-wide transmission resource occupancy status is a hash mapping structure, where the key is the optical cable segment unit identifier and the value is the corresponding transmission channel occupancy distribution information.

[0097] Step S520: Perform a fragmentation analysis on the snapshot of the network's transmission resource occupancy status to identify scattered idle fiber segments with fiber core utilization rates below a preset fragmentation threshold and congested fiber segments with fiber core utilization rates above a preset overload threshold.

[0098] Next, the computing environment calculates the fiber core utilization rate for each optical cable segment unit in the snapshot of the network-wide transmission resource occupancy status. The fiber core utilization rate is calculated as follows: the total number of occupied fiber cores is counted from the transmission channel occupancy distribution information, and then divided by the fiber core capacity parameter of the optical cable segment unit. The computing environment marks optical cable segment units with a fiber core utilization rate below a preset fragmentation threshold as scattered idle fiber core segments, and optical cable segment units with a fiber core utilization rate above a preset overload threshold as congested fiber core segments. The preset fragmentation threshold is a positive real number less than the preset overload threshold.

[0099] Step S530: Generate a transmission resource balancing adjustment instruction. The transmission resource balancing adjustment instruction includes a list of source optical cable segment unit identifiers to be adjusted, destination optical cable segment unit identifiers, and transmission channel identifiers to be migrated.

[0100] Then, the computing environment executes a greedy migration matching algorithm. For each congested fiber segment, it is sorted in descending order of fiber core utilization as source optical cable segment units. For each source optical cable segment unit, the optical cable segment unit with the lowest fiber core utilization and a greater than zero available transmission resource margin is selected from the set of scattered idle fiber segments as the destination optical cable segment unit. The computing environment selects the transmission channel identifier with the most occupied fiber core numbers from the transmission channel occupancy distribution information of the source optical cable segment units as the transmission channel to be migrated. The computing environment generates a transmission resource balancing adjustment instruction, which is a data structure containing a source optical cable segment unit identifier field, a destination optical cable segment unit identifier field, and a list of transmission channel identifiers to be migrated field.

[0101] Step S540: According to the transmission resource balancing adjustment instruction, some transmission channels in the congested fiber chip segment are migrated to scattered idle fiber chip segments to achieve the redistribution of transmission channel occupancy.

[0102] Finally, the computing environment sends a transmission resource balancing adjustment command to the network element management system. Upon receiving the command, the network element management system performs a path rerouting operation on the transmission channel corresponding to the transmission channel identifier to be migrated, switching the carrying fiber of that transmission channel from the occupied fiber core of the source optical cable segment unit to the idle fiber core of the destination optical cable segment unit. After the operation is completed, the network element management system returns a successful operation confirmation message to the computing environment. Upon receiving the confirmation message, the computing environment updates the snapshot of the entire network's transmission resource occupancy status and synchronizes it to the resource management database.

[0103] Step S190: This method also includes a core fragment sorting and scheduling step.

[0104] For example, step S610: Obtain the fiber core capacity parameters, occupied fiber core sequence number and transmission channel occupancy distribution information of each optical cable segment unit in the data center optical cable transmission architecture, and construct a fiber core-level transmission resource occupancy grid.

[0105] In this embodiment, the computing environment extracts fiber core capacity parameters and current transmission channel occupancy distribution information for each optical cable segment unit. The transmission channel occupancy distribution information includes the serial numbers of all occupied fiber cores. The computing environment creates a one-dimensional Boolean array with a length equal to the fiber core capacity parameter and an array index representing the fiber core serial number. During initialization, all array elements are set to false, and then all occupied fiber core serial numbers are traversed, with the array element at the corresponding index set to true. This one-dimensional Boolean array is the fiber-level transmission resource occupancy grid; each element being true indicates that the fiber core is occupied, and false indicates that the fiber core is idle.

[0106] Step S620: Perform spatial continuity analysis on the fiber core-level transmission resource occupancy grid to identify fiber core island regions formed by unoccupied fibers surrounded by occupied fibers and fiber core fragment regions formed by the discontinuous distribution of occupied fibers along the fiber core number axis.

