A device cluster health detection method, system, device and storage medium

Abstract

The present disclosure relates to the technical field of device testing, and particularly relates to a device cluster health detection method and system, a device and a storage medium. The method comprises the following steps: obtaining a plurality of governance items and corresponding running data of a test device; performing window aggregation based on the running data to obtain a governance value, and determining a governance item state of each governance item according to the governance value and a governance rule; determining a health assessment result according to the governance item state and a preset governance item weight, and assigning a test task to each test device to obtain a test task execution result; generating a test device cluster health result according to the health assessment result, determining an expected task execution quality according to the cluster health result, determining an actual task execution quality based on the test task execution result, generating a governance gap result and determining a governance optimization object, optimizing the governance rule of the governance item, and re-detecting each test device. Through real-time detection of the running state of the test device cluster, effective allocation of the test task is realized.

Description

Technical Field

[0001] This disclosure relates to the field of equipment testing technology, and in particular to a method, system, device and storage medium for health detection of equipment clusters. Background Technology

[0002] With the rapid development of intelligent connected vehicles, the software scale of in-vehicle infotainment systems is constantly expanding, functions are continuously enriching, and the software version update frequency is constantly increasing. To ensure the functional correctness and operational stability of the in-vehicle software, automakers typically build in-vehicle testing laboratories, utilizing multiple testing devices to execute automated testing tasks in parallel, thereby shortening the testing cycle and improving testing efficiency.

[0003] During long-term continuous operation, the vehicle infotainment testing laboratory may experience test task failures due to reasons such as software service anomalies, insufficient computing resources, changes in the system environment, hardware failures, or abnormal external device connections. To ensure the continuous and stable operation of the laboratory, maintenance personnel typically need to periodically check the equipment's operating status and perform maintenance on any malfunctioning equipment based on the inspection results.

[0004] In existing technologies, the monitoring of the operational status of test equipment mainly relies on equipment monitoring software to collect operational information such as processor utilization, memory usage, disk space, network connection status, and service operation status, combined with manual inspection to determine whether the equipment is abnormal. When an abnormality is found, maintenance personnel typically suspend the equipment, locate the cause of the fault, and repair it before resuming testing. Due to the large number of laboratory test devices and the complex operating environment, the causes of failures vary significantly among different devices. The equipment status inspection and fault handling processes generally rely on manual experience, which may lead to abnormal equipment continuing to perform testing tasks, resulting in decreased testing efficiency and wasted testing resources.

[0005] Therefore, existing technologies cannot detect the operating status of testing equipment, resulting in a lack of effective basis for the allocation of testing tasks. Summary of the Invention

[0006] To address the aforementioned technical issues, this disclosure provides a method, system, device, and storage medium for health monitoring of a device cluster.

[0007] This disclosure provides a method for health detection of a device cluster, including: Multiple governance items of the test equipment are obtained, each governance item includes a monitoring indicator, a duration window and a corresponding governance rule, the operation data of each test equipment is collected, and the operation data corresponding to each governance item is determined according to the monitoring indicator of each governance item; Based on the running data and duration window corresponding to each governance item, window aggregation is performed to obtain the governance value of each governance item, and the governance item status corresponding to each governance item is determined according to the governance value and the corresponding governance rule. Based on the status of each governance item and the preset weight of each governance item, the health assessment result of each test device is determined, and test tasks are assigned to each test device based on the health assessment results, and the test task execution results of each test device are obtained. Based on the health assessment results of each test device, a test device cluster health result is generated. Based on the cluster health result, the expected task execution quality within the corresponding time window is determined. Based on the test task execution result, the actual task execution quality is determined. Based on the expected task execution quality and the actual task execution quality, a governance gap result is generated. Based on the results of the governance gaps, the governance optimization targets are determined, and the governance rules of the governance items are optimized based on the governance optimization targets. The optimized governance items are then used to re-test each testing device.

[0008] Further, determining the governance item status corresponding to each governance item based on the governance value and the corresponding governance rule includes: The governance item type is determined based on the data type of the operational data, and the governance item type includes binary governance items and interval governance items; If the running data is a finite state, then the corresponding governance item is determined as a binary governance item, and the state of the first governance item is determined according to the available state of the corresponding detection object; If the operating data is a continuous measurement value, the corresponding governance item is determined as an interval governance item. Each interval governance item includes a warning threshold and an unavailability threshold. The governance value is compared with the warning threshold and the unavailability threshold of the corresponding interval governance item to obtain a first comparison result and a second comparison result. The status of the second governance item is determined based on the first comparison result and the second comparison result.

