High-parallel test method of optical comprehensive test platform

Abstract

The application provides a high-parallel test method of an optical comprehensive test platform, relates to the field of active optical testing, realizes parallel test of multiple measured objects or multiple test items through N independent test channels, avoids waiting time of traditional serial test, effectively improves manufacturing efficiency in combination with the high-parallel characteristics of the platform, dynamically allocates tasks based on real-time monitoring of channel resource states, avoids idle resources, cooperates with the high-integration characteristics of the platform, multiple modules share the platform, effectively shares costs, improves overall utilization of the equipment, the optical power meter, the adjustable optical attenuator and other modules integrated in each channel of the hardware architecture support hot plugging, the task analysis algorithm can flexibly map test items, adapts to different measured object requirements, conforms to the advantages of on-demand matching and easy expansion of the platform, can protect investment and adapt to dynamic changes of test scenes, a distributed data acquisition mechanism synchronously acquires channel data, and provides support for generating reliable test reports.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of active optical testing, in particular to a high-parallel testing method of an optical comprehensive testing platform. BACKGROUND

[0002] With the rapid development of the Internet and broadband services, active optical modules are evolving towards high speed and multi-channel, which puts forward higher requirements on testing efficiency. The traditional testing method adopts a serial testing mode, that is, a single testing channel completes the testing of multiple indicators of the device under test one by one, which has problems such as long testing period and low equipment utilization.

[0003] Although the existing optical testing platform has certain parallel testing capability, it has the following limitations:

[0004] The hardware resource allocation is fixed and cannot be dynamically adjusted according to the testing task;

[0005] The software algorithm does not realize real parallel scheduling, and there is data conflict during multi-channel testing;

[0006] The centralized architecture is used for data processing, which is difficult to meet the throughput demand of large-scale parallel testing.

[0007] Therefore, there is an urgent need for a high-parallel testing method that can realize efficient resource scheduling and synchronous data processing to meet the efficiency requirements of optical module mass production scenarios. SUMMARY

[0008] The present application provides a high-parallel testing method of an optical comprehensive testing platform to solve the problems raised in the background art.

[0009] A high-parallel testing method of an optical comprehensive testing platform, comprising:

[0010] S1: obtaining a multi-channel parallel testing hardware architecture of the optical comprehensive testing platform, wherein the parallel testing hardware architecture comprises N independent testing channels, each independent testing channel integrating an optical power meter, an adjustable optical attenuator, an optical switch and a bit error rate instrument module;

[0011] S2: generating parallel testing items based on the object under test, and mapping the parallel testing items to the N independent testing channels based on a task analysis algorithm;

[0012] S3: real-time monitoring the resource state of the N independent testing channels, and dynamically scheduling the resources based on the resource state;

[0013] S4: synchronously obtaining the testing data of each channel based on a distributed data acquisition mechanism, and generating a testing report based on the testing data of each channel.

[0014] Preferably, in the S1, each independent test channel integrates an optical power meter, an adjustable optical attenuator, an optical switch and a bit error rate instrument module, which are specifically used for:

[0015] The optical power meter is used for testing the optical power of the measured object;

[0016] The adjustable optical attenuator is used for testing the attenuation characteristics of the measured object;

[0017] The optical switch is used for realizing fast switching between independent test channels;

[0018] The bit error rate instrument module is used for generating a standard test signal for the measured object;

[0019] Each independent test channel realizes synchronous control through a backplane bus.

[0020] Preferably, in the S2, parallel test items are generated based on the measured object, including:

[0021] The type and nominal parameters of the measured object are determined from the core information of the measured object;

[0022] Based on the type and nominal parameters of the measured object, test indicators are determined from industry standards and user requirements;

[0023] Based on user requirements, special test requirements are determined;

[0024] Based on the test indicators and special test requirements, parallel test items are established.