[0107] Next, the computing environment scans the fiber-level transmission resource occupancy grid. The scanning method is to traverse the one-dimensional array from left to right. The computing environment maintains a state machine for consecutive occupied segments and consecutive free segments. When an unoccupied element is found to be flanked by occupied elements, the unoccupied element and its largest consecutive unoccupied segment constitute a fiber-core island region. When a single or a few unoccupied elements exist between consecutive occupied segments, the discontinuity between the consecutive occupied segments constitutes a fiber-core fragment region. Both fiber-core island regions and fiber-core fragment regions are recorded as structures containing a starting fiber core number and an ending fiber core number.

[0108] Step S630: Generate a fiber core fragment sorting and scheduling scheme, which includes the source fiber core location to be sorted, the destination fiber core location, and the sorting execution time window.

[0109] Then, the computing environment generates a fiber fragmentation reorganization scheduling scheme. For each fiber fragmentation area, the occupied fiber cores are designated as the source fiber core locations. For each fiber island area, the unoccupied fiber cores are designated as the destination fiber core locations. The reorganization execution time window is set to a low-traffic period, determined by querying historical transmission traffic statistics. The fiber fragmentation reorganization scheduling scheme is a list, where each element contains three fields: source fiber core location, destination fiber core location, and reorganization execution time window.

[0110] Step S640: Within the time window for organizing, the transmission channels in the fiber core fragment area and the fiber core island area are migrated to the continuous fiber core area to form an optimized fiber core occupancy layout with continuous fiber core occupancy and concentrated unoccupied fiber cores.

[0111] Finally, at the start of the cleanup execution window, the computing environment sends fiber core migration instructions to the network element management system one by one. Each migration instruction specifies the migration of the transmission channel carried at the source fiber core location to the destination fiber core location. The network element management system completes the migration through fiber cross-connection operations. The computing environment executes all migration instructions sequentially according to the time order of the cleanup execution window. After all migrations are completed, the fiber core occupancy status of the optical cable segment unit is redistributed, with occupied fibers clustered in one continuous interval and unoccupied fibers concentrated in another continuous interval, forming an optimized fiber core occupancy layout.

[0112] In some embodiments, the intelligent optimization system for optical fiber transmission in data centers used to perform the above-described methods can be any electronic device with data computing, processing, and storage capabilities. This intelligent optimization system for optical fiber transmission in data centers can be used to implement the intelligent optimization method for optical fiber transmission in data centers provided in the above embodiments.

[0113] Typically, intelligent optimization systems for fiber optic transmission in data centers include a processor and memory. The processor may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor can be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field Programmable Gate Array), or PLA (Programmable Logic Array). The processor may also include a main processor and coprocessors. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor may also include an AI (Artificial Intelligence) processor, which handles computational operations related to machine learning.

[0114] The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory is used to store a computer program configured to be executed by one or more processors to implement the above-described intelligent optimization method for fiber optic transmission in data centers.

[0115] In an illustrative embodiment, a computer-readable storage medium is also provided, wherein a computer program is stored in the storage medium, and the computer program, when executed by a processor of a computer device, implements the above-described intelligent optimization method for optical cable transmission in data centers. Optionally, the computer-readable storage medium may be ROM (Read-Only Memory), RAM (Random Access Memory), CD-ROM (CompactDisc Read-Only Memory), magnetic tape, floppy disk, and optical data storage device, etc.

[0116] This application provides a computer program product, which includes computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, the processor will execute the intelligent optimization method for optical cable transmission in data centers provided in this application.

[0117] This application provides a computer-readable storage medium storing computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by a processor, the processor will execute the intelligent optimization method for optical cable transmission in data centers provided in this application.

[0118] In some embodiments, the computer-readable storage medium may be a read-only memory (ROM), random access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic surface memory, optical disk, or CD-ROM, etc.; or it may be a device that includes one or any combination of the above-mentioned memories.

[0119] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

[0120] As an example, computer-executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., files that store one or more modules, subroutines, or code sections).

[0121] As an example, computer-executable instructions can be deployed to execute on a single electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed across multiple locations and interconnected via a communication network.