[0009] Further, determining the governance item status corresponding to each governance item based on the governance value and the corresponding governance rule includes: When the duration of time during which the first governance item state or the second governance item state continuously meets the corresponding governance rule reaches the duration window, the state is updated to the first governance item state or the second governance item state; otherwise, the current governance item state is maintained.

[0010] Further, determining the health assessment result of each test device based on the governance item status and preset governance item weights for each governance item includes: If the status of the governance item is unavailable, the corresponding test device is determined to be an unschedulable device; otherwise, the device health score is determined according to the status of each governance item and the preset governance item weight, and the test device is determined to be a schedulable device and have a health level based on the device health score.

[0011] Further, generating a health result for the test equipment cluster based on the health assessment results of each test equipment includes: Determine the percentage of test devices in the unavailable state in the health assessment results; Health statistics are determined based on the health score of the test equipment in the available status described in the health assessment results; The health results of the test device cluster are generated based on the percentage of test devices in the unavailable state and the health statistics.

[0012] Furthermore, the step of generating governance gap results based on the expected task execution quality and the actual task execution quality includes: Based on the expected task execution quality and the actual task execution quality, determine the deviation between the expected task success rate and the actual task success rate; If the deviation exceeds a preset deviation threshold for multiple consecutive time windows, a governance gap result is generated.

[0013] Further, the step of determining the governance optimization targets based on the governance gap results, optimizing the governance rules of the governance items based on the governance optimization targets, and re-testing each detection device using the optimized governance items includes: When the governance gap result indicates that all governance items are in a normal state, a suggestion to add new governance items is generated. When the governance gap result indicates the existence of a governance item in a warning state, threshold adjustment suggestions for the corresponding governance item are generated based on historical operating data to obtain governance optimization suggestions; Based on the governance optimization suggestions, obtain optimization confirmation results, optimize the configuration of the governance items according to the optimization confirmation results, and re-test each testing device based on the optimized governance items.

[0014] This disclosure also provides a device cluster health monitoring system, including: The acquisition module is used to acquire multiple governance items of the test equipment. Each governance item includes a monitoring indicator, a duration window and a corresponding governance rule. The module collects the operating data of each test equipment and determines the operating data corresponding to each governance item based on the monitoring indicator of each governance item. The aggregation module is used to perform window aggregation based on the running data and duration window corresponding to each governance item to obtain the governance value of each governance item, and to determine the governance item status corresponding to each governance item according to the governance value and the corresponding governance rule; The evaluation module is used to determine the health evaluation result of each test device based on the status of each governance item and the preset governance item weight, and to allocate test tasks to each test device based on the health evaluation result, and obtain the test task execution result of each test device. The execution module is used to generate a test device cluster health result based on the health assessment results of each test device, determine the expected task execution quality within the corresponding time window based on the cluster health result, determine the actual task execution quality based on the test task execution result, and generate a governance gap result based on the expected task execution quality and the actual task execution quality. The detection module is used to determine the objects to be optimized for governance based on the results of the governance gaps, optimize the governance rules of the governance items based on the objects to be optimized for governance, and re-detect each detection device through the optimized governance items.

[0015] Furthermore, the aggregation module is specifically used for: The governance item type is determined based on the data type of the operational data, and the governance item type includes binary governance items and interval governance items; If the running data is a finite state, then the corresponding governance item is determined as a binary governance item, and the state of the first governance item is determined according to the available state of the corresponding detection object; If the operating data is a continuous measurement value, the corresponding governance item is determined as an interval governance item. Each interval governance item includes a warning threshold and an unavailability threshold. The governance value is compared with the warning threshold and the unavailability threshold of the corresponding interval governance item to obtain a first comparison result and a second comparison result. The status of the second governance item is determined based on the first comparison result and the second comparison result.

[0016] Furthermore, the aggregation module is specifically used for: When the duration of time during which the first governance item state or the second governance item state continuously meets the corresponding governance rule reaches the duration window, the state is updated to the first governance item state or the second governance item state; otherwise, the current governance item state is maintained.

[0017] Furthermore, the evaluation module is specifically used for: If the status of the governance item is unavailable, the corresponding test device is determined to be an unschedulable device; otherwise, the device health score is determined according to the status of each governance item and the preset governance item weight, and the test device is determined to be a schedulable device and have a health level based on the device health score.