[0025] Preferably, in the S2, the parallel test items are mapped to N independent test channels based on a task analysis algorithm, including:

[0026] The parallel test items are analyzed to obtain a plurality of sub-tasks containing one test indicator;

[0027] Based on resource independence and time sequence dependence, the sub-tasks are comprehensively analyzed to determine the parallel characteristics and time sequence characteristics between the sub-tasks, and the sub-tasks are parallel marked and time sequence marked based on the parallel characteristics and time sequence characteristics;

[0028] The test duration of the test indicators of the sub-tasks is obtained, and the sub-tasks are combined and packaged based on the parallel marking and time sequence marking to obtain balanced task packages, and the balanced task packages are optimized based on special test requirements to obtain target balanced task packages;

[0029] Based on the time sequence marking, the target balanced task packages are prioritized to obtain the priority order of each target balanced task package, based on user requirements, the priority weight of each target balanced task package is determined, and based on the priority order and priority weight, the target priority of the target balanced task package is determined;

[0030] Obtaining historical test data completed by N independent test channels, determining an optimal test index sequence of each independent test channel based on a relationship between the independent test channel and the historical test data;

[0031] Obtaining a resource occupation state of each independent test channel, and establishing an intelligent mapping model based on the resource occupation state and the optimal test index sequence, with a goal of maximizing parallel efficiency and minimizing resource conflicts;

[0032] Mapping a target balanced task package to the N independent test channels based on the intelligent mapping model.

[0033] Preferably, the method further comprises periodically optimizing the intelligent mapping model, specifically:

[0034] Obtaining a task mapping result based on the intelligent mapping model, and determining a mapping effect corresponding to the task mapping result;

[0035] When the mapping effect meets a preset effect requirement, rewarding the intelligent mapping model for this mapping, otherwise, punishing the intelligent mapping model for this mapping;

[0036] Optimizing the intelligent mapping model based on the reward and the punishment result, to obtain a latest intelligent mapping model.

[0037] Preferably, in the S3, the resource state of the N independent test channels is monitored in real time, including:

[0038] Real-time monitoring of a module occupation state of each independent test channel when testing a parallel test item;

[0039] Generating a resource occupation table based on the module occupation state.

[0040] Preferably, in the S3, dynamic resource scheduling is performed based on the resource state, including:

[0041] Timestamp aligning the resource state of each independent test channel to obtain synchronized resource occupation information, and determining a load condition of each independent test channel based on the synchronized resource occupation information, and establishing a dynamic load migration strategy in combination with a constraint that an index difference of a load of each channel is within ±15%;

[0042] Obtaining a resource scheduling log of the independent test channel in a preset historical time period, and determining a test preference of the independent test channel based on the resource scheduling log;

[0043] Establishing an initial resource scheduling strategy based on the test preference and the dynamic load migration strategy;

[0044] Scheduling resources for N independent test channels according to the initial resource scheduling strategy, and determining whether there is a resource conflict after completing the resource scheduling once;

[0045] If yes, performing secondary resource scheduling;

[0046] Otherwise, completing the test according to the current resource state distribution;

[0047] The secondary resource scheduling specifically includes:

[0048] Determining whether the resource conflict can be replaced by other hardware, and if yes, performing resource scheduling based on the other hardware replacement;

[0049] Otherwise, determining whether the current task can be split and reorganized, and if yes, performing resource scheduling based on the initial resource scheduling and / or the secondary resource scheduling after splitting and reorganizing the current task to complete the test;

[0050] Otherwise, performing resource simplification on the compressive non-critical steps of the current task to obtain a simplified task, and then performing resource scheduling based on the initial resource scheduling and / or the secondary resource scheduling to complete the test.

[0051] Preferably, the initial resource scheduling strategy is established based on the test preference and the dynamic load migration strategy, including:

[0052] Establishing a scheduling priority of the resource scheduling to the corresponding independent test channel based on the test preference;

[0053] Determining the resource schedulable condition based on the dynamic load migration strategy;

[0054] Establishing the initial resource scheduling strategy based on the resource schedulable condition and the scheduling priority.

[0055] Preferably, in the S4, the test data of each channel is synchronously obtained based on a distributed data acquisition mechanism, including:

[0056] A high-speed serial transmission interface is configured for each independent test channel, and an acquisition instruction is sent through a backplane bus;

[0057] Based on the acquisition instruction, the collected data is transmitted through the high-speed serial transmission interface to obtain the test data of each channel.

[0058] Preferably, in the S4, a test report is generated based on the test data of each channel, including:

[0059] The test data of each channel is preprocessed to obtain target test data;

[0060] The target test data is divided and processed according to the measured objects and test items to obtain object test reports and item test reports;

[0061] The object test report and the project test report are integrated to generate a table of contents, and the final test report is obtained based on the table of contents, the object test report and the project test report.

