PON system load balancing switching method, device, equipment and medium
By building a predictive model and deep learning algorithm in the PON system, calculating the bandwidth utilization ratio and performing load balancing switching between OLTs, the problem of wasted service bearings of backup OLTs in the TYPE D daisy-chain protection architecture is solved, improving resource utilization and system performance.
Patent Information
- Application Number
- CN202411193370.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-22
AI Technical Summary
In PON systems, the service bearer of the backup OLT in the TYPE D daisy-chain protection architecture is wasted, and the user bandwidth limitation is indiscriminate, which affects resource utilization and system performance.
A predictive model is built using deep learning algorithms to calculate the bandwidth utilization ratio of the PON port and perform load balancing switching between OLTs when the load is unbalanced. Intelligent analysis and switching decisions are made in conjunction with the device operating status.
It improves system resource utilization and overall performance, ensures load balancing while achieving protection objectives, and enhances system reliability and stability.
Smart Images

Figure CN119012056A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of PON protection technology, and in particular to a PON system load balancing switching method, device, equipment and medium. Background Art
[0002] Among the current passive optical network (PON) protection technologies, one of the full protection solutions is to adopt the TYPE D hand-in-hand protection architecture. The TYPE D hand-in-hand protection architecture refers to the protection mechanism used in passive optical network systems such as power. In the TYPE D protection architecture, there are usually two PON ports, one as the main one and the other as the backup. They back up each other like hand in hand. When one port fails, the other port immediately takes over. This protection mechanism is to improve the reliability and stability of the network. When a failure occurs, it can quickly switch to the backup link to ensure uninterrupted communication in critical services, thereby ensuring the continuity and security of power communication. It is often deployed in scenarios with high protection requirements such as power distribution networks.
[0003] like Figure 1 As shown in the figure, in general, the TYPE D hand-in-hand protection architecture in the PON system involves two optical line terminals (OLTs), two 1:N optical splitters and two optical network units (ONUs) of PON MAC, which can comprehensively protect the OLT PON MAC, optical links and ONU PON MAC in the PON system deployment. This conventional deployment can only achieve the purpose of protection, but in many cases, there is a great waste of service carrying for the backup OLT, and the bandwidth limit that users can use is the same as when using one OLT, which is not conducive to the effective use of resources and may also affect the overall performance of the system. Summary of the invention
[0004] The purpose of this application is to address the above problems and provide a PON system load balancing switching method, device, equipment and medium to load balance the business flows carried by the PON ports of two OLTs in the PON system while achieving the protection purpose, thereby effectively improving the system resource utilization and overall performance.
[0005] In a first aspect, the present application provides a load balancing switching method for a PON system, wherein the PON system includes a first OLT and a first PON port corresponding thereto, and a second OLT and a second PON port corresponding thereto, a link switching protection mechanism is established between the first OLT and the second OLT, the first PON port is communicatively connected to a main PON port of a plurality of first ONUs and a backup PON port of a plurality of second ONUs, the second PON port is communicatively connected to a main PON port of a plurality of second ONUs and a backup PON port of a plurality of first ONUs, and the method includes:
[0006] Obtaining a first service parameter of the first OLT and a first traffic parameter corresponding to the first PON port, and calculating a first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first traffic parameter and a preset bandwidth prediction model; the preset bandwidth prediction model is obtained by training based on sample service parameters and sample bandwidth usage ratio values;
[0007] If it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, obtaining a second service parameter of the second OLT and a second traffic parameter corresponding to the second PON port, and calculating a second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model;
[0008] A load balancing characteristic parameter is calculated according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than a second preset threshold, load balancing switching is performed between the first OLT and the second OLT.
[0009] According to the technical solution provided by this application, the method also includes:
[0010] Collecting the sample service parameters and bandwidth usage percentage values corresponding to the sample service parameters;
[0011] An initial prediction model is constructed based on a deep learning algorithm, and the sample business parameters and the bandwidth usage ratio are input into the initial prediction model for training to obtain the preset bandwidth prediction model.
[0012] According to the technical solution provided by the present application, the first bandwidth occupancy ratio corresponding to the first PON port is calculated according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model, including:
[0013] Inputting the first service parameter into the preset bandwidth prediction model to obtain a first predicted bandwidth usage ratio value corresponding to each first ONU connected to the first PON port;
[0014] Calculate a first predicted occupied bandwidth value corresponding to each of the first ONUs connected to the first PON port according to the maximum bandwidth value corresponding to each of the first ONUs and the first predicted bandwidth usage ratio;
[0015] Calculate a first bearer occupied bandwidth value according to the first traffic parameter;
[0016] The ratio of the sum of the first bearer occupied bandwidth value and the first predicted occupied bandwidth values corresponding to all the first ONUs to the first PON port capacity is used as the first bandwidth occupancy ratio.
[0017] According to the technical solution provided by the present application, the calculating the second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model includes:
[0018] Inputting the second service parameter into the preset bandwidth prediction model to obtain a second predicted bandwidth usage ratio value corresponding to each second ONU connected to the second PON port;
[0019] Calculate the second predicted occupied bandwidth value corresponding to each of the second ONUs connected to the second PON port according to the maximum bandwidth value corresponding to each of the second ONUs and the second predicted bandwidth usage ratio value;
[0020] Calculate a second bearer occupied bandwidth value according to the second traffic parameter;
[0021] The ratio of the sum of the second bearer occupied bandwidth value and the second predicted occupied bandwidth values corresponding to all the second ONUs to the second PON port capacity is used as the second bandwidth occupancy ratio.
[0022] According to the technical solution provided in the present application, the load balancing switching between the first OLT and the second OLT includes:
[0023] Acquire a third flow parameter of each first ONU accessed to the first PON port and a fourth flow parameter of each second ONU accessed to the second PON port;
[0024] Calculate a first probability of the first ONU to be switched and a second probability of the second ONU to be switched according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter;
[0025] Acquire a first device operating parameter of the first OLT and a second device operating parameter of the second OLT, and respectively calculate a first failure probability of the first OLT and a second failure probability of the second OLT according to the first device operating parameter, the second device operating parameter and a preset fault prediction model; wherein the preset fault prediction model is obtained by training based on sample device operating parameters and sample fault probabilities;
[0026] A first switching probability of the first ONU is calculated according to the first probability to be switched and the second failure probability, a second switching probability of the second ONU is calculated according to the second probability to be switched and the first failure probability, a target ONU is determined according to the first switching probability and the second switching probability, and a main PON port of the target ONU is switched between the first OLT and the second OLT.