[0018] Furthermore, the execution module is specifically used for: Determine the percentage of test devices in the unavailable state in the health assessment results; Health statistics are determined based on the health score of the test equipment in the available status described in the health assessment results; The health results of the test device cluster are generated based on the percentage of test devices in the unavailable state and the health statistics.

[0019] In some possible implementations, the execution module is specifically used to: determine the deviation between the expected task success rate and the actual task success rate based on the expected task execution quality and the actual task execution quality; and generate a governance gap result if the deviation exceeds a preset deviation threshold for multiple consecutive time windows.

[0020] Furthermore, the detection module is specifically used for: When the governance gap result indicates that all governance items are in a normal state, a suggestion to add new governance items is generated. When the governance gap result indicates the existence of a governance item in a warning state, threshold adjustment suggestions for the corresponding governance item are generated based on historical operating data to obtain governance optimization suggestions; Based on the governance optimization suggestions, obtain optimization confirmation results, optimize the configuration of the governance items according to the optimization confirmation results, and re-test each testing device based on the optimized governance items.

[0021] This disclosure also provides an electronic device, including: a processor; and a memory for storing executable instructions; wherein the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the device cluster health detection method described above.

[0022] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor implements the device cluster health detection method described above.

[0023] The technical solution provided in this disclosure has the following advantages compared with the prior art: By acquiring multiple governance items from the test equipment, each including monitoring indicators, duration windows, and corresponding governance rules, the operational data of each test equipment is collected, and the operational data corresponding to each governance item is determined based on the monitoring indicators of each governance item. Based on the operational data and duration windows corresponding to each governance item, window aggregation is performed to obtain the governance value of each governance item. The status of each governance item is determined based on the governance value and the corresponding governance rules. Based on the status of each governance item and the preset governance item weights, the health assessment results of each test equipment are determined, and test tasks are allocated to each test equipment based on the health assessment results, obtaining the test task execution results of each test equipment. A test equipment cluster health result is generated based on the health assessment results of each test equipment. The expected task execution quality within the corresponding time window is determined based on the cluster health result, and the actual task execution quality is determined based on the test task execution results. A governance gap result is generated based on the expected task execution quality and the actual task execution quality. The governance gap result determines the governance optimization object, and the governance rules of the governance item are optimized based on the governance optimization object. Each test equipment is re-tested using the optimized governance items. By monitoring the operational status of the test equipment cluster in real time, effective allocation of test tasks is achieved. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0025] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0026] Figure 1 A schematic diagram of the device cluster health detection method provided in the embodiments of this disclosure; Figure 2 A schematic diagram illustrating the health assessment results of various testing devices provided in the embodiments of this disclosure; Figure 3 A schematic diagram illustrating the configuration of optimized governance items provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram of a device cluster health monitoring system provided in an embodiment of this disclosure. Detailed Implementation

[0027] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0028] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0029] Figure 1 This is a schematic diagram of the device cluster health detection method provided in the embodiments of this disclosure; as shown Figure 1 As shown, a method for health detection of a device cluster includes: S1, acquire multiple governance items of the test equipment. Each governance item includes monitoring indicators, duration window and corresponding governance rules. Collect the operation data of each test equipment and determine the operation data corresponding to each governance item based on the monitoring indicators of each governance item. In this embodiment, a data foundation for health monitoring of test equipment is established, providing a unified data source for subsequent governance value calculation, health status determination, and health assessment. First, the system acquires multiple governance items corresponding to the test equipment. Each governance item is a basic monitoring unit describing the health status of the test equipment in a specific operational dimension. Each governance item corresponds to a specific monitoring object and its judgment logic, such as CPU (Central Processing Unit) utilization, remaining disk space, Worker service running status, test bench hardware connection status, and driver status. Each governance item includes monitoring metrics, a duration window, and corresponding governance rules.

[0030] Among them, monitoring indicators are specific parameters used to reflect the operating status of equipment, which can be objectively collectable data such as CPU utilization, memory usage, disk capacity, service process status, and network connection status; duration window is a parameter used to limit the statistical time range of monitoring data, such as 10 seconds, 30 seconds, 1 minute, or 5 minutes, to avoid frequent changes in health status due to instantaneous fluctuations and improve the stability of health monitoring results; governance rules are judgment rules pre-configured for the corresponding monitoring indicators, used to specify the aggregation method of operating data, status judgment conditions, and the basis for subsequent health assessment.