[0062] Compared with the prior art, the present application has the following beneficial effects:

[0063] Parallel testing of multiple testees or multiple test projects is realized through N independent test channels, avoiding the waiting time of traditional serial testing, and combining the high parallelism of the platform, the manufacturing efficiency can be effectively improved, especially suitable for mass production testing scenarios under the scale growth of active optical modules, dynamic resource scheduling allocates tasks based on the real-time monitoring of channel resource states, avoiding resource idling, and cooperating with the high integration characteristics of the platform, multiple modules share the platform, effectively sharing the cost and improving the overall utilization of the equipment, the optical power meter, the adjustable optical attenuator and other modules integrated in the hardware architecture support hot plugging, and the task analysis algorithm can flexibly map the test projects to adapt to the needs of different testees, conforming to the advantages of platform on-demand matching and easy expansion, protecting investment and adapting to the dynamic changes of the testing scene, the distributed data acquisition mechanism synchronously acquires channel data, and combining the stable hardware performance of the platform, the accuracy of the test data is ensured, providing support for generating reliable test reports.

[0064] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and achieved by the structure specifically pointed out in the present application document.

[0065] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0066] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, used to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0067] Figure 1 The flowchart of the high parallel testing method of the optical comprehensive testing platform in the embodiment of the present application is shown in the figure.

[0068] Figure 2 The flowchart of obtaining channel test data in the embodiment of the present application is shown in the figure.

[0069] Figure 3 The flowchart of generating a test report in the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0070] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, in which it is understood that the preferred embodiments described below merely for the purpose of illustrating and explaining the present application, and are not intended to limit the present application.

[0071] Embodiment 1

[0072] The embodiment of the present application provides a high-parallel test method of an optical comprehensive test platform, as shown in the figure, comprising: Figure 1

[0073] S1: acquiring a multi-channel parallel test hardware architecture of the optical comprehensive test platform, wherein the parallel test hardware architecture comprises N independent test channels, each independent test channel integrating an optical power meter, an adjustable optical attenuator, an optical switch and a bit error rate instrument module;

[0074] S2: generating a parallel test item based on a measured object, and mapping the parallel test item to the N independent test channels based on a task analysis algorithm;

[0075] S3: performing real-time monitoring on resource states of the N independent test channels, and performing dynamic resource scheduling based on the resource states;

[0076] S4: synchronously acquiring test data of each channel based on a distributed data acquisition mechanism, and generating a test report based on the test data of each channel.

[0077] In this embodiment, the parallel test hardware architecture specifically comprises 16 independent test channels, each channel integrating an optical power meter, an adjustable optical attenuator, a 2*2 optical switch and a bit error rate instrument module; a main control module adopts an ARM processor+Linux operating system, supports communication interfaces such as an RJ-45 LAN and a USB2.0; a backplane bus adopts a high-speed serial interface, realizes synchronous control of each channel, and the synchronous precision is less than or equal to 100ns; supports a hot plug function, can flexibly increase or decrease the number of channels according to test requirements, and the maximum attenuation value of a single channel can reach 60dB.

[0078] In this embodiment, the parallel test item comprises requirements for items such as optical power, extinction ratio and bit error rate.

[0079] ​The beneficial effects of the above design scheme are: the parallel testing of multiple measured objects or multiple test items is realized through N independent test channels, the waiting time of traditional serial testing is avoided, the manufacturing efficiency can be effectively improved in combination with the high parallelism of the platform, especially suitable for mass production testing scenarios under the scale growth of active optical modules, dynamic resource scheduling allocates tasks based on the real-time monitoring of channel resource states, avoids resource idling, and cooperates with the high integration characteristics of the platform, multiple modules share the platform, effectively share the cost, and improve the overall utilization of the equipment, the optical power meter, the adjustable optical attenuator and other modules integrated in each channel of the hardware architecture support hot plugging, and the task analysis algorithm can flexibly map test items to adapt to the needs of different measured objects, conforming to the advantages of on-demand matching and easy expansion of the platform, protecting investment and adapting to the dynamic changes of the testing scenario, the distributed data acquisition mechanism synchronously acquires channel data, and in combination with the stable hardware performance of the platform, the accuracy of the test data is ensured to provide support for generating reliable test reports.