[0122] Finally, it should be noted that the above-disclosed embodiments are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart optimization method for optical cable transmission in data centers, characterized in that, The method includes: The optical cable transmission architecture of the data center is analyzed at the optical cable segment level to obtain the set of optical cable segment units contained in the optical cable transmission architecture of the data center. The initial physical attribute record and the initial transmission carrying status record of each optical cable segment unit are extracted. The initial physical attribute record includes the spatial topology start and end point identifiers of the optical cable segment unit and the fiber core capacity parameters of the optical cable segment unit. The initial transmission carrying status record includes the current transmission channel occupancy distribution information of the optical cable segment unit and the historical transmission interruption event markers of the optical cable segment unit. Based on the initial physical attribute records of each optical cable segment unit in the optical cable segment unit set, a physical topology connectivity diagram of the data center optical cable transmission architecture is constructed. Based on the initial transmission carrying status record of each optical cable segment unit, a transmission health status degradation trend curve of the corresponding optical cable segment unit is generated. Based on the physical topology connectivity diagram and the transmission health status degradation trend curve, a transmission path elastic reconfigurability analysis is performed on the data center optical cable transmission architecture to obtain the elastic reconfiguration applicability identifier of each optical cable segment unit in the data center optical cable transmission architecture. Based on the elastic reconfiguration applicability identifier, a subset of candidate elastic reconfiguration optical cable segment units is selected from the set of optical cable segment units, and the available transmission resource margin record of each candidate elastic reconfiguration optical cable segment unit in the subset of candidate elastic reconfiguration optical cable segment units is determined. Using the physical topology connectivity graph as the constraint boundary, a preset transmission path elastic reconfiguration strategy generation model is called to perform transmission path replanning on the subset of candidate elastic reconfiguration optical cable segment units and the available transmission resource margin record, generating an elastic transmission path configuration scheme that includes a primary transmission path and a backup transmission path. When a transmission link degradation trigger signal is detected, the path switching node sequence and path switching timing arrangement information of the backup transmission path in the elastic transmission path configuration scheme are parsed, and the transmission carrying traffic is migrated from the primary transmission path to the backup transmission path segment by segment according to the path switching node sequence and the path switching timing arrangement information.

2. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, The process involves performing fiber optic segment granularity analysis on the data center fiber optic transmission architecture to obtain the set of fiber optic segment units contained in the data center fiber optic transmission architecture, and extracting the initial physical attribute record and initial transmission bearer status record for each fiber optic segment unit, including: The fiber optic patch panel port mapping table and fiber optic cable routing diagram information of the data center optical cable transmission architecture are analyzed. The fiber optic patch panel port identifiers at both ends of the optical cable segment are extracted as spatial topology start and end point identifiers, and the total number of fiber cores contained in the optical cable segment is extracted as the fiber core capacity parameter. Fiber bundles with the same spatial topology start and end point identifiers are aggregated into a fiber segment unit. All fiber bundles in the data center optical cable transmission architecture are traversed to form the set of fiber segment units. Access the network element management system of the optical cable transmission architecture of the data center, extract the active transmission channel identifier currently carried by each optical cable segment unit and the fiber core number occupied by the corresponding transmission channel, and summarize them to form the transmission channel occupancy distribution information; Access the historical fault work order database of the data center optical cable transmission architecture, extract the timestamp and duration of optical cable interruption events that occurred in each optical cable segment unit within a preset backtracking time window, and statistically form the historical transmission interruption event markers. The spatial topology start and end point identifiers and the fiber core capacity parameters are combined to form the initial physical attribute record, and the transmission channel occupancy distribution information and the historical transmission interruption event markers are combined to form the initial transmission bearer status record.

3. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, The process involves constructing a physical topology connectivity diagram of the data center optical cable transmission architecture based on the initial physical attribute records of each optical cable segment unit in the optical cable segment unit set, generating a transmission health status degradation trend curve for each optical cable segment unit based on its initial transmission bearer status record, and performing a transmission path elastic reconfigurability analysis on the data center optical cable transmission architecture based on the physical topology connectivity diagram and the transmission health status degradation trend curve. This yields an elastic reconfigurability applicability identifier for each optical cable segment unit in the data center optical cable transmission architecture, including: The spatial topology start and end point identifiers of each optical cable segment unit in the optical cable segment unit set are abstracted as graph nodes, and the optical cable segment units are abstracted as graph edges connecting the corresponding graph nodes. The graph edges are assigned fiber core capacity parameters as edge weights to construct the physical topology connectivity graph. Extract the timestamp and duration of the optical cable interruption event from the historical transmission interruption event markers of each optical cable segment unit, arrange them in time sequence to form an interruption event sequence, and fit the degradation trend of the interruption event sequence in the time dimension to generate the transmission health status degradation trend curve. The inflection point of the transmission health status degradation trend curve is identified, the critical time point of the degradation rate change in the transmission health status degradation trend curve is extracted, and the transmission health status degradation trend curve is divided into multiple degradation stage intervals according to the critical time point. Assign a stage degradation coefficient to each degradation stage interval, and use the stage degradation coefficient of the degradation stage interval at the current time point as the instantaneous degradation degree value of the corresponding optical cable segment unit. Using the instantaneous degradation degree value as the failure probability factor, failure simulation is performed on the graph edge corresponding to each optical cable segment unit in the physical topology connectivity graph, and the maximum connected subgraph size change of the remaining graph node in the physical topology connectivity graph after removing the graph edge is calculated. The maximum change in the size of the connected subgraph is compared with a preset connectivity impact threshold. When the change in the size of the maximum connected subgraph exceeds the preset connectivity impact threshold, the elastic reconfiguration applicability flag of the corresponding optical cable segment unit is set as the priority participation flag; otherwise, it is set as the regular participation flag.

4. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, The process involves selecting a subset of candidate reconfigurable optical cable segment units from the set of optical cable segment units based on the reconfigurability identifier, determining the available transmission resource reserve record for each candidate reconfigurable optical cable segment unit in the subset, using the physical topology connectivity graph as a constraint boundary, and invoking a preset transmission path reconfiguration strategy generation model to perform transmission path replanning on the candidate reconfigurable optical cable segment unit subset and the available transmission resource reserve record, generating a reconfigurable transmission path configuration scheme that includes a primary transmission path and a backup transmission path, including: Extract the optical cable segment units whose elastic reconfiguration applicability identifier is a priority participation identifier, and form the candidate elastic reconfiguration optical cable segment unit subset; For each candidate resilient reconfigurable optical cable segment unit in the candidate reconfigurable optical cable segment unit subset, the difference between its fiber core capacity parameter and the total number of occupied fiber core numbers in the transmission channel occupancy distribution information is extracted as the available transmission resource reserve record. Obtain the sequence of optical cable segment units traversed by the current primary transmission path in the data center optical cable transmission architecture and the start and end node identifiers of the transmission service to be optimized, and map the start and end node identifiers to the corresponding start and end graph nodes in the physical topology connectivity graph. The candidate flexible reconfigurable optical cable segment unit subset, the available transmission resource margin record of each candidate flexible reconfigurable optical cable segment unit, the starting graph node and the ending graph node are input into the transmission path flexible reconfiguration strategy generation model. The transmission path flexible reconfiguration strategy generation model performs path search under the constraints of the physical topology connectivity graph and outputs a transmission path set composed of candidate flexible reconfigurable optical cable segment units. The transmission path with the fewest shared optical cable segment units with the current primary transmission path is selected from the transmission path set as the backup transmission path, and the current primary transmission path and the backup transmission path are combined to form the flexible transmission path configuration scheme.

5. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, When a transmission link degradation trigger signal is detected, the method of parsing the path switching node sequence and path switching timing arrangement information of the backup transmission path in the resilient transmission path configuration scheme, and performing a segmented migration operation of the transmission bearer traffic from the primary transmission path to the backup transmission path according to the path switching node sequence and the path switching timing arrangement information, includes: The system receives a transmission link degradation trigger signal sent by the performance monitoring platform of the data center optical cable transmission architecture. The transmission link degradation trigger signal carries the identifier of the target optical cable segment unit where degradation has occurred and the current bit error rate parameter of the target optical cable segment unit. The primary transmission path and the backup transmission path are determined according to the elastic transmission path configuration scheme, and the starting fork node where the primary transmission path and the backup transmission path begin to fork and the ending rejoining node are extracted. The backup transmission path is traversed hop by hop from the starting fork node to the ending rendezvous node, and the port identifier of each optical cable segment unit traversed is extracted and arranged in traversal order to form the path switching node sequence. Configure a switching trigger time for each path switching node in the path switching node sequence. The switching trigger time is the system time after the previous path switching node completes traffic migration plus a preset fiber link stabilization waiting time, forming the path switching timing arrangement information. Starting from the initial fork node, according to the path switching node sequence and the path switching timing arrangement information, at each path switching node, the transmission channel carried by the corresponding optical cable segment unit on the primary transmission path is switched to the idle fiber core indicated by the available transmission resource reserve record of the corresponding optical cable segment unit on the backup transmission path.

6. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, The method further includes: Extract the optical power attenuation sampling data and the receiving end bit error rate sampling data of each optical cable segment unit within a preset monitoring period from the transmission performance log of the data center optical cable transmission architecture. The optical power attenuation sampling data is fitted to an attenuation trend function of optical power attenuation over time, and the bit error rate sampling data at the receiving end is discretized into a time series to generate a discrete time series of bit error rate. Analyze the slope of the decay trend function within a unit time interval. When the slope of the decay change exceeds a preset decay acceleration threshold for multiple consecutive unit time intervals, generate a pre-deterioration warning mark. Analyze the bit error rate jump amplitude between adjacent time points in the discrete time series of bit error rate. When the bit error rate jump amplitude exceeds the preset jump tolerance limit, generate a burst degradation alarm flag. When the same optical cable segment unit has both the pre-degradation warning mark and the sudden degradation alarm mark, a transmission link degradation trigger signal is triggered for that optical cable segment unit. The transmission link degradation trigger signal carries the optical cable segment unit identifier and the current bit error rate parameter of that optical cable segment unit.

7. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, The method further includes: Extract the primary optical cable segment unit sequence corresponding to the primary transmission path and the backup optical cable segment unit sequence corresponding to the backup transmission path in the elastic transmission path configuration scheme; The primary optical cable segment unit sequence and the backup optical cable segment unit sequence are used to identify shared optical cable segments. Optical cable segment units that appear in both sequences are extracted as shared risk optical cable segment units, and optical cable segment units that appear in only one sequence are selected as independent optical cable segment units. Extract the historical transmission interruption event markers of the shared risk optical cable segment unit, and calculate the number of common cause failure events in the historical transmission interruption events in which the shared risk optical cable segment unit simultaneously caused the primary transmission path and the backup transmission path to become unusable. The common cause failure events are counted as a percentage of the total number of historical transmission interruption events. When the percentage of common cause failures exceeds a preset risk tolerance limit, the elastic transmission path configuration scheme is reselected. During the path reselection process, with the goal of minimizing the number of shared optical cable segment units between the new backup transmission path and the primary transmission path, the preset transmission path elastic reconstruction strategy is invoked again to generate the model and output the updated backup transmission path. The updated backup transmission path replaces the backup transmission path in the original elastic transmission path configuration scheme.

8. The intelligent optimization method for optical cable transmission in data centers according to claim 1, characterized in that, The method further includes: After the segmented migration operation of the transmission carrying traffic from the primary transmission path to the backup transmission path is completed, the transmission channel occupancy distribution information of each optical cable segment unit is analyzed and updated to form the transmission channel occupancy distribution information after migration. Extract the released optical cable segment units and released fiber core numbers that no longer carry transmission traffic in the primary transmission path, and mark the released fiber core numbers as idle transmission resources; Extract the occupied optical cable segment units and occupied fiber core numbers of the newly added transmission traffic in the backup transmission path, and mark the occupied fiber core numbers as active transmission resources; The idle transmission resources and the active transmission resources are aggregated, and the transmission channel occupancy distribution information of the corresponding optical cable segment unit in the data center optical cable transmission architecture is updated to form the transmission channel occupancy distribution information after migration. The migrated transmission channel occupancy distribution information is synchronized to the resource management database of the data center optical cable transmission architecture, replacing the original transmission channel occupancy distribution information of the corresponding optical cable segment unit.

9. The intelligent optimization method for optical cable transmission in data centers according to claim 8, characterized in that, The method further includes: Periodically collect the transmission channel occupancy distribution information of all optical cable segment units in the data center optical cable transmission architecture to form a snapshot of the network transmission resource occupancy status; Perform a transmission resource fragmentation analysis on the snapshot of the network-wide transmission resource occupancy status to identify scattered idle fiber segments with fiber core utilization rates below a preset fragmentation threshold and congested fiber segments with fiber core utilization rates above a preset overload threshold. Generate a transmission resource balancing adjustment instruction, which includes a list of source optical cable segment unit identifiers to be adjusted, destination optical cable segment unit identifiers, and transmission channel identifiers to be migrated. According to the transmission resource balancing adjustment instruction, some transmission channels in the congested fiber chip segment are migrated to the scattered idle fiber chip segment, thereby realizing the redistribution of transmission channel occupancy.

10. A smart optimization system for optical fiber transmission in data centers, characterized in that, The method includes a processor and a computer-readable storage medium storing machine-executable instructions, which, when executed by a computer, implement the intelligent optimization method for optical cable transmission in data centers as described in any one of claims 1-9.

Images