[0031] Subsequently, operational data from each test device is collected in real time through a data acquisition program deployed on the test equipment. This operational data can originate from multiple data sources, such as operating system performance interfaces, hardware monitoring interfaces, service status interfaces, log systems, or test platform status interfaces, forming a unified dataset. Since different governance items focus on different monitoring objects, target operational data matching each governance item is filtered from the collected operational data based on the monitoring metrics corresponding to each governance item, establishing a one-to-one correspondence between governance items and operational data. For example, when a governance item corresponds to CPU utilization, CPU utilization data is extracted; when a governance item corresponds to Worker service status, Worker service operational status data is extracted; and when a governance item corresponds to bench connection status, the corresponding hardware interface status data is extracted. This unified organization of operational data from different sources ensures that each governance item receives corresponding operational data. It enables comprehensive collection of operational status at the software, system, and hardware layers of the test equipment, improving the data integrity of health monitoring. Furthermore, by establishing a clear mapping relationship between governance items and monitoring metrics, interference between different monitoring metrics is avoided, improving the maintenance efficiency of the entire test equipment health monitoring system.

[0032] S2, perform window aggregation based on the running data and duration window corresponding to each governance item to obtain the governance value of each governance item, and determine the governance item status corresponding to each governance item based on the governance value and the corresponding governance rule; In this embodiment, the collected raw operational data is converted into governance results that reflect the health status of the test equipment, providing a unified data foundation for health assessment.

[0033] Because the operational data collected during the test equipment's operation is characterized by real-time changes, high sampling frequency, and susceptibility to instantaneous fluctuations, directly using single sampling results to judge the equipment status can easily lead to misjudgments due to occasional resource fluctuations or network jitter. Therefore, window aggregation is performed on the operational data based on the duration window corresponding to each governance item. Window aggregation refers to using a duration window as the statistical period to statistically analyze multiple operational data samples collected within that time period, forming statistical results that represent the operational status of that time window, such as calculating the average, maximum, minimum, median, duration of continuous over-limit, or number of over-limit occurrences. Window aggregation can effectively filter out instantaneous anomalies and improve the stability of subsequent health status assessments. The statistical results obtained from window aggregation are governance values, which are quantitative indicators reflecting the overall operational status of the corresponding governance item within the current time window, such as average CPU utilization, duration of continuous over-limit, continuous online status of Worker services, and hardware connection stabilization time.

[0034] Subsequently, the status of a governance item is determined based on the governance value and the pre-configured governance rules for that governance item. The governance item status represents the operational status of the monitored object corresponding to the current governance item, and can include different levels such as normal status, warning status, and unavailable status. By converting raw operational data into standardized governance values ​​and further mapping them uniformly to governance item statuses, not only is the impact of different units of measurement between different types of operational data avoided, but a unified health judgment standard is also established. This allows the operational status of different levels, such as software services, system resources, and hardware devices, to be described in a consistent manner, improving the consistency of health monitoring results and reducing the impact of transient abnormal data on overall monitoring results.

[0035] Specifically, S2, the governance item status corresponding to each governance item is determined according to the governance value and the corresponding governance rule, including: determining the governance item type according to the data type of the running data, the governance item type includes binary governance item and interval governance item; if the running data is a finite state, the corresponding governance item is determined as a binary governance item, and the first governance item status is determined according to the availability status of the corresponding detection object; if the running data is a continuous measurement value, the corresponding governance item is determined as an interval governance item, each interval governance item includes a warning threshold and an unavailability threshold, the governance value is compared with the warning threshold and unavailability threshold of the corresponding interval governance item respectively to obtain a first comparison result and a second comparison result, and the second governance item status is determined according to the first comparison result and the second comparison result.

[0036] Specifically, different health status determination methods are used for different types of monitoring data to improve the accuracy of health monitoring results. Since the testing equipment comprises multiple components such as control nodes, working nodes, and test benches, and the data types generated by these different components vary significantly, the governance item type is first determined based on the data type of the operational data.

[0037] Based on data characteristics, operational data can be divided into two categories: finite-state data and continuous measurement data. Finite-state data refers to data that can only take a limited number of discrete states, such as whether a service is started, whether a device is online, or whether a driver has been successfully loaded. Its state typically only includes a limited number of values ​​such as "normal," "abnormal," "online," and "offline," thus requiring the establishment of a binary governance item. The binary governance item is mainly used to describe whether the monitored object has basic availability capabilities. The state of the first governance item can be determined based on whether the monitored object is currently in an available state. For example, if the Worker service is running, it is determined to be in a normal state; if the service stops, it is determined to be in an unavailable state.