[0080] Embodiment 2

[0081] Based on the basis of embodiment 1, the embodiment of the application provides a high-parallel testing method of an optical comprehensive testing platform, in S1, the optical power meter, the adjustable optical attenuator, the optical switch and the error code instrument module integrated in each independent test channel are specifically used for:

[0082] The optical power meter is used for testing the optical power of the measured object;

[0083] The adjustable optical attenuator is used for testing the attenuation characteristics of the measured object;

[0084] The optical switch is used for realizing the rapid switching between the independent test channels;

[0085] The error code instrument module is used for generating a standard detection signal for the measured object;

[0086] Each independent test channel realizes synchronous control through a backplane bus.

[0087] The beneficial effects of the above design scheme are: the functions of each module are designed to be specialized, avoiding the loss of test precision caused by the multifunctional reuse of general modules, and ensuring the measurement accuracy of each test index.

[0088] Embodiment 3

[0089] Based on the basis of embodiment 1, the embodiment of the application provides a high-parallel testing method of an optical comprehensive testing platform, in S2, the parallel test items are generated based on the measured objects, including:

[0090] The type and nominal parameter of the measured object are determined from the core information of the measured object;

[0091] The test index is determined from the industry standard and the user demand based on the type and nominal parameter of the measured object;

[0092] determine special test requirements based on user requirements;

[0093] establish parallel test projects based on the test indicators and special test requirements.

[0094] In this embodiment, the type of the object under test is, for example, an optical fiber link, an optical chip, etc., and the nominal parameters are, for example, operating wavelength, transmission rate, interface type, rated optical power range, etc.

[0095] In this embodiment, the test indicators include optical power, extinction ratio, eye diagram, bit error rate, attenuation characteristics, temperature stability, etc.

[0096] In this embodiment, the special test requirements are, for example, test temperature range, signal modulation method, minimum / maximum input optical power, etc.

[0097] The above design scheme has the beneficial effects that: by extracting the type and nominal parameters of the object under test, the generated test project directly matches the core characteristics of the object under test, avoiding the redundancy of irrelevant test items, reducing invalid test operations, improving test efficiency while ensuring the relevance of the test results to the actual performance of the object under test, through the systematic test project generation logic, ensuring test compliance and accuracy while providing a scientific and executable task basis for high-parallel testing, effectively supporting efficient implementation of multi-channel parallel testing.

[0098] Embodiment 4:

[0099] Based on the basis of Embodiment 1, the present embodiment provides a high-parallel test method of an optical comprehensive test platform, and in S2, the parallel test project is mapped to N independent test channels based on a task analysis algorithm, comprising:

[0100] The parallel test project is analyzed to obtain a plurality of sub-tasks containing one test indicator;

[0101] Based on resource independence and time sequence dependence, the sub-tasks are comprehensively analyzed to determine the parallel characteristics and time sequence characteristics between the sub-tasks, and the sub-tasks are parallel marked and time sequence marked based on the parallel characteristics and time sequence characteristics;

[0102] The test duration of the test indicator of the sub-task is obtained, the sub-tasks are combined and packaged in combination with the parallel marking and the time sequence marking to obtain a balanced task package, and the balanced task package is optimized based on the special test requirements to obtain a target balanced task package;

[0103] Based on the time sequence mark, the target balancing task package is prioritized to obtain a priority order of each target balancing task package, a priority weight of each target balancing task package is determined based on user demand, and a target priority of the target balancing task package is determined based on the priority order and the priority weight.

[0104] The historical test data completed by the N independent test channels is acquired, and the optimal test index sequence of each independent test channel is determined based on the relationship between the independent test channel and the historical test data.

[0105] The resource occupation state of each independent test channel is acquired, and an intelligent mapping model is established based on the resource occupation state and the optimal test index sequence, with the goal of maximizing parallel efficiency and minimizing resource conflicts.

[0106] The target balancing task package is mapped into the N independent test channels based on the intelligent mapping model.

[0107] In this embodiment, resource independence analysis is that if two test items do not need to share the same hardware module, such as optical power test relying on optical power meter and bit error rate test relying on bit error instrument, they can be parallel; if they share the same module, such as the same optical switch, they need to be staggered through a scheduling algorithm.

[0108] In this embodiment, the time sequence dependency is, for example, that there is a dependency between test items, such as attenuation characteristic test needing to set the attenuation value through an adjustable optical attenuator first and then performing the receiving sensitivity test, which is split into a pre-task and a parallel task, the pre-task is executed in series, and the subsequent task is parallel in different channels.

[0109] In this embodiment, the sub-tasks are combined and packaged to obtain the balancing task package, which is specifically to combine the tasks into a balanced task package according to the time consumption of the test items, such as bit error rate test consuming longer time and optical power test consuming shorter time, to avoid overloading of a certain channel.

[0110] In this embodiment, the balancing task package is optimized based on special test requirements, which arranges the same special test requirements in the same task package.

[0111] In this embodiment, the intelligent mapping model specifies the modules required to be called by each test item, ensuring that the task matches the hardware capability of the channel.

[0112] In this embodiment, the priority order determines the order that must be followed between the target balancing task packages, and the order corresponding to the same level can be finally determined according to the priority weight.

[0113] The beneficial effects of the above design scheme are: through analyzing and testing the parallel project to obtain a single index subtask, and based on resource independence and time sequence dependence, marking parallel and time sequence characteristics, it can be clearly defined which subtasks can be executed synchronously and which subtasks have a prior dependence, pseudo-parallelism is avoided from the bottom layer, the utilization rate of truly parallel subtasks is improved to more than 90%, and the foundation for high parallel testing is laid. Combined with the test length, parallel / time sequence marking, the subtask combination is packaged, long-time-consuming tasks such as error rate testing and short-time-consuming tasks such as optical power testing are cross-combined, the load backlog of a single channel caused by concentrating on processing long-time-consuming tasks is avoided, the total time consumption difference of each channel is controlled within 10%, the overall parallel efficiency of N channels is significantly improved, based on special test requirements such as stability test under extreme temperature and self-defined modulation mode test, the balanced task package is optimized, the task combination logic can be adjusted specifically, such as concentrating the tasks that need to share the oven resources to adjacent channels to avoid the adaptation limitation of general packaging logic to special scenarios, improve the scene coverage capability of the test system, determine the basic priority order through the time sequence marking, and give priority weight combined with user demand, the final target priority can guarantee the correctness of the test logic and meet the priority test demand of the user on the core index, avoid the key task being blocked by the low priority task, determine the optimal test index sequence of each channel based on the historical test data, which can accurately match the task package with the hardware capability of the channel, reduce the test error caused by the mismatch between the channel and the task, and improve the efficiency of the channel special test, the intelligent mapping model aims to maximize the parallel efficiency and minimize the resource conflict, dynamically allocates the task package combined with the real-time resource occupation state, which can avoid the conflict of multiple tasks competing for the same module through pre-judgment, and can improve the channel resource utilization rate to more than 95% through dynamic adjustment, which significantly shortens the overall test cycle.

[0114] Embodiment 5:

[0115] Based on the basis of embodiment 4, the embodiment of the application provides a high parallel test method of an optical comprehensive test platform, further comprising: periodically optimizing the intelligent mapping model, specifically:

[0116] Obtaining the task mapping result obtained based on the intelligent mapping model, and determining the mapping effect corresponding to the task mapping result;

[0117] When the mapping effect meets the preset effect demand, rewarding the intelligent mapping model for this mapping, otherwise, punishing the intelligent mapping model for this mapping;

[0118] Based on the reward and punishment results, the intelligent mapping model is optimized to obtain the latest intelligent mapping model.

[0119] The beneficial effects of the above design scheme are: through periodic optimization of the intelligent mapping model, the latest intelligent mapping model obtained can adaptively change according to actual requirements, and the accuracy of mapping the parallel test project to the N independent test channels is ensured.

[0120] Embodiment 6:

[0121] Based on the basis of embodiment 1, the embodiment of the application provides a high-parallel test method of an optical comprehensive test platform, in S3, resource state real-time monitoring is performed on the N independent test channels, including:

[0122] Real-time monitoring of module occupation of each independent test channel when testing the parallel test project;

[0123] Generating a resource occupation table based on the module occupation.

[0124] The beneficial effects of the above design scheme are: through the combination of real-time monitoring and the resource occupation table, dynamic perception of the resource state is realized, and data support is provided for efficient scheduling, smooth operation of multi-channel parallel testing is ensured from the two dimensions of information transparency and conflict avoidance, and the resource utilization rate and test efficiency of the optical comprehensive test platform are significantly improved.

[0125] Embodiment 7:

[0126] Based on the basis of embodiment 1, the embodiment of the application provides a high-parallel test method of an optical comprehensive test platform, in S3, dynamic resource scheduling is performed based on the resource state, including:

[0127] Timestamp alignment is performed on the resource state of each independent test channel to obtain synchronous resource occupation information, and the load condition of each independent test channel is determined based on the synchronous resource occupation information, a dynamic load migration strategy is established in combination with the constraint that the load index difference of each channel is allowed to be within ±15%, and the dynamic load migration strategy is established based on the test preference and the dynamic load migration strategy;

[0128] Obtaining a resource scheduling log of the independent test channel in a preset historical time period, determining the test preference of the independent test channel based on the resource scheduling log;

[0129] Establishing an initial resource scheduling strategy based on the test preference and the dynamic load migration strategy;

[0130] Performing resource scheduling on the N independent test channels according to the initial resource scheduling strategy, and determining whether there is a resource conflict condition after completing the resource scheduling once;

[0131] If yes, performing secondary resource scheduling;

[0132] Otherwise, completing the test according to the current resource state distribution;

[0133] The secondary resource scheduling is specifically:

[0134] It is judged whether the resource conflict can be replaced by other hardware. If yes, resource scheduling is performed based on the replacement of other hardware;

[0135] Otherwise, it is judged whether the current task can be split and reorganized. If yes, the current task is split and reorganized, and then resource scheduling is performed again based on the primary resource scheduling and / or the secondary resource scheduling to complete the test;

[0136] Otherwise, the current task is subjected to resource simplification by compression non-critical steps, and then resource scheduling is performed again based on the primary resource scheduling and / or the secondary resource scheduling to complete the test.

[0137] In this embodiment, the dynamic load migration strategy is to identify low-priority tasks that can be migrated, and migrate them to a light-load channel. During the migration process, the parameters are tested synchronously through the backplane bus to ensure data consistency.

[0138] In this embodiment, the test preference of the independent test channel is determined, such as the completion time of a certain type of task in channel 3 is 20% longer than in other channels. The scheduling preference is updated through a reinforcement learning model, such as reducing the allocation of this type of task to channel 3.

[0139] In this embodiment, when there is still a resource conflict after the primary resource scheduling of the N independent test channels according to the initial resource scheduling strategy, secondary resource scheduling is introduced.

[0140] In this embodiment, the resource simplification by compression non-critical steps, for example, reduces the optical power sampling rate from 100 times per second to 50 times per second, which does not affect the result but shortens the time consumption.

[0141] In this embodiment, task split and reorganization, for example, a certain task must use a specific wavelength error code instrument, which is split into two parts: pre-preparation and core test. The pre-preparation, such as parameter configuration, is executed in the current channel, and the core test waits for resource release, reducing the invalid waiting time.

[0142] The beneficial effects of the above design scheme are: by timestamping the resource state of each independent test channel, the time reference of the synchronous resource occupation information is ensured to be consistent, the load of each channel is calculated based on the synchronous resource occupation information, and the dynamic load migration strategy is established with the constraint of allowing the load index difference to be ± 15%, which can effectively avoid the problems of task accumulation in part of the channel and long-term idle of part of the channel in traditional scheduling, and by analyzing the resource scheduling log in the preset historical time period to determine the test preference of the channel, the initial resource scheduling strategy can preferentially allocate the matched test task to the channel, fully exert the special advantages of each channel, improve the single-channel test efficiency, and the secondary resource scheduling realizes efficient resolution of resource conflicts through the hierarchical strategy of hardware replacement, task splitting and reorganization, and compression of non-critical steps. The scheme maximizes the resource potential of the multi-channel parallel architecture on the basis of ensuring test accuracy and continuity, significantly improves the overall test efficiency and scene adaptation capability of the optical comprehensive test platform, and is especially suitable for complex scheduling scenes in large-scale optical module mass production testing.

[0143] Embodiment 8:

[0144] Based on the basis of embodiment 7, the embodiment of the application provides a high-parallel test method of an optical comprehensive test platform, and an initial resource scheduling strategy is established based on the test preference and the dynamic load migration strategy, including:

[0145] Establishing a scheduling priority of resource scheduling to the corresponding independent test channel based on the test preference;

[0146] Determining resource schedulability based on the dynamic load migration strategy;

[0147] Establishing an initial resource scheduling strategy based on the resource schedulability and the scheduling priority.

[0148] The beneficial effects of the above design scheme are: by the cooperative design of preference direction and load control boundary, the initial resource scheduling strategy realizes organic unification in aspects of exerting hardware characteristics, guaranteeing global efficiency, adapting to scene changes, etc., which not only improves the quality of initial decision, but also lays a foundation for subsequent process lightweight, and significantly enhances the overall performance of resource scheduling in high-parallel test.

[0149] Embodiment 9:

[0150] Based on the basis of embodiment 1, the embodiment of the application provides a high-parallel test method of an optical comprehensive test platform, as shown in Figure 3 In S4, the test data of each channel is synchronously acquired based on a distributed data acquisition mechanism, including:

[0151] A high-speed serial transmission interface is configured for each independent test channel, and a collection instruction is sent through a backplane bus;

[0152] Based on the acquisition instruction, the collected data is transmitted through the high-speed serial transmission interface to obtain the test data of each channel.

[0153] The beneficial effects of the above design scheme are that the distributed data acquisition mechanism synchronously acquires the data of each channel, the stable hardware performance of the platform is combined to ensure the accuracy of the test data, and support is provided for generating a reliable test report.

[0154] Embodiment 10:

[0155] Based on the basis of embodiment 1, the embodiment of the application provides a high-parallel test method of an optical comprehensive test platform, as shown in Figure 3 In S4, a test report is generated based on the test data of each channel, including:

[0156] The test data of each channel is preprocessed to obtain target test data.

[0157] The target test data is divided and processed according to the measured objects and test items to obtain object test reports and item test reports.

[0158] The object test reports and the item test reports are integrated to generate a directory table, and the final test report is obtained based on the directory table, the object test reports and the item test reports.

[0159] The beneficial effects of the above design scheme are that through the design of preprocessing for accuracy, grouping for needs, integration for efficiency, and structure for traceability, the test report can accurately reflect the results of multi-channel parallel test and can be flexibly adapted to different user scenarios, thereby providing an efficient and reliable information carrier for the production screening, research and development optimization and quality control of optical modules.

[0160] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application also intends to include these modifications and variations.

Claims

1. A highly parallel testing method for an optical synthesis test platform, characterized in that, include: S1: Obtain the multi-channel parallel test hardware architecture of the optical integrated test platform. The parallel test hardware architecture includes N independent test channels. Each independent test channel integrates an optical power meter, an adjustable optical attenuator, an optical switch and a bit error rate tester module. S2: Generate parallel test projects based on the object under test, and map the parallel test projects to N independent test channels based on the task parsing algorithm; S3: Real-time monitoring of resource status for N independent test channels, and dynamic resource scheduling based on resource status, including: The resource status of each independent test channel is timestamped to obtain synchronous resource occupancy information. Based on the synchronous resource occupancy information, the load status of each independent test channel is determined. Combined with the constraint that the load index difference between each channel is allowed to be within ±15%, a dynamic load migration strategy is established. Obtain the resource scheduling logs of the independent test channel within a preset historical time period, and determine the test preferences of the independent test channel based on the resource scheduling logs; An initial resource scheduling strategy is established based on the aforementioned test preferences and dynamic load migration strategy, specifically as follows: Based on the aforementioned testing preferences, a scheduling priority is established for resource allocation to the corresponding independent testing channels; Determine resource schedulability based on dynamic load migration strategies; An initial resource scheduling strategy is established based on the resource schedulability and scheduling priority. According to the initial resource scheduling strategy, perform resource scheduling on N independent test channels once, and determine whether there is a resource conflict after completing one resource scheduling; If so, perform secondary resource scheduling; Otherwise, complete the test according to the current resource status distribution; The secondary resource scheduling is as follows: Determine whether resource conflicts can be replaced by other hardware; if so, perform resource scheduling based on alternative hardware. Otherwise, determine whether the current task can be split and reorganized. If so, split and reorganize the current task, and then re-perform resource scheduling based on one and / or two resource scheduling to complete the test. Otherwise, perform resource simplification by compressing non-critical steps on the current task, obtain a simplified task, and then re-perform resource scheduling based on one and / or two resource scheduling to complete the test; S4: Synchronously acquire test data from each channel based on a distributed data acquisition mechanism, and generate a test report based on the test data from each channel.

2. The high-parallel testing method for an optical synthesis test platform according to claim 1, characterized in that, In S1, each independent test channel integrates an optical power meter, an adjustable optical attenuator, an optical switch, and a bit error rate meter module, specifically used for: An optical power meter is used to test the optical power of an object under test. Adjustable light attenuators are used to test the attenuation characteristics of the object under test; Optical switches are used to enable rapid switching between independent test channels; The bit error rate tester module is used to generate standard detection signals for the tested object; Each independent test channel is synchronously controlled via a backplane bus.

3. The high-parallel testing method for an optical synthesis test platform according to claim 1, characterized in that, In step S2, generating parallel test projects based on the object under test includes: Determine the type and nominal parameters of the object under test from its core information; Based on the type and nominal parameters of the object under test, test indicators are determined from industry standards and user requirements; Specific testing requirements are determined based on user needs; A parallel test project is established based on the aforementioned test metrics and specific test requirements.

4. The high-parallel testing method for an optical synthesis test platform according to claim 1, characterized in that, In step S2, the parallel test items are mapped to N independent test channels based on the task parsing algorithm, including: The parallel test project is parsed to obtain multiple subtasks, each containing a test metric. Based on resource independence and temporal dependence, a comprehensive analysis of subtasks is performed to determine the parallel and temporal characteristics between subtasks. Based on these parallel and temporal characteristics, the subtasks are then marked with parallel and temporal tags. The test duration of the test metrics of the subtasks is obtained. The subtasks are combined and packaged with parallel and timing markers to obtain a balanced task package. Based on specific test requirements, the balanced task package is optimized to obtain the target balanced task package. Based on the time sequence marker, the target balancing task packages are prioritized to obtain the priority order of each target balancing task package. Based on user needs, the priority weight of each target balancing task package is determined. Based on the priority order and priority weight, the target priority of the target balancing task package is determined. Obtain historical test data completed in N independent test channels, and determine the optimal test index sequence for each independent test channel based on the relationship between the independent test channels and the historical test data; Obtain the resource usage status of each independent test channel, and based on the resource usage status and the optimal test index sequence, establish an intelligent mapping model with the goal of maximizing parallel efficiency and minimizing resource conflicts; The target balanced task package is mapped to N independent test channels based on the intelligent mapping model.

5. The high-parallel testing method for an optical synthesis test platform according to claim 4, characterized in that, Also includes: The intelligent mapping model is optimized regularly, specifically as follows: Obtain the task mapping results based on the intelligent mapping model, and determine the mapping effect corresponding to the task mapping results; When the mapping effect meets the preset effect requirements, the intelligent mapping model is rewarded for this mapping; otherwise, the intelligent mapping model is penalized for this mapping. The intelligent mapping model is optimized based on the reward and punishment results to obtain the latest intelligent mapping model.

6. The high-parallel testing method for an optical synthesis test platform according to claim 1, characterized in that, In S3, real-time monitoring of resource status is performed on N independent test channels, including: Real-time monitoring of module usage when each independent test channel tests parallel test items; A resource usage table is generated based on the module's usage status.

7. The high-parallel testing method for an optical synthesis test platform according to claim 1, characterized in that, In step S4, test data from each channel is acquired synchronously based on a distributed data acquisition mechanism, including: Each independent test channel is configured with a high-speed serial transmission interface to send acquisition commands via the backplane bus; Based on the acquisition command, the acquired data is transmitted through a high-speed serial transmission interface to obtain test data for each channel.

8. The high-parallel testing method for an optical synthesis test platform according to claim 1, characterized in that, In step S4, a test report is generated based on the test data from each channel, including: Preprocess the test data from each channel to obtain the target test data; The test data is divided and processed according to the tested objects and test items to obtain object test reports and project test reports. The object test report and the project test report are integrated to generate a table of contents. The final test report is then obtained based on the table of contents, the object test report, and the project test report.

Images