[0027] According to the technical solution provided by the present application, the first probability of the first ONU to be switched and the second probability of the second ONU to be switched are calculated according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter, including:
[0028] According to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter, taking the difference between the number of ONUs accessing the first OLT through the main PON port and the number of ONUs accessing the second OLT through the main PON port after switching as a constraint condition being less than a third preset threshold, and taking the minimum difference in service flow occupancy ratio between the first OLT and the second OLT after switching as a goal, a first probability of the first ONU to be switched and a second probability of the second ONU to be switched are calculated.
[0029] According to the technical solution provided by the present application, the target ONU includes a first target ONU, or a first target ONU and a second target ONU; accordingly, determining the target ONU according to the first switching probability and the second switching probability, and switching the main PON port of the target ONU between the first OLT and the second OLT, includes:
[0030] Acquire a first ONU whose first switching probability is greater than a fourth preset threshold as a first target ONU;
[0031] Acquire a second ONU whose second switching probability is greater than a fifth preset threshold as a second target ONU;
[0032] Switch the main PON port of the first target ONU to the second OLT; or switch the main PON port of the first target ONU to the second OLT and switch the main PON port of the second target ONU to the first OLT.
[0033] In a second aspect, the present application provides a PON system load balancing switching device, which is applied to the PON system, including:
[0034] an acquisition module, configured to acquire a first service parameter of the first OLT and a first flow parameter corresponding to the first PON port, and calculate a first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first flow parameter and a preset bandwidth prediction model; the preset bandwidth prediction model is obtained by training based on sample service parameters and sample bandwidth usage ratio values;
[0035] a processing module, configured to obtain a second service parameter of the second OLT and a second traffic parameter corresponding to the second PON port if it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, and calculate a second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model;
[0036] The switching module is used to calculate a load balancing characteristic parameter according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and perform load balancing switching between the first OLT and the second OLT if it is determined that the load balancing characteristic parameter is greater than a second preset threshold.
[0037] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described above when executing the computer program.
[0038] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and the computer program implements the method described above when executed by a processor.
[0039] Compared with the prior art, the present application has the following beneficial effects: the PON system load balancing switching method, device, electronic device and computer-readable storage medium provided by the present application calculate the first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model obtained; if it is determined that the first bandwidth occupancy ratio is greater than the first preset threshold, the second bandwidth occupancy ratio corresponding to the second PON port is calculated according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model obtained; the load balancing characteristic parameter is calculated according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than the second preset threshold, load balancing switching is performed between the first OLT and the second OLT, and the bandwidth usage of the ONU corresponding to the OLT is intelligently analyzed through a deep learning algorithm, thereby improving the accuracy and efficiency of the analysis; the feasibility and necessity of load balancing switching are comprehensively evaluated by combining two independent machine training models that respectively characterize the load condition of the local OLT and the device status of the opposite OLT, thereby improving the accuracy and reliability of the load balancing switching condition judgment; at the same time, the service flow size is balanced by actively switching the main PON port of the ONU, which is suitable for TYPE The PON system with D hand-in-hand protection structure can achieve the purpose of system protection while performing load balancing switching on the business flows carried by the PON ports of the two OLTs in the PON system, effectively improving the utilization rate of system resources and the reliability and stability of system operation.
[0040] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in this application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it is understood that the description of features or beneficial effects means that specific technical features, technical solutions or beneficial effects are included in at least one embodiment. Therefore, the description of technical features, technical solutions or beneficial effects in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and beneficial effects described in the present embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that the embodiment can be realized without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects can also be identified in a specific embodiment that does not embody all embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the technical solution in this embodiment, the drawings required for use in the description of the embodiment will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 A schematic diagram of a PON system provided in one embodiment of the present application;
[0043] Figure 2 A flow chart of a PON system load balancing switching method provided in one embodiment of the present application;
[0044] Figure 3 A flow chart of a PON system load balancing switching method provided by another embodiment of the present application;
[0045] Figure 4 A schematic diagram of a load balancing switching device for a PON system provided in an embodiment of the present application;
[0046] Figure 5 A schematic diagram of a computer system of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0047] In order to enable those skilled in the art to better understand the technical solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings. The description in this section is only exemplary and explanatory and should not have any limiting effect on the protection scope of the present application. Specifically, the described embodiments are only embodiments of a part of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work should fall within the scope of protection of the present application.
[0048] It should be noted that similar reference numerals and letters represent similar items in the following figures, so once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0049] In order to facilitate the understanding of this application, a brief introduction to the related technology is first given:
[0050] Among the current PON protection technologies, a full protection solution is to use the TYPE D hand-in-hand protection architecture. The TYPE D hand-in-hand protection architecture refers to a protection mechanism used in the PON system. This protection mechanism is designed to improve the reliability and stability of the network. When a fault occurs, it can quickly switch to the backup link to ensure uninterrupted communication in key services, thereby ensuring the continuity and security of communication. It is often deployed in scenarios with high protection requirements, such as power distribution networks. Figure 1 As shown in the figure, in general, the TYPE D hand-in-hand protection architecture in the PON system involves two OLTs, two 1:N optical splitters and two ONUs with PON ports, which can comprehensively protect the OLT PON MAC, optical links and ONU PON MAC in the PON system deployment. If it is deployed according to the conventional method, it can only achieve the protection purpose. In many cases, there is a great waste of business carrying for the backup OLT, and the bandwidth restriction that can be used by users is no different from that when using one OLT. This is not conducive to the effective use of resources and may also affect the overall performance of the system.
[0051] In view of this, an embodiment of the present application provides a load balancing switching method for a PON system, wherein the method calculates a first bandwidth occupancy ratio corresponding to a first PON port according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model obtained; if it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, the second bandwidth occupancy ratio corresponding to the second PON port is calculated according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model obtained; a load balancing characteristic parameter is calculated according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than the second preset threshold, load balancing switching is performed between the first OLT and the second OLT, so that while achieving the protection purpose, the PON ports of the two OLTs in the PON system carry service flows and load balancing switching is performed, effectively improving the system resource utilization and overall performance.
[0052] In order to make the technical solution of the present application clearer and easier to understand, the load balancing switching method of the PON system provided in the embodiment of the present application is introduced below.
[0053] like Figure 2As shown, the figure is a flow chart of a load balancing switching method for a PON system provided by an embodiment of the present application, the PON system includes a first OLT and a first PON port corresponding thereto and a second OLT and a second PON port corresponding thereto, a link switching protection mechanism is established between the first OLT and the second OLT, the first PON port is communicatively connected with a plurality of main PON ports of the first ONUs and a plurality of backup PON ports of the second ONUs, the second PON port is communicatively connected with a plurality of main PON ports of the second ONUs and a plurality of backup PON ports of the first ONUs, the method includes the following steps:
[0054] S101. Obtain a first service parameter of the first OLT and a first traffic parameter corresponding to the first PON port, and calculate a first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first traffic parameter and a preset bandwidth prediction model; the preset bandwidth prediction model is obtained by training based on sample service parameters and sample bandwidth usage ratio values;
[0055] Specifically, the current first service parameters of one of the OLTs (i.e., the first OLT) in the PON system TYPE D hand-in-hand protection architecture are obtained in real time every preset time period, and the first service parameters include the number of MAC addresses in the OLT whole machine MAC address table (macCnt); the number of PON port MAC addresses (ponMacCnt); the number of MAC addresses corresponding to each ONU under the PON port (because the maximum number of ONUs at the PON port in the current technology is 256, the range is 0-255); historical performance statistics: the average uplink and downlink rates and peak rates of the uplink port of the OLT protection group (upPortInAvgRate, upPortOutAvgRate, upPortInPeakRate, upPortOutAvgRate, upPortInPeakRate, upPortOutAvgRate, upPortOutPeak ... tOutPeakRate), the average and peak rates of the upstream and downstream of the PON port (ponPortInAvgRate, ponPortOutAvgRate, ponPortInPeakRate, ponPortOutPeakRate); the fixed bandwidth, guaranteed bandwidth, and best-effort bandwidth values configured for each ONU under the PON port (onuFixBw[i], onuConfirmBw[i], and onuBestEffortBw[i]); port speed limit configuration: the speed limit values of the OLT uplink port and the PON port (upPortRate and ponPortRate); service VLAN ID value (srvVlan[j], j ranges from 1 to 4094); timestamp (timeVal); other business data may also be included, which may be set and adjusted according to actual needs, and are not specifically limited here, but need to be consistent with the data type selected when the preset bandwidth prediction model is trained based on sample business parameters and sample occupied bandwidth values; it is understandable that the above business data can be extracted and collected through the original SNMP interface or internal API of the first OLT; and, the MAC address table data does not require specific MAC address values, but only needs to extract the total number of MAC addresses learned by the OLT, the number of MAC addresses of the protection group PON port, and the number of MAC addresses of each ONU under the protection group, and the historical performance data only needs to extract the protection group PON port / uplink port / each ONU.
[0056] At the same time, the business flow carried by the first PON port corresponding to the first OLT is obtained and marked with the flow size of the first PON port, that is, the first flow parameter, and this value can be obtained through the performance statistics function of the first PON port; it should be noted that the first flow parameter can be an instantaneous peak value or an average value within a certain time period. At the same time, other non-business interference parameters need to be excluded in the performance statistics process (for example, whether an ONU upgrade is in progress, whether a loop is generated, etc.), which can be set and adjusted according to actual conditions, and no specific limitation is made here.
[0057] A first predicted occupied bandwidth value corresponding to each first ONU connected to the first PON port is calculated according to the first service parameter and a preset bandwidth prediction model; then a first bearer occupied bandwidth value is calculated according to the first traffic parameter, and then the first bandwidth occupancy ratio is calculated according to the first bearer occupied bandwidth value and the first predicted occupied bandwidth values corresponding to all the first ONUs.
[0058] S102: If it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, a second service parameter of the second OLT and a second traffic parameter corresponding to the second PON port are obtained, and a second bandwidth occupancy ratio corresponding to the second PON port is calculated according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model;
[0059] Specifically, if it is determined that the first bandwidth occupancy ratio is greater than the first preset threshold, it is determined that the current load of the first PON port is high, and a load balancing switch needs to be performed with the second OLT at the opposite end. However, whether to perform the load balancing switch also needs to consider the load condition of the second OLT at the opposite end. Therefore, the second service parameter of the second OLT and the second flow parameter corresponding to the second PON port are obtained, and the second predicted occupied bandwidth value corresponding to each second ONU connected to the second PON port is calculated according to the second service parameter and the preset bandwidth prediction model; then the second bearer occupied bandwidth value is calculated according to the second flow parameter, and then the second bandwidth occupancy ratio is calculated according to the second bearer occupied bandwidth value and the second predicted occupied bandwidth value corresponding to all the second ONUs. Among them, the attributes and acquisition methods of the second service parameter and the second flow parameter are similar to those of the first service parameter and the first flow parameter, and are not repeated here; the first preset threshold can be set according to experience or obtained through model training. It can be understood that the first preset threshold can be 50% or 80%, which can be set and adjusted according to actual conditions, and is not specifically limited here.
[0060] S103: Calculate a load balancing characteristic parameter according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than a second preset threshold, perform load balancing switching between the first OLT and the second OLT.
[0061] Specifically, the difference between the first bandwidth occupancy ratio and the second bandwidth occupancy ratio is used as the load balancing characteristic parameter to characterize the load balancing degree between the first OLT and the second OLT. The higher the load balancing characteristic parameter is, the more unbalanced the load between the first OLT and the second OLT is, and vice versa, the more balanced the load between the first OLT and the second OLT is. When the load balancing characteristic parameter is greater than the second preset threshold, it indicates that the load balancing degree between the first OLT and the second OLT is too low, and load balancing switching is performed between the first OLT and the second OLT. Among them, the second preset threshold can be set according to experience or obtained through model training. The second preset threshold cannot be too small to avoid frequent switching, but it cannot be too large, otherwise the purpose of traffic balancing cannot be achieved. It can be set and adjusted according to actual conditions, and is not specifically limited here.
[0062] The embodiment of the present application provides a load balancing switching method for a PON system. According to the service parameters of the first OLT obtained and the preset bandwidth prediction model, the predicted occupied bandwidth value corresponding to each first ONU connected to the first PON port is calculated, and then according to the predicted occupied bandwidth value and the first traffic parameter corresponding to the first PON port, the first bandwidth occupancy ratio corresponding to the first PON port is calculated. When it is determined that the first bandwidth occupancy ratio is greater than the first preset threshold, the second bandwidth occupancy ratio corresponding to the second PON port is calculated according to the second traffic parameter corresponding to the second PON port. Then, if it is determined that the difference between the first bandwidth occupancy ratio and the second bandwidth occupancy ratio is greater than the second preset threshold, load balancing switching is performed between the first OLT and the second OLT, so that while achieving the protection purpose, the PON port-borne service flows of the two OLTs in the PON system are load balanced switched, effectively improving the system resource utilization and overall performance.
[0063] In the above embodiment, the method further includes:
[0064] Collecting the sample service parameters and bandwidth usage percentage values corresponding to the sample service parameters;
[0065] Specifically, in each preset time period, the original data is extracted and collected through the original SNMP interface or internal API of the first OLT, and the parameters that can represent the bandwidth usage of the first ONU are screened out for preprocessing, including but not limited to data cleaning (for example, removing noise and outliers), data conversion (such as converting non-numerical data into numerical data), data normalization (for example, scaling the data to the same range), etc., to obtain the sample service parameters, and at the same time, the bandwidth usage percentage value of each first ONU connected to the first PON port corresponding to the first OLT corresponding to the sample service parameters is collected, and the sample service parameters and the bandwidth usage percentage value both carry timestamps to ensure the mutual correspondence between the two and the accuracy of model training. It can be understood that in the actual application process, the type of sample service parameters can be selected according to the specific situation, and then the model training and verification can be carried out, and then it can be continuously iterated and corrected in combination with the actual situation to continuously improve the accuracy of the model output results.
[0066] An initial prediction model is constructed based on a deep learning algorithm, and the sample business parameters and the bandwidth usage ratio are input into the initial prediction model for training to obtain the preset bandwidth prediction model.
[0067] Specifically, an initial prediction model is constructed based on deep learning algorithms such as a multi-layer perceptron (MLP) or a convolutional neural network (CNN), including determining the number of network layers, the number of neurons in each layer, an activation function (such as a Softmax function), a loss function, etc. The collected sample business parameters and the bandwidth usage percentage values corresponding to the sample business parameters are divided into a training set, a validation set, and a test set, and the initial prediction model is trained and learned. At the same time, the test set is used to evaluate each trained model. The model parameters are continuously adjusted through an optimization algorithm (such as a gradient descent algorithm, a stochastic gradient descent algorithm, etc.) to optimize the model, and the preset bandwidth prediction model is obtained.
[0068] The PON system load balancing switching method provided in the present application obtains a preset bandwidth prediction model through deep learning algorithm construction and training, which is used to perform intelligent analysis on the ONU bandwidth usage corresponding to the OLT, thereby improving the accuracy and efficiency of the analysis.
[0069] In the above embodiment, the calculating the first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model includes:
[0070] Inputting the first service parameter into the preset bandwidth prediction model to obtain a first predicted bandwidth usage ratio value corresponding to each first ONU connected to the first PON port;
[0071] Specifically, the first service parameter of the first PON port obtained in real time is input into the preset bandwidth prediction model obtained through training, and a first predicted bandwidth usage ratio value corresponding to each first ONU connected to the first PON port is output. The output value of the preset bandwidth prediction model is a percentage value. The first predicted bandwidth usage ratio value of each ONU under the first PON port is expressed in a range of 1% to 100%. The larger the value, the larger the bandwidth it will occupy.
[0072] Calculate a first predicted occupied bandwidth value corresponding to each of the first ONUs connected to the first PON port according to the maximum bandwidth value corresponding to each of the first ONUs and the first predicted bandwidth usage ratio;
[0073] Specifically, the first predicted bandwidth usage percentage value corresponding to each of the first ONUs is multiplied by its corresponding maximum bandwidth value to obtain the first predicted occupied bandwidth value corresponding to each of the first ONUs accessing the first PON port. It can be understood that the maximum bandwidth value is a performance attribute parameter of the ONU, which is related to the ONU itself and can be determined based on the ONU used in the actual application process, and is not specifically limited here.
[0074] Calculate a first bearer occupied bandwidth value according to the first traffic parameter;
[0075] The ratio of the sum of the first bearer occupied bandwidth value and the first predicted occupied bandwidth values corresponding to all the first ONUs to the first PON port capacity is used as the first bandwidth occupancy ratio.
[0076] Specifically, a bandwidth value occupied by a service flow currently carried by the first PON port, that is, the first bearer occupied bandwidth value, is calculated according to the first traffic parameter, and a sum of the first bearer occupied bandwidth value and a first predicted occupied bandwidth value corresponding to the first ONU is used as the predicted occupied bandwidth value corresponding to the current first PON port, and a ratio of the predicted occupied bandwidth value corresponding to the first PON port to the capacity of the first PON port is used as the first bandwidth occupancy ratio corresponding to the first PON port.
[0077] In the above embodiment, the calculating the second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model includes:
[0078] Inputting the second service parameter into the preset bandwidth prediction model to obtain a second predicted bandwidth usage ratio value corresponding to each second ONU connected to the second PON port;
[0079] Specifically, the second service parameter of the second PON port is input into the preset bandwidth prediction model obtained by training, and the second predicted bandwidth usage ratio corresponding to each second ONU connected to the second PON port is output. The output value of the preset bandwidth prediction model is a percentage value. The second predicted bandwidth usage ratio value for each ONU under the second PON port is expressed by 1% to 100%. The larger the value, the larger the bandwidth it will occupy.
[0080] Calculate the second predicted occupied bandwidth value corresponding to each of the second ONUs connected to the second PON port according to the maximum bandwidth value corresponding to each of the second ONUs and the second predicted bandwidth usage ratio value;
[0081] Specifically, the second predicted bandwidth usage percentage value corresponding to each second ONU is multiplied by its corresponding maximum bandwidth value to obtain the second predicted occupied bandwidth value corresponding to each second ONU accessing the second PON port. It can be understood that the maximum bandwidth value is a performance attribute parameter of the ONU, which is related to the ONU itself and can be determined according to the ONU used in the actual application process, and is not specifically limited here.
[0082] Calculate a second bearer occupied bandwidth value according to the second traffic parameter;
[0083] The ratio of the sum of the second bearer occupied bandwidth value and the second predicted occupied bandwidth values corresponding to all the second ONUs to the second PON port capacity is used as the second bandwidth occupancy ratio.
[0084] Specifically, the bandwidth value occupied by the service flow currently carried by the second PON port, that is, the second bearer occupied bandwidth value, is calculated according to the second traffic parameter, and the sum of the second bearer occupied bandwidth value and the second predicted occupied bandwidth value corresponding to the second ONU is used as the predicted occupied bandwidth value corresponding to the current second PON port, and the ratio of the predicted occupied bandwidth value corresponding to the second PON port to the capacity of the second PON port is used as the second bandwidth occupancy ratio corresponding to the second PON port.
[0085] In the above embodiment, the load balancing switching between the first OLT and the second OLT includes:
[0086] Acquire a third flow parameter of each first ONU accessed to the first PON port and a fourth flow parameter of each second ONU accessed to the second PON port;
[0087] Specifically, the service flow carried by each first ONU connected to the first PON port and the service flow carried by each second ONU connected to the second PON port are marked with traffic size, that is, the third traffic parameter and the fourth traffic parameter, and this value can be obtained through the performance statistics function of the PON port of the first ONU and the PON port of the second ONU respectively; it should be noted that the third traffic parameter and the fourth traffic parameter can be both instantaneous peak values or average values within a certain time period. At the same time, other non-service interference parameters (for example, whether an ONU upgrade is being performed, whether a loop is generated, etc.) need to be excluded in the performance statistics process. They can be set and adjusted according to actual conditions, and are not specifically limited here.
[0088] Calculate a first probability of the first ONU to be switched and a second probability of the second ONU to be switched according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter;
[0089] Specifically, according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter, and the fourth traffic parameter, the difference between the number of ONUs connected to the first OLT through the main PON port and the number of ONUs connected to the second OLT through the main PON port after switching is less than a third preset threshold as a constraint condition, and the difference between the service flow occupancy ratio of the first OLT and the second OLT after switching is minimized as a goal, a first probability of the first ONU to be switched and a second probability of the second ONU to be switched are calculated. The first probability to be switched and the second probability to be switched respectively represent the probability of switching the main PON port of the first ONU to the second OLT and the probability of switching the main PON port of the second ONU to the first OLT when the difference between the service flow occupancy ratio of the first OLT and the second OLT after switching is minimized as a goal.
[0090] Acquire a first device operating parameter of the first OLT and a second device operating parameter of the second OLT, and respectively calculate a first failure probability of the first OLT and a second failure probability of the second OLT according to the first device operating parameter, the second device operating parameter and a preset fault prediction model; wherein the preset fault prediction model is obtained by training based on sample device operating parameters and sample fault probabilities;
[0091] Specifically, the first device operating parameters of the first OLT and the second device operating parameters of the second OLT are obtained, and the first device operating parameters and the second device operating parameters both include the power-on operation time of the OLT device, the historical alarm information of the OLT, the temperature of the OLT device, the storage space occupancy rate, the CPU occupancy rate, the optical module data, etc., and may also include other parameters used to characterize the operation status of the device, which may be set and adjusted according to the actual situation, and are not specifically limited here. Then, the first device operating parameters and the second device operating parameters are respectively input into the preset fault prediction model, and the first fault probability of the first OLT and the second fault probability of the second OLT are output. Among them, the preset fault prediction model is obtained by training with sample device operating parameters and sample fault probabilities: an initial prediction model can be constructed by using multiple multi-layer perceptrons (MLP), including an input layer, a hidden layer and an output layer, wherein the input layer includes multiple nodes, namely, the device operating parameters; the output layer includes a node, and the output value is mapped to between 0 and 1 using the Sigmoid activation function, namely, the probability of the current first OLT failing; the hidden layer is one or more layers, which are responsible for learning the complex relationship between the input (device operating parameters) and the output (fault probability). The collected sample equipment operating parameters and sample failure probabilities are divided into a training set, a validation set and a test set. The initial prediction model is trained and learned through the sample equipment operating parameters and the sample failure probabilities. At the same time, the test set is used to evaluate each trained model. The model parameters are continuously adjusted through optimization algorithms (such as gradient descent algorithm, stochastic gradient descent algorithm, etc.) to optimize the model and obtain the preset fault prediction model.
[0092] A first switching probability of the first ONU is calculated according to the first probability to be switched and the second failure probability, a second switching probability of the second ONU is calculated according to the second probability to be switched and the first failure probability, a target ONU is determined according to the first switching probability and the second switching probability, and a main PON port of the target ONU is switched between the first OLT and the second OLT.
[0093] Specifically, the average value of the first probability to be switched and the second probability of failure can be used as the first switching probability of the first ONU, or the first probability to be switched and the second probability of failure can be weighted averaged according to different weights according to a preset formula to calculate the switching probability of the first ONU, the preset formula is: P = P1 × ω1 + P2 × ω2; wherein P1 is the first probability to be switched, ω1 is the first weight, P2 is the second probability of failure, ω2 is the second weight, and ω1 + ω2 = 1. Similarly, the average value of the second probability to be switched and the first probability of failure can be used as the second switching probability of the second ONU, or the second probability to be switched and the first probability of failure can be weighted averaged according to a preset formula to calculate the switching probability of the second ONU, the preset formula is: P = P3 × ω1 + P4 × ω2; wherein P3 is the second probability to be switched, ω1 is the first weight, P4 is the first probability of failure, ω2 is the second weight, and ω1 + ω2 = 1. The first switching probability of the first ONU and the second switching probability of the second ONU can also be calculated according to other strategies, which can be set and adjusted according to actual conditions, and are not specifically limited here. The target ONU is determined according to the first switching probability and the second switching probability. The target ONU may include only one or more first ONUs whose current main PON port is connected to the first PON port corresponding to the first OLT, that is, the first target ONU, and then unidirectional switching is performed, that is, only the main PON port of the first target ONU is switched to the second OLT; or the target ONU includes both one or more first ONUs whose current main PON port is connected to the first PON port corresponding to the first OLT, that is, the first target ONU, and one or more second ONUs whose current main PON port is connected to the second PON port corresponding to the second OLT, that is, the second target ONU, and then bidirectional switching is performed, that is, not only the main PON port of the first target ONU is switched to the second OLT, but also the main PON port of the second target ONU is switched to the first OLT.
[0094] The PON system load balancing switching method provided in the present application combines two independent machine training models that respectively characterize the load condition of the local OLT and the device status of the opposite OLT, comprehensively evaluates the feasibility and necessity of load balancing switching, and improves the accuracy and reliability of load balancing switching condition judgment.
[0095] In the above embodiment, the calculating, according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter, the first probability of the first ONU to be switched and the second probability of the second ONU to be switched includes:
[0096] According to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter, taking the difference between the number of ONUs accessing the first OLT through the main PON port and the number of ONUs accessing the second OLT through the main PON port after switching as a constraint condition being less than a third preset threshold, and taking the minimum difference in service flow occupancy ratio between the first OLT and the second OLT after switching as a goal, a first probability of the first ONU to be switched and a second probability of the second ONU to be switched are calculated.
[0097] Specifically, when performing load balancing switching, it is necessary to select a suitable first ONU or a suitable first ONU and second ONU to switch simultaneously, so as to try to ensure that after part of the traffic of the current first ONU is switched to the second OLT, the difference (absolute value) of the bandwidth occupancy ratio between the two OLTs is reduced to a minimum value, that is, the difference in the service flow occupancy ratio is the smallest, so each of the first ONUs is sequentially used as the first ONU to be switched, and the first service flow occupancy ratio difference corresponding to switching only the main PON port of the first ONU to be switched to the second OLT and the second service flow occupancy ratio difference corresponding to switching the main PON port of the first ONU to be switched to the second OLT and sequentially switching the main PON port of each of the second ONUs to the first OLT are calculated respectively; taking the difference between the number of ONUs connected to the first OLT through the main PON port and the number of ONUs connected to the second OLT through the main PON port after switching as less than a third preset threshold as a constraint condition, the first probability of each of the first ONUs to be switched and the second probability of each of the second ONUs to be switched are calculated according to the first service flow occupancy ratio difference and the second service flow occupancy ratio difference. Since multiple first ONUs may be switched at the same time, the specific switching selection also needs to be optimized through training, including: whether to choose unidirectional switching (i.e., switching only from the first OLT to the second OLT) or bidirectional switching (switching between the first OLT and the second OLT) in the same switching to minimize the difference in the proportion of business flow occupancy after switching. At the same time, it is also necessary to consider the difference in the number of ONUs carried by the two OLTs after switching. The constraint condition is that the difference between the number of ONUs connected to the first OLT through the main PON port and the number of ONUs connected to the second OLT through the main PON port after switching is less than a third preset threshold. In most cases, the third preset threshold should not be too large to cope with the problem of triggering excessively frequent switching when some ONU traffic changes suddenly.
[0098] In the above embodiment, the target ONU includes a first target ONU, or a first target ONU and a second target ONU; accordingly, determining the target ONU according to the first switching probability and the second switching probability, and switching the main PON port of the target ONU between the first OLT and the second OLT, includes:
[0099] Acquire a first ONU whose first switching probability is greater than a fourth preset threshold as a first target ONU;
[0100] Acquire a second ONU whose second switching probability is greater than a fifth preset threshold as a second target ONU;
[0101] Switch the main PON port of the first target ONU to the second OLT; or switch the main PON port of the first target ONU to the second OLT and switch the main PON port of the second target ONU to the first OLT.
[0102] Specifically, the first ONU whose first switching probability is greater than the fourth preset threshold is used as the first target ONU, and the second ONU whose second switching probability is greater than the fifth preset threshold is used as the second target ONU. If the number of the second target ONUs is zero, only the main PON port of the first target ONU is switched to the second OLT; if the number of the second target ONUs is not zero, the main PON port of the first target ONU is switched to the second OLT and the main PON port of the second target ONU is switched to the first OLT. It can be understood that the fourth preset threshold and the fifth preset threshold can be equal or unequal in size, and can be set and adjusted according to actual conditions, and are not specifically limited here.
[0103] A PON system load balancing switching method provided by an embodiment of the present application calculates a first bandwidth occupancy ratio corresponding to a first PON port according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model obtained; if it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, the second bandwidth occupancy ratio corresponding to a second PON port is calculated according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model obtained; a load balancing characteristic parameter is calculated according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than the second preset threshold, load balancing switching is performed between the first OLT and the second OLT, and the bandwidth usage of the ONU corresponding to the OLT is intelligently analyzed through a deep learning algorithm, thereby improving the accuracy and efficiency of the analysis; two independent machine training models that respectively characterize the load condition of the local OLT and the device status of the opposite OLT are combined to comprehensively evaluate the feasibility and necessity of the load balancing switching, thereby improving the accuracy and reliability of the load balancing switching condition judgment; at the same time, the service flow size is balanced by actively switching the main PON port of the ONU, which is suitable for TYPE The PON system with D hand-in-hand protection structure can achieve the purpose of system protection while performing load balancing switching on the business flows carried by the PON ports of the two OLTs in the PON system, effectively improving the utilization rate of system resources and the reliability and stability of system operation.
[0104] like Figure 3 As shown, the figure is a flow chart of a PON system load balancing switching method provided by this embodiment, and the method includes the following steps:
[0105] S201, obtaining a first service parameter of a first OLT, and obtaining and calculating a first predicted occupied bandwidth value corresponding to each first ONU connected to a first PON port according to the first service parameter and a preset bandwidth prediction model;
[0106] S202, obtaining a first traffic parameter corresponding to the first PON port, and calculating a first bandwidth occupancy ratio corresponding to the first PON port according to the first predicted occupied bandwidth value and the first traffic parameter;
[0107] S203, determining whether the first bandwidth occupancy ratio is greater than a first preset threshold, if so, executing step S204; otherwise, returning to step S201;
[0108] S204, obtaining a second service parameter of the second OLT, and obtaining and calculating a second predicted occupied bandwidth value corresponding to each second ONU connected to the second PON port according to the second service parameter and a preset bandwidth prediction model;
[0109] S205. Obtain a second traffic parameter corresponding to the second PON port, and calculate a second bandwidth occupancy ratio corresponding to the second PON port according to the second predicted occupied bandwidth value and the second traffic parameter;
[0110] S206. Calculate a load balancing characteristic parameter according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio;
[0111] S207, determine whether the load balancing characteristic parameter is greater than a second preset threshold; if so, execute step S208; otherwise, return to step S201;
[0112] S208. Acquire a third flow parameter of each first ONU connected to the first PON port, and a fourth flow parameter of each second ONU connected to the second PON port;
[0113] S209, calculating a first probability of the first ONU to be switched and a second probability of the second ONU to be switched according to the first predicted bandwidth value, the second predicted bandwidth value, the third flow parameter and the fourth flow parameter;
[0114] S210, obtaining a first device operating parameter of the first OLT and a second device operating parameter of the second OLT; respectively calculating a first failure probability of the first OLT and a second failure probability of the second OLT according to the first device operating parameter, the second device operating parameter and a preset fault prediction model;
[0115] S211, calculating a first switching probability of the first ONU according to the first probability to be switched and the second failure probability, and calculating a second switching probability of the second ONU according to the second probability to be switched and the first failure probability;
[0116] S212, obtaining a first ONU whose first switching probability is greater than a fourth preset threshold as a first target ONU, obtaining a second ONU whose second switching probability is greater than a fifth preset threshold as a second target ONU; then executing step S2131 or step S2132;
[0117] S2131. Switch the primary PON port of the first target ONU to the second OLT;
[0118] S2132: Switch the main PON port of the first target ONU to the second OLT and switch the main PON port of the second target ONU to the first OLT.
[0119] A PON system load balancing switching method provided by an embodiment of the present application calculates a first bandwidth occupancy ratio corresponding to a first PON port according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model obtained; if it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, the second bandwidth occupancy ratio corresponding to a second PON port is calculated according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model obtained; a load balancing characteristic parameter is calculated according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than the second preset threshold, load balancing switching is performed between the first OLT and the second OLT, and the bandwidth usage of the ONU corresponding to the OLT is intelligently analyzed through a deep learning algorithm, thereby improving the accuracy and efficiency of the analysis; two independent machine training models that respectively characterize the load condition of the local OLT and the device status of the opposite OLT are combined to comprehensively evaluate the feasibility and necessity of the load balancing switching, thereby improving the accuracy and reliability of the load balancing switching condition judgment; at the same time, the service flow size is balanced by actively switching the main PON port of the ONU, which is suitable for TYPE The PON system with D hand-in-hand protection structure can achieve the purpose of system protection while performing load balancing switching on the business flows carried by the PON ports of the two OLTs in the PON system, effectively improving the utilization rate of system resources and the reliability and stability of system operation.
[0120] Combination of the above Figure 1 , Figure 2 and Figure 3 The load balancing switching method of the PON system provided in the embodiment of the present application is introduced in detail. The load balancing switching device, electronic device and computer-readable storage medium of the PON system provided in the embodiment of the present application will be introduced in conjunction with the accompanying drawings.
[0121] like Figure 4 As shown, this figure is a schematic diagram of a load balancing switching device for a PON system provided by the present application, the device is applied to the PON system, and the device includes:
[0122] The acquisition module 301 is used to obtain a first service parameter of the first OLT and a first flow parameter corresponding to the first PON port, and calculate a first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first flow parameter and a preset bandwidth prediction model; the preset bandwidth prediction model is obtained by training based on sample service parameters and sample bandwidth usage ratio values;
[0123] The processing module 302 is configured to obtain a second service parameter of the second OLT and a second traffic parameter corresponding to the second PON port if it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, and calculate a second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model;
[0124] The switching module 303 is used to calculate a load balancing characteristic parameter according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and perform load balancing switching between the first OLT and the second OLT if it is determined that the load balancing characteristic parameter is greater than a second preset threshold.
[0125] The PON system load balancing switching device provided in the embodiment of the present application can correspond to executing the PON system load balancing switching method described in the embodiment of the present application, and the above functions of each module of the device correspond to implementing Figure 2 , 3 For the sake of brevity, the corresponding process of the method shown will not be repeated here.
[0126] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the PON system load balancing switching method as described in the above embodiment is implemented.
[0127] like Figure 5 As shown, the computer system 400 of the electronic device includes a CPU 401, which can perform various appropriate actions and processes according to the program stored in the ROM 402 or the program loaded from the storage part 408 to the RAM 403. In the RAM 403, various programs and data required for system operation are also stored. The CPU 401, the ROM 402 and the RAM 403 are connected to each other through a bus 404. The I / O interface 405 is also connected to the bus 404. Among them, CPU 401 represents a central processing unit, ROM 402 represents a read-only memory, RAM 403 represents a random access memory, and I / O represents input / output.
[0128] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube, a liquid crystal display, etc., and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed, so that a computer program read therefrom is installed into the storage section 408 as needed.
[0129] In particular, the process of the PON system load balancing switching method described in the above embodiment can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains program code for executing the PON system load balancing switching method described in the above embodiment. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 409, and / or installed from the removable medium 411. When the computer program is executed by the CPU 401, the above functions defined in the present computer system 400 are executed.
[0130] The embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the load balancing switching method of the PON system as described in the above embodiment is implemented.
[0131] Specifically, the computer-readable storage medium may be included in the electronic device described in the above embodiment; or it may exist independently without being assembled into the electronic device. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the PON system load balancing switching method described in the above embodiment.
[0132] It should be noted that the computer-readable storage medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above.
[0133] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and its core ideas of this application. The above is only the preferred implementation method of this application. It should be pointed out that due to the limitations of textual expression and the objective existence of infinite specific structures, ordinary technicians in this technical field can make several improvements, modifications or changes without departing from the principles of the present invention, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the inventive concept and technical solution to other occasions without improvement, should be regarded as the scope of protection of this application.
Claims
1. A load balancing switching method for a PON system, wherein the PON system comprises a first OLT and a first PON port corresponding thereto and a second OLT and a second PON port corresponding thereto, wherein a link switching protection mechanism is established between the first OLT and the second OLT, wherein the first PON port is communicatively connected to a plurality of primary PON ports of first ONUs and a plurality of backup PON ports of second ONUs, and the second PON port is communicatively connected to a plurality of primary PON ports of second ONUs and a plurality of backup PON ports of first ONUs, wherein: The method comprises: Obtaining a first service parameter of the first OLT and a first traffic parameter corresponding to the first PON port, and calculating a first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first traffic parameter and a preset bandwidth prediction model; the preset bandwidth prediction model is obtained by training based on sample service parameters and sample bandwidth usage ratio values; If it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, obtaining a second service parameter of the second OLT and a second traffic parameter corresponding to the second PON port, and calculating a second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model; A load balancing characteristic parameter is calculated according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and if it is determined that the load balancing characteristic parameter is greater than a second preset threshold, load balancing switching is performed between the first OLT and the second OLT.
2. The method according to claim 1, characterized in that The method further comprises: Collecting the sample service parameters and bandwidth usage percentage values corresponding to the sample service parameters; An initial prediction model is constructed based on a deep learning algorithm, and the sample business parameters and the bandwidth usage ratio are input into the initial prediction model for training to obtain the preset bandwidth prediction model.
3. The method according to claim 1, characterized in that The calculating the first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first traffic parameter and the preset bandwidth prediction model includes: Inputting the first service parameter into the preset bandwidth prediction model to obtain a first predicted bandwidth usage ratio value corresponding to each first ONU connected to the first PON port; Calculate a first predicted occupied bandwidth value corresponding to each of the first ONUs connected to the first PON port according to the maximum bandwidth value corresponding to each of the first ONUs and the first predicted bandwidth usage ratio; Calculate a first bearer occupied bandwidth value according to the first traffic parameter; The ratio of the sum of the first bearer occupied bandwidth value and the first predicted occupied bandwidth values corresponding to all the first ONUs to the first PON port capacity is used as the first bandwidth occupancy ratio.
4. The method according to claim 1, characterized in that: The calculating the second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model includes: Inputting the second service parameter into the preset bandwidth prediction model to obtain a second predicted bandwidth usage ratio value corresponding to each second ONU connected to the second PON port; Calculate the second predicted occupied bandwidth value corresponding to each of the second ONUs connected to the second PON port according to the maximum bandwidth value corresponding to each of the second ONUs and the second predicted bandwidth usage ratio value; Calculate a second bearer occupied bandwidth value according to the second traffic parameter; The ratio of the sum of the second bearer occupied bandwidth value and the second predicted occupied bandwidth values corresponding to all the second ONUs to the second PON port capacity is used as the second bandwidth occupancy ratio.
5. The method according to claim 1, characterized in that The performing load balancing switching between the first OLT and the second OLT includes: Acquire a third flow parameter of each first ONU accessed to the first PON port and a fourth flow parameter of each second ONU accessed to the second PON port; Calculate a first probability of the first ONU to be switched and a second probability of the second ONU to be switched according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter; Acquire a first device operating parameter of the first OLT and a second device operating parameter of the second OLT, and respectively calculate a first failure probability of the first OLT and a second failure probability of the second OLT according to the first device operating parameter, the second device operating parameter and a preset fault prediction model; wherein the preset fault prediction model is obtained by training based on sample device operating parameters and sample fault probabilities; A first switching probability of the first ONU is calculated according to the first probability to be switched and the second failure probability, a second switching probability of the second ONU is calculated according to the second probability to be switched and the first failure probability, a target ONU is determined according to the first switching probability and the second switching probability, and a main PON port of the target ONU is switched between the first OLT and the second OLT.
6. The method according to claim 5, characterized in that The calculating, according to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter, and the fourth traffic parameter, a first probability of the first ONU to be switched and a second probability of the second ONU to be switched, comprises: According to the first predicted bandwidth value, the second predicted bandwidth value, the third traffic parameter and the fourth traffic parameter, taking the difference between the number of ONUs accessing the first OLT through the main PON port and the number of ONUs accessing the second OLT through the main PON port after switching as a constraint condition being less than a third preset threshold, and taking the minimum difference in service flow occupancy ratio between the first OLT and the second OLT after switching as a goal, a first probability of the first ONU to be switched and a second probability of the second ONU to be switched are calculated.
7. The method according to claim 6, characterized in that The target ONU includes a first target ONU, or a first target ONU and a second target ONU; accordingly, determining the target ONU according to the first switching probability and the second switching probability, and switching the main PON port of the target ONU between the first OLT and the second OLT, includes: Acquire a first ONU whose first switching probability is greater than a fourth preset threshold as a first target ONU; Acquire a second ONU whose second switching probability is greater than a fifth preset threshold as a second target ONU; Switch the main PON port of the first target ONU to the second OLT; or switch the main PON port of the first target ONU to the second OLT and switch the main PON port of the second target ONU to the first OLT.
8. A load balancing switching device for a PON system, applied to the PON system, characterized in that: include: an acquisition module, configured to acquire a first service parameter of the first OLT and a first flow parameter corresponding to the first PON port, and calculate a first bandwidth occupancy ratio corresponding to the first PON port according to the first service parameter, the first flow parameter and a preset bandwidth prediction model; the preset bandwidth prediction model is obtained by training based on sample service parameters and sample bandwidth usage ratio values; a processing module, configured to obtain a second service parameter of the second OLT and a second traffic parameter corresponding to the second PON port if it is determined that the first bandwidth occupancy ratio is greater than a first preset threshold, and calculate a second bandwidth occupancy ratio corresponding to the second PON port according to the second service parameter, the second traffic parameter and the preset bandwidth prediction model; The switching module is used to calculate a load balancing characteristic parameter according to the first bandwidth occupancy ratio and the second bandwidth occupancy ratio, and perform load balancing switching between the first OLT and the second OLT if it is determined that the load balancing characteristic parameter is greater than a second preset threshold.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