[0038] Continuous measurement data refers to numerical data that changes continuously, such as CPU utilization, disk space, response time, and network bandwidth. Therefore, corresponding interval governance items are established. In addition to monitoring indicators, interval governance items include warning thresholds and unavailability thresholds. The warning threshold represents the critical value at which device performance begins to decline but can still continue executing test tasks, while the unavailability threshold represents the boundary at which the device can no longer stably execute test tasks. The aggregated governance value is compared with the corresponding warning threshold and unavailability threshold to form a first comparison result and a second comparison result. The status of the second governance item is determined based on a comprehensive analysis of the two comparison results. For example, when CPU utilization exceeds the warning threshold but does not reach the unavailability threshold, it can be determined as a warning state; when it reaches the unavailability threshold, it is determined as an unavailability state. By establishing binary governance items and interval governance items based on data types, the problem of inaccurate evaluation of different data types due to a uniform judgment method is avoided. Furthermore, health judgment strategies that conform to actual operating characteristics can be adopted for different monitored objects, improving the accuracy of health status judgment.

[0039] In some possible implementations, S2 determines the state of each governance item based on the governance value and the corresponding governance rule, including: when the state of the first governance item or the state of the second governance item continuously satisfies the corresponding governance rule for a duration window, it is updated to the state of the first governance item or the state of the second governance item; otherwise, the current state of the governance item is maintained.

[0040] Specifically, this approach aims to avoid frequent changes in the status of governance items due to momentary anomalies or short-term fluctuations, thereby improving the stability of health monitoring results. During operation, test equipment is affected by factors such as system scheduling, resource contention, network jitter, and test task switching. Some operational metrics may experience abnormal changes within a very short period, such as a sudden increase in CPU utilization, short-term disk I / O congestion, or a brief increase in network latency. If the status of governance items is updated immediately based on only a single sampling result, normal fluctuations can easily be misjudged as equipment failures, leading to frequent entry into warning states or even unavailability states, affecting the normal scheduling of test tasks. Therefore, a duration window mechanism is introduced.

[0041] Based on the status of the first or second governance item, the current governance item status is not immediately updated, but rather continuously monitored to ensure its stability. When the first or second governance item status continuously meets the corresponding governance rule for a duration window, it indicates that the corresponding anomaly is not an occasional fluctuation but has been persistent. In this case, the current governance item status is updated to the new governance item status. If the duration has not yet reached the duration window, the current anomaly is considered a short-term fluctuation, and the original governance item status is maintained without triggering a status switch. Governance item status updates refer to switching the current governance item's status from normal to warning or unavailable, and also include the process of restoring from a warning state to a normal state. By introducing a duration window as a status update condition, transient anomalies and occasional noise can be effectively filtered, improving the stability of governance item status switching, avoiding excessive fluctuations in health assessment results due to frequent status changes, and reducing false alarms and false shutdowns.

[0042] S3. Based on the status of each governance item and the preset weight of each governance item, determine the health assessment results of each test device, and allocate test tasks to each test device based on the health assessment results, and obtain the test task execution results of each test device. In this embodiment, the overall health level of the test equipment is formed based on the operating status of each governance item, and the health assessment results are applied to the test task scheduling to achieve linkage control between equipment health status and test resource management.

[0043] Since the test equipment comprises multiple components such as control nodes, working nodes, and test benches, and each component has varying degrees of importance to the test task execution, corresponding governance item weights are pre-configured for each governance item. The governance item weight reflects the degree of impact of the corresponding governance item on the overall health of the equipment. Key operational indicators can be assigned higher weights, while general operational indicators can be assigned lower weights, thereby ensuring that the health assessment results more accurately reflect the actual usability of the test equipment.

[0044] Based on the status and weight of each governance item, a comprehensive calculation is performed on all governance items to obtain the health assessment result for the corresponding test equipment. The health assessment result comprehensively describes the current overall operating status of the test equipment, and may include equipment health score, health level, and equipment scheduling status. Specifically, the equipment health score is a quantitative evaluation result calculated based on all governance items; the health level indicates the overall health level of the equipment, such as healthy, warning, or unavailable; and the equipment scheduling status indicates whether the equipment is allowed to participate in test task scheduling.

[0045] Subsequently, the scheduling system allocates test tasks based on the health assessment results of each test device, assigning tasks only to devices that meet the scheduling conditions. For devices with abnormal health status or that are unschedulable, new test task allocation is suspended to prevent test task failure due to device malfunction. After a test task is completed, the system further collects the execution results for each test device. These results include information such as whether the test was successfully completed, execution time, reason for failure, and resource usage, reflecting the device's operational performance during the actual testing process. By establishing a correlation between health assessment results and test task scheduling, test resources can be dynamically allocated based on the device's operating status, preventing abnormal devices from continuously participating in testing and improving the success rate of test task execution.

[0046] Specifically, S3 determines the health assessment result of each test device based on the status of each governance item and the preset weight of each governance item, including: if the status of a governance item is unavailable, the corresponding test device is determined as an unschedulable device; otherwise, the device health score is determined based on the status of each governance item and the preset weight of each governance item, and the test device is determined as a schedulable device and its health level based on the device health score.

[0047] Specifically, when any governance item enters an unavailable state, it means that the critical function corresponding to that governance item can no longer work properly, such as the test service stopping, the test bench going offline, or the critical driver failing. At this time, even if other governance items are still in a normal state, the test equipment cannot reliably complete the test task. Therefore, the test equipment is directly identified as an unschedulable device, and the allocation of new test tasks is suspended.

[0048] When all governance items are not in an unavailable state, the overall equipment health score is calculated based on the status and weight of each governance item. The equipment health score is a quantitative evaluation index reflecting the overall health of the tested equipment, which can comprehensively reflect the operation of different governance items. The higher the score, the more stable the overall operation of the equipment.

[0049] Based on the equipment health score, the corresponding health level of the equipment is further determined, for example, it can be divided into different levels such as healthy and warning. The health level is then used to determine whether the equipment is allowed to continue participating in test task scheduling. By establishing a mapping relationship between governance item status, governance item weight, and equipment health score, a unified transformation from local operating status to overall health status is achieved. This not only avoids the one-sidedness caused by judging based on a single monitoring indicator, but also enables the scheduling system to rationally allocate test resources based on the overall health level of the equipment, reducing the large number of test failures caused by abnormal equipment participating in testing, and improving the overall testing efficiency and test resource utilization rate of the laboratory.

[0050] S4. Generate test equipment cluster health results based on the health assessment results of each test equipment, determine the expected task execution quality within the corresponding time window based on the cluster health results, determine the actual task execution quality based on the test task execution results, and generate governance gap results based on the expected task execution quality and the actual task execution quality. In this embodiment, the operational status of the testing equipment is analyzed from the perspective of the laboratory as a whole, and deficiencies in the health monitoring system are identified by comparing the theoretical and actual performance. Since laboratories typically deploy multiple testing devices that share automated testing tasks, focusing solely on the health status of a single device is insufficient to comprehensively reflect the overall operational capabilities of the laboratory.

[0051] The health result of the test equipment cluster is generated based on the health assessment results of all test equipment. The health result of the test equipment cluster represents the overall operational level of the entire laboratory within the current time window, and can be used to form a unified evaluation index by comprehensively considering factors such as the number of schedulable equipment, the health distribution of equipment, and the proportion of unschedulable equipment.

[0052] Subsequently, the expected task execution quality for the current time window is determined based on the cluster health results. Expected task execution quality is the theoretical execution effect predicted based on the laboratory's current overall health status, and may include indicators such as expected task success rate, expected number of completed tasks, or expected average execution time. It reflects the testing capability that the laboratory should theoretically achieve under the current equipment operating conditions. On the other hand, the actual task execution quality within the corresponding time window is statistically obtained based on the test task execution results. The actual task execution quality reflects the true execution of the test tasks, such as actual success rate, average execution time, and number of failed tasks. When there is a significant deviation between the expected and actual task execution quality, it indicates that the current health monitoring results fail to fully reflect the actual operating conditions of the equipment. Therefore, a governance gap result is generated based on the difference between the two. The governance gap result indicates the degree of difference between the health monitoring results and the actual operating effect. When the governance gap is small, it indicates that the current health monitoring can accurately reflect the equipment's operating status; when the governance gap continues to increase, it indicates that the existing monitoring rules may have monitoring blind spots or insufficient judgment capabilities, requiring further optimization of the health monitoring strategy. By establishing the correlation between cluster health results, expected performance quality, and actual performance quality, the analysis has been expanded from single-device health analysis to overall laboratory operational quality analysis. This enables health monitoring to not only reflect the status of the equipment itself, but also to evaluate whether the health monitoring system can truly reflect the test execution effect, thereby improving the accuracy of overall laboratory health management.

[0053] Specifically, S4 generates a test equipment cluster health result based on the health assessment results of each test equipment, including: determining the proportion of test equipment in an unavailable state in the health assessment results; determining health statistics based on the health scores of test equipment in an available state in the health assessment results; and generating a test equipment cluster health result based on the proportion of test equipment in an unavailable state and the health statistics.

[0054] Specifically, since automated testing tasks are typically completed collaboratively by multiple testing devices, even if a single device is in good health, it cannot fully reflect the overall operational capability of the laboratory. The number of testing devices in an unavailable state according to the health assessment results is counted, and the proportion of these unavailable devices to all testing devices is calculated. This proportion directly reflects the effective proportion of equipment currently available for testing tasks in the laboratory. When the proportion of unavailable devices is high, it indicates a significant decline in the overall operational capability of the laboratory, and maintenance personnel should be prioritized for troubleshooting.

[0055] Subsequently, the health scores of all schedulable devices are statistically analyzed to generate corresponding health statistics. The percentage of unavailable test devices and the health statistics are then combined to generate the corresponding test device cluster health results. By establishing cluster health results, laboratory managers can quickly understand the overall operational status and rationally arrange equipment maintenance plans, test task scale, and equipment expansion strategies based on the cluster health level.

[0056] In some possible implementations, S4 generates governance gap results based on the expected task execution quality and the actual task execution quality, including: determining the deviation between the expected task success rate and the actual task success rate based on the expected task execution quality and the actual task execution quality; generating governance gap results if the deviation exceeds a preset deviation threshold for multiple consecutive time windows.

[0057] In this embodiment, a small deviation indicates that the current health monitoring system can accurately reflect the equipment's operating status; a continuously increasing deviation indicates that the current health assessment results cannot fully explain actual test failures, and the existing health monitoring system may have problems such as newly discovered anomalies not yet covered or unreasonable judgment rules. To avoid misjudgments caused by accidental anomalies, the corresponding governance gap result is only generated when the deviation exceeds a preset deviation threshold for multiple consecutive time windows. By continuously comparing the differences between theoretical predictions and actual execution results, the effectiveness of the health monitoring system itself is continuously evaluated, enabling it not only to monitor the equipment's health status but also to automatically identify deficiencies in the monitoring system.

[0058] S5. Based on the results of the governance gap, determine the governance optimization targets, optimize the governance rules of the governance items based on the governance optimization targets, and re-test each testing device through the optimized governance items.

[0059] In this embodiment, the health monitoring system is continuously optimized, allowing the health monitoring rules to be constantly improved based on laboratory operational experience, thus forming a health governance process. Once governance gap results are generated, the corresponding governance optimization targets are determined based on these results. The governance optimization target refers to the target object that caused the governance gap, which can be a specific governance item, a type of monitoring indicator, or a corresponding governance rule.

[0060] When the governance gap results indicate that the overall health status of the current test equipment is normal, but the quality of test task execution continues to decline, it indicates that there may be new types of anomalies that have not yet been covered by the existing monitoring system. In this case, the newly added monitoring objects can be identified as governance optimization objects. When the governance gap results indicate that some governance items are in a warning state for a long time and the actual task failure rate continues to increase, it indicates that the corresponding governance rules may have unreasonable judgment conditions or inaccurate threshold settings. The corresponding governance rules can be identified as governance optimization objects.

[0061] Subsequently, based on historical operational data, task execution results, and manual maintenance experience, the governance rules corresponding to the governance optimization targets are optimized. This includes adding new monitoring rules, adjusting status judgment conditions, modifying corresponding thresholds, or improving status transition logic. After optimizing the governance rules, the updated governance items are used to re-monitor the health of each test device. This involves re-executing steps such as operational data collection, governance value calculation, governance status judgment, and health assessment, thus forming a complete health monitoring closed loop. By continuously using actual operational results to optimize the health monitoring rules, the health monitoring system can accumulate experience over time, improving the accuracy of health monitoring, reducing manual maintenance workload, minimizing test resource waste caused by recurring failures, and continuously enhancing the overall health management capabilities of the vehicle-mