Resource scheduling system based on optical communication network

Abstract

The invention relates to the technical field of optical communication networks, and discloses a resource scheduling system based on an optical communication network, which comprises a multi-user equipment capability difference dynamic evaluation module for evaluating energy level resources of each equipment port in a park, constructing a capability archive library, arranging detection equipment at each equipment port, and determining the energy level resources of each equipment port in the park; the method comprises the following steps that: a resource scheduling module receives a plurality of equipment ports, monitors optical performance and link quality, constructs a link quality detection library of each equipment port, triggers a resource recovery strategy when any equipment port is abnormal, and reserves an initial resource for a newly accessed equipment port based on the link quality detection library and a capability archive library of each equipment port; and intelligent scheduling is carried out. According to the system, intelligent scheduling is carried out based on a link quality detection library of each equipment port and reserved initial resources of each equipment port, so that resource allocation of different ports is completed, and rapid matching of tasks is realized according to different resources held by the ports.

Description

Technical Field

[0001] This invention relates to the field of optical communication network technology, specifically to a resource scheduling system based on optical communication networks. Background Technology

[0002] In the field of optical communication, traditional communication scheduling systems are usually static and difficult to adapt to constantly changing communication needs and network conditions. These systems often fail to effectively manage and allocate network resources, leading to resource waste and decreased communication quality. In addition, with the continuous increase in communication types and data traffic, the complexity of the network is also increasing. There are shortcomings in the allocation of resources between different ports within the campus and the scheduling between tasks. Therefore, a resource scheduling system based on optical communication networks is proposed. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a resource scheduling system based on optical communication networks, which has the advantages of allocating, scheduling, or reclaiming resources held by different ports in the optical communication process, thus solving the aforementioned technical problems.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a resource scheduling system based on an optical communication network, comprising: The multi-user device capability difference dynamic assessment module assesses the capability resources of each device port in the park and builds a capability archive. It deploys detection equipment at each device port to monitor optical performance and link quality, and builds a link quality detection library for each device port. Based on the link quality detection library for each device port, it builds a fault assessment prediction coefficient in real time. When any device port is abnormal, a resource reclamation strategy is triggered. The resource scheduling module reserves initial resources for newly connected device ports based on the link quality detection library and capability archive library of each device port, and performs intelligent scheduling. The link fault scheduling module is used to monitor the dynamic evaluation module for the capability differences of multiple user devices and, in conjunction with the link quality detection library, execute link aging migration strategies.

[0005] As a preferred technical solution of the present invention, the multi-user device capability difference dynamic evaluation module includes an energy level management unit and a fault location and early warning unit. The energy level management unit is used to evaluate the energy level resources of each device port in the park and build a capability archive. The specific steps of the energy level management unit in assessing the energy level resources of each device port within the park and constructing a capability archive are as follows: The tasks sent from various device ports within the park are aggregated, and the parameters of each device port are obtained and stored accordingly. The ratio of the average success rate of each device port to the sum of the average success rates of all device ports, the... Average transmission success rate of each device port The ratio of the maximum and minimum transmission success rates of each device port and the number of... The ratio of the number of port calls to a specific device to the total number of port calls across all devices is used to construct the first... The capability evaluation coefficients of each device port are stored together, and a capability archive is constructed.

[0006] As a preferred embodiment of the present invention, the fault location and early warning unit deploys detection devices at each device port and monitors optical performance and link quality. The specific steps for constructing a link quality detection library for each device port are as follows: Based on the device quality parameters stored in the capability archive, obtain the... The device quality parameters of each device port are calculated, and the first device port is calculated. The fault deviation coefficient of the device port is obtained. The performance parameters of each device port are calculated, and the performance parameters of the first device port are calculated. The performance deviation coefficient of each device port, based on the first... The fault deviation coefficient of each device port and the first The performance deviation coefficients of each device port are weighted to construct the first... Individual device port fault assessment prediction coefficient .

[0007] As a preferred embodiment of the present invention, the fault location and early warning unit in the first... Individual device port fault assessment prediction coefficient Exceeding the set fault assessment prediction threshold Timely warning, determination of the first The first device port is abnormal, and the location of the first one is determined. Each device port simultaneously triggers the resource reclamation policy; When the Individual device port fault assessment prediction coefficient No warning will be issued or the resource recovery strategy will be triggered if the failure assessment and prediction threshold is not exceeded.

[0008] As a preferred embodiment of the present invention, the specific steps of the resource recycling strategy are as follows: The first Individual device port fault assessment prediction coefficient With the set fault assessment prediction threshold Compare values ​​to the first Fault deviation coefficient of individual device ports According to Article Fault deviation coefficient of individual device ports For the first The current resources of each device port are reclaimed, specifically: During each sampling interval, the resources of the first device port are reclaimed. Resource reclamation of individual device ports Until all items are recycled, and the number of items is updated during the iterative recycling process. Individual device port fault assessment prediction coefficient And adjust the next resource recycling process If the update is performed during the iterative recycling process, Individual device port fault assessment prediction coefficient The failure assessment prediction threshold was not exceeded. The recycling process will terminate at the designated time, and the recycled resources will be redistributed to the first... Each device port. This represents the total number of sampling intervals since the resource reclamation policy was triggered.

[0009] As a preferred embodiment of the present invention, the resource scheduling module includes a resource reservation unit and an intelligent scheduling unit. The resource reservation unit reserves initial resources for newly connected device ports based on the link quality detection library and capability archive of each device port. The intelligent scheduling unit performs intelligent scheduling based on the link quality detection library of each device port and the reserved initial resources of each device port.

[0010] As a preferred embodiment of the present invention, the specific steps of the resource reservation unit in reserving initial resources for newly connected device ports based on the link quality detection library and capability archive of each device port are as follows: Match newly connected device ports to the capability archive and construct similarity coefficients. After traversing and calculating, a similar sequence is obtained, and the initial resource corresponding to the most similar device port in the similar sequence is selected as the first initial resource. After removing the device port with the most similar sequence, the top-ranked ports are retained. One device port, and construct a second initial resource. Based on the first initial resource Second initial resources Reserve initial resources for the ports of newly connected devices.

[0011] As a preferred embodiment of the present invention, the intelligent scheduling unit performs intelligent scheduling based on the link quality detection library of each device port and the initial resources reserved for each device port. The specific steps are as follows: Step D1: Obtain the resources and duration required for the task; Step D2: Filter the device ports that meet the conditions of Step D1 from all device ports, and construct a set of device ports. After the construction is complete, proceed to step D3. If the set of device ports cannot be filtered and constructed from all device ports, proceed accordingly. Then proceed to step D4; Step D3: Calculate the first... Evaluation coefficients for each device port The corresponding storage is located in the device port set. Then, collect from the device ports. Select the one with the smallest evaluation coefficient for scheduling; Step D4: Determine whether the data stream of the current task supports parallel transmission. If it does, proceed to step D5. If it does not, terminate and jump to step D7 and wait. Step D5: Under the condition of meeting the task duration, read the remaining resources of all device ports and construct a set of remaining resources for each device port. , and sorted in descending order; Step D6: Set the remaining resources of the device ports The first element of each element is incremented sequentially until the incremented result exceeds the resources required by the task. The current task is then fragmented and matched for transmission. Step D7: Obtain the first The unallocated resources of the device port are determined, and the first one is determined. If the unallocated resources of each device port can meet the resource requirements of the task, skip it until a suitable device port is found. If no suitable device port is found, enter the waiting queue. Step D8: Call the first The unallocated resources of each device port are used to execute tasks, and after the tasks are completed, the unallocated resources are retained for k cycles before being reclaimed.

[0012] As a preferred technical solution of the present invention, the first Evaluation coefficients for each device port The construction process specifically involves: combining the resources required by the task with the first... The remaining resources of each device port are compared, and the first... The device port failure assessment prediction coefficient and the first The capability evaluation coefficients of each device port are weighted and obtained.

[0013] As a preferred embodiment of the present invention, the link fault scheduling module is used to monitor the fault location and early warning unit and, in conjunction with the link quality detection library, execute a link aging migration strategy. The specific steps are as follows: Triggered when the fault location and early warning unit is invoked, and recorded. Time of the first Individual device port fault assessment prediction coefficient When the fault location and early warning unit is called again, it will be recorded. Time of the first Individual device port fault assessment prediction coefficient ,like Less than or equal to If the time is right, then the current judgment will be terminated, and... Time of the first Clear the prediction coefficients for individual device port fault assessments, if Greater than If so, the link aging migration strategy will be triggered; The specific steps of the link aging migration strategy are as follows: mark the port as aging risk, remove the scheduling rights of the port, start the link resource migration mechanism, and migrate the original tasks to other ports.

[0014] Compared with existing technologies, the present invention provides a resource scheduling system based on optical communication networks, which has the following advantages: This invention evaluates the data of each device port in the current park, monitors optical performance and link quality, and constructs a link quality detection library for each device port. Based on the link quality detection library of each device port, it constructs fault assessment and prediction coefficients in real time. When any device port is abnormal, a resource reclamation strategy is triggered. Based on the link quality detection library of each device port and the reserved initial resources of each device port, intelligent scheduling is performed, thereby completing the resource allocation of different ports and realizing the rapid matching of tasks according to the different resources held by the ports. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 A resource scheduling system based on optical communication networks includes: The multi-user device capability difference dynamic evaluation module is used to evaluate the data of each device port in the current park. The multi-user device capability difference dynamic assessment module includes a capability management unit and a fault location and early warning unit. The capability management unit is used to assess the capability resources of each device port in the park and build a capability archive. The fault location and early warning unit deploys detection devices at each device port, monitors optical performance and link quality, and builds a link quality detection library for each device port. Based on the link quality detection library for each device port, it builds fault assessment and prediction coefficients in real time. When any device port is abnormal, a resource reclamation strategy is triggered. The resource scheduling module is used to prioritize the current task and perform resource scheduling in conjunction with the link quality detection library of each device port. The resource scheduling module includes a resource reservation unit and an intelligent scheduling unit. The resource reservation unit reserves initial resources for newly connected device ports based on the link quality detection library and capability archive of each device port. The intelligent scheduling unit performs intelligent scheduling based on the link quality detection library of each device port and the reserved initial resources of each device port. The link fault scheduling module is used to monitor the fault location and early warning unit and, in conjunction with the link quality detection library, execute the link aging migration strategy.

[0018] Furthermore, the specific steps taken by the energy level management unit to assess the energy level resources of various device ports within the park and build a capacity archive are as follows: Step A1: Summarize the tasks sent from each device port in the park, obtain the parameters of each device port, including transmission success rate, bandwidth, etc., and store them for the corresponding device port. Step A2: Construct the first The capability evaluation coefficient for each device port is expressed as follows: in, Indicates the first Average transmission success rate per device port This represents the sum of the average transmission success rates across all device ports. Indicates the total number of device ports. Indicates total Maximum success rate of transmission per device port Indicates the first Number of port calls per device This represents the total number of calls to all device ports. Evaluating the capability of the i-th device port using its capability assessment coefficient ensures that an evaluation source is provided for each device port during subsequent intelligent scheduling. Furthermore, the square root of the sum of the success rates of the i-th device port and its mean effectively reflects the port's capability. A larger value indicates that the i-th port is used more frequently, has a higher success rate, and therefore possesses stronger capability. See Table 1 below for details. Table 1 Capability Evaluation Coefficients for Device Ports Step A3: Combine and store the parameters of each device port with the capability evaluation coefficients of the device port, and build a capability archive.

[0019] Furthermore, the fault location and early warning unit deploys detection equipment at each device port to monitor optical performance and link quality, and the specific steps for building a link quality detection library for each device port are as follows: Step B1: Based on the equipment quality parameters stored in the capability archive, obtain the... The device quality parameters of each device port are calculated, and the first device port is calculated. The fault deviation coefficient for each device port is expressed as follows: in, Indicates the first Individual device port fault deviation coefficient This indicates the standard value of equipment quality deviation. Indicates the first The m-th fault deviation value of a device port. This indicates the total number of equipment quality parameters. See Table 2 below for details. Table 2 only provides examples and does not represent all types of faults. Similar faults can be calculated using the above expression: Table 2 Types of Device Port Faults Step B2: Obtain the first The performance parameters of each device port are calculated, and the performance parameters of the first device port are calculated. The performance deviation coefficient for each device port is expressed as follows: in, Indicates the first Performance deviation coefficient of each device port Indicates the first The device port number Each device's performance parameters Indicates the first Initial values ​​for the performance of each device port. The total number of performance parameters specifically includes: output optical power, receiver sensitivity, wavelength shift, temperature deviation, modulation current deviation, response time deviation, etc., and ensures that the first... The fault deviation coefficient of each device port and the first The consistency of the performance deviation coefficient of each device port is sufficient; Step B3: Based on the first step in step B1 The fault deviation coefficient of each device port and the first The performance deviation coefficients of each device port are comprehensively constructed to form the first... Individual device port fault assessment prediction coefficient The specific expression is as follows: in, , These represent the weight coefficients that sum to 1. The weighted calculation using these two parameters can reflect the... The fault assessment prediction coefficients for each device port are used to estimate the pre-fault situation. When a fault location early warning unit is activated, it is used to locate the fault in the first... Individual device port fault assessment prediction coefficient Exceeding the set fault assessment prediction threshold Early warnings are issued in a timely manner. This threshold can be obtained based on logs stored at the time of fault reporting, and also based on errors reported during historical processes or experiments, to determine the first... The first device port is abnormal, and the location of the first one is determined. In this embodiment, multiple device ports simultaneously trigger the resource reclamation policy. , As shown in Table 3 below: Table 3 Prediction coefficients for device port failure assessment When the Individual device port fault assessment prediction coefficient No warning will be issued or the resource recovery strategy will be triggered if the failure assessment and prediction threshold is not exceeded. The set fault assessment prediction threshold is obtained by acquiring it from historical fault logs. That is, when a fault occurs at a device port, it is recorded in real time, and the corresponding device port fault assessment prediction coefficient is obtained. The average of all the acquired device port fault assessment prediction coefficients is then calculated.

[0020] Furthermore, the specific steps of the resource recycling strategy are as follows: The first Individual device port fault assessment prediction coefficient With the set fault assessment prediction threshold Compare values ​​to the first Fault deviation coefficient of individual device ports The specific expression is as follows: According to the Fault deviation coefficient of individual device ports For the first The current resources of each device port are reclaimed, specifically: During each sampling interval, the resources of the first device port are reclaimed. Resource reclamation of individual device ports Until all items are recycled, and the number of items is updated during the iterative recycling process. Individual device port fault assessment prediction coefficient And adjust the next resource recycling process If the update is performed during the iterative recycling process, Individual device port fault assessment prediction coefficient The failure assessment prediction threshold was not exceeded. The recycling process will terminate at the designated time, and the recycled resources will be redistributed to the first... Each device port. This represents the total number of sampling intervals since the resource reclamation policy was triggered. By incrementally triggering reclamation, it is possible to ensure that bandwidth is quickly allocated to other ports before the ports stabilize, and that the reclaimed resources are redistributed after the ports stabilize, thus ensuring resource utilization. At this point, in this embodiment, the device port 5 fault assessment prediction coefficient in Table 3 is... Exceeding the set fault assessment prediction threshold And calculate the first iteration. At this point, the first iteration reclaims resources for... During the second sampling If the threshold is exceeded, the resources to be recovered in the second iteration are calculated. When a stable trend exists, resource recycling will decrease, and so on until all resources are recycled, or the device port 5 device port failure assessment prediction coefficient. Below the threshold.

[0021] Furthermore, the specific steps by which the resource reservation unit reserves initial resources for newly connected device ports based on the link quality detection library and capability archive for each device port are as follows: Step C1: Match the newly connected device ports to the capability archive and construct a similarity coefficient. After traversing and calculating, a similar sequence is obtained, and the initial resource corresponding to the most similar device port in the similar sequence is selected as the first initial resource. Construct similarity coefficient The specific expression is as follows: in, Represents the similarity coefficient. Indicates the first The device port number Each device's performance parameters Indicates the port number of the newly connected device. Each device's performance parameters This represents the minimum function, where the selected device performance parameters are the same as those constructed above. The performance parameters selected for each device port are the same; Step C2: After removing the device port with the most similar sequence, keep the top-ranked ones. One device port, and construct a second initial resource. The specific expression is as follows: in, Indicates the top ranking The initial resources of each device port are summed. By summing the resources of the top n ports and taking the average, a relatively accurate initial value can be obtained. The similarity of these n different ports is relatively high. Step C3: Based on the first initial resource Second initial resources The initial resources are reserved for the newly connected device ports in the overall construction, and the first initial resources are... Second initial resources Calculating the mean can, to some extent, balance the relationship between the most similar sequence and the relatively similar sequences, thus providing a benchmark value. The specific expression is as follows: in, This indicates that initial resources are reserved for the ports of newly connected devices.

[0022] The intelligent scheduling unit performs intelligent scheduling based on the link quality detection library of each device port and the initial resources reserved for each device port. The specific steps are as follows: Step D1: Obtain the resources and duration required for the task; Step D2: Filter the device ports that meet the conditions of Step D1 from all device ports, and construct a set of device ports. After the construction is complete, proceed to step D3. If the set of device ports cannot be filtered and constructed from all device ports, proceed accordingly. Then proceed to step D4; Step D3: Calculate the first... Evaluation coefficients for each device port The corresponding storage is located in the device port set. Then, collect from the device ports. Select the one with the smallest evaluation coefficient for scheduling; Step D4: Determine whether the data stream of the current task supports parallel transmission. If it does, proceed to step D5. If it does not, terminate and jump to step D7 and wait. Step D5: Under the condition of meeting the task duration, read the remaining resources of all device ports and construct a set of remaining resources for each device port. , and sorted in descending order; Step D6: Set the remaining resources of the device ports The first element of each element is incremented sequentially until the incremented result exceeds the resources required by the task. The current task is then fragmented and matched for transmission. Step D7: Obtain the first The unallocated resources of the device port are determined, and the first one is determined. If the unallocated resources of each device port can meet the resource requirements of the task, skip it until a suitable device port is found. If no suitable device port is found, enter the waiting queue. Step D8: Call the first Unallocated resources on each device port are used to execute tasks, and after the tasks are completed, the unallocated resources are retained for k cycles before being reclaimed. If, after k cycles, the unallocated resources are... If the resources added to a device port are still being utilized, then adjust the... Each device port x new initial resources ,in This indicates the kth period within the kth period. Resources added to each device port This represents a balance coefficient greater than 1; In addition, the intelligent scheduling unit can manually reschedule and redistribute resources that have been idle for a long time according to the needs of the management personnel, which will not be elaborated on here.

[0023] Furthermore, the first Evaluation coefficients for each device port The specific expression is as follows: in, Indicates the first Individual device port fault assessment prediction coefficients Indicates the resources required for the task. Indicates the first Remaining resources for each device port , These represent the weight coefficients that sum to 1, and are combined with... This reflects that the larger the ratio of remaining resources to task resources, the worse the matching is. This results in tasks with smaller resource requirements occupying a large amount of remaining resources, forcing subsequent tasks that require more resources to be processed in segments or enter a waiting queue, which may lead to congestion.

[0024] The link fault scheduling module is used to monitor the fault location and early warning unit and, in conjunction with the link quality detection library, execute the link aging migration strategy. The specific steps are as follows: Triggered when the fault location and early warning unit is invoked, and recorded. Time of the first Individual device port fault assessment prediction coefficient When the fault location and early warning unit is called again, it will be recorded. Time of the first Individual device port fault assessment prediction coefficient ,like Less than or equal to If the time is right, then the current judgment will be terminated, and... Time of the first Clear the prediction coefficients for individual device port fault assessments, if Greater than This will trigger the link aging migration strategy.

[0025] Furthermore, the specific steps of the link aging migration strategy are as follows: mark the port as aging risk, remove the scheduling rights of the port, start the link resource migration mechanism, and migrate the original tasks to other ports.

[0026] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A resource scheduling system based on optical communication networks, characterized in that: include: The multi-user device capability difference dynamic assessment module assesses the capability resources of each device port in the park and builds a capability archive. It deploys detection equipment at each device port to monitor optical performance and link quality, and builds a link quality detection library for each device port. Based on the link quality detection library for each device port, it builds a fault assessment prediction coefficient in real time. When any device port is abnormal, a resource reclamation strategy is triggered. The resource scheduling module reserves initial resources for newly connected device ports based on the link quality detection library and capability archive library of each device port, and performs intelligent scheduling. The link fault scheduling module is used to monitor the dynamic evaluation module for the capability differences of multiple user devices and, in conjunction with the link quality detection library, execute link aging migration strategies.

2. The resource scheduling system based on optical communication networks according to claim 1, characterized in that: The multi-user device capability difference dynamic assessment module includes a capability level management unit and a fault location and early warning unit. The capability level management unit is used to assess the capability level resources of each device port in the park and build a capability archive. The specific steps of the energy level management unit in assessing the energy level resources of each device port within the park and constructing a capability archive are as follows: The tasks sent from various device ports within the park are aggregated, and the parameters of each device port are obtained and stored accordingly. The ratio of the average success rate of each device port to the sum of the average success rates of all device ports, the... Average transmission success rate of each device port The ratio of the maximum and minimum transmission success rates of each device port and the number of... The ratio of the number of port calls to a specific device to the total number of port calls across all devices is used to construct the first... The capability evaluation coefficients of each device port are stored together, and a capability archive is constructed.

3. The resource scheduling system based on optical communication networks according to claim 2, characterized in that: The fault location and early warning unit deploys detection devices at each device port to monitor optical performance and link quality, and constructs a link quality detection library for each device port. The specific steps are as follows: Based on the device quality parameters stored in the capability archive, obtain the... The device quality parameters of each device port are calculated, and the first device port is calculated. The fault deviation coefficient of the device port is obtained. The performance parameters of each device port are calculated, and the first... The performance deviation coefficient of each device port, based on the first... The fault deviation coefficient of each device port and the first The performance deviation coefficients of each device port are weighted to construct the first... Individual device port fault assessment prediction coefficient .

4. The resource scheduling system based on optical communication networks according to claim 3, characterized in that: The fault location and early warning unit in the first Individual device port fault assessment prediction coefficient Exceeding the set fault assessment prediction threshold Timely warning, determination of the first The first device port is abnormal, and the location of the first one is determined. Each device port simultaneously triggers the resource reclamation policy; When the Individual device port fault assessment prediction coefficient No warning will be issued or the resource recovery strategy will be triggered if the failure assessment and prediction threshold is not exceeded.

5. The resource scheduling system based on optical communication networks according to claim 4, characterized in that: The specific steps of the resource recycling strategy are as follows: The first Individual device port fault assessment prediction coefficient With the set fault assessment prediction threshold Compare values ​​to the first Fault deviation coefficient of individual device ports According to Article Fault deviation coefficient of individual device ports For the The current resources of each device port are reclaimed, specifically: During each sampling interval, the resources of the first device port are reclaimed. Resource reclamation of individual device ports Until all items are recycled, and the number of items is updated during the iterative recycling process. Individual device port fault assessment prediction coefficient And adjust the next resource recycling process If the update is performed during the iterative recycling process... Individual device port fault assessment prediction coefficient The failure assessment prediction threshold was not exceeded. The recycling process will terminate at the designated time, and the recycled resources will be redistributed to the first... Each device port. This represents the total number of sampling intervals since the resource reclamation policy was triggered.

6. The resource scheduling system based on optical communication networks according to claim 4, characterized in that: The resource scheduling module includes a resource reservation unit and an intelligent scheduling unit. The resource reservation unit reserves initial resources for newly connected device ports based on the link quality detection library and capability archive of each device port. The intelligent scheduling unit performs intelligent scheduling based on the link quality detection library of each device port and the reserved initial resources of each device port.

7. The resource scheduling system based on optical communication networks according to claim 6, characterized in that: The specific steps by which the resource reservation unit reserves initial resources for newly connected device ports based on the link quality detection library and capability archive for each device port are as follows: Match newly connected device ports to the capability archive and construct similarity coefficients. After traversing and calculating, a similar sequence is obtained, and the initial resource corresponding to the most similar device port in the similar sequence is selected as the first initial resource. After removing the device port with the most similar sequence, the top-ranked ports are retained. One device port, and construct a second initial resource. Based on the first initial resource Second initial resources Reserve initial resources for the ports of newly connected devices.

8. The resource scheduling system based on optical communication networks according to claim 6, characterized in that: The intelligent scheduling unit performs intelligent scheduling based on the link quality detection library of each device port and the initial resources reserved for each device port. The specific steps are as follows: Step D1: Obtain the resources and duration required for the task; Step D2: Filter the device ports that meet the conditions of Step D1 from all device ports, and construct a set of device ports. After the construction is complete, proceed to step D3. If the set of device ports cannot be filtered and constructed from all device ports, proceed accordingly. Then proceed to step D4; Step D3: Calculate the first... Evaluation coefficients for each device port The corresponding storage is located in the device port set. Then, collect from the device port. Select the one with the smallest evaluation coefficient for scheduling; Step D4: Determine whether the data stream of the current task supports parallel transmission. If it does, proceed to step D5. If it does not, terminate and jump to step D7 and wait. Step D5: Under the condition of meeting the task duration, read the remaining resources of all device ports and construct a set of remaining resources for each device port. , and sorted in descending order; Step D6: Set the remaining resources of the device ports The first element of each element is incremented sequentially until the incremented result exceeds the resources required by the task. The current task is then fragmented and matched for transmission. Step D7: Obtain the first The unallocated resources of the device port are determined, and the first one is determined. If the unallocated resources of each device port can meet the resource requirements of the task, skip it until a suitable device port is found. If no suitable device port is found, enter the waiting queue. Step D8: Call the first The unallocated resources of each device port are used to execute tasks, and after the tasks are completed, the unallocated resources are retained for k cycles before being reclaimed.

9. The resource scheduling system based on optical communication networks according to claim 8, characterized in that: The first Evaluation coefficients for each device port The construction process specifically involves: combining the resources required by the task with the first... The remaining resources of each device port are compared, and the first... The device port failure assessment prediction coefficient and the first The capability evaluation coefficients of each device port are weighted and obtained.

10. The resource scheduling system based on optical communication networks according to claim 1, characterized in that: The link fault scheduling module is used to monitor the fault location and early warning unit and, in conjunction with the link quality detection library, execute the link aging migration strategy. The specific steps are as follows: Triggered when the fault location and early warning unit is invoked, and recorded. Time of the first Individual device port fault assessment prediction coefficient When the fault location and early warning unit is called again, it will be recorded. Time of the first Individual device port fault assessment prediction coefficient ,like Less than or equal to If the time is right, then the current judgment will be terminated, and... Time of the first Clear the prediction coefficients for individual device port fault assessments, if Greater than If so, the link aging migration strategy will be triggered; The specific steps of the link aging migration strategy are as follows: mark the port as aging risk, remove the scheduling rights of the port, start the link resource migration mechanism, and migrate the original tasks to other ports.

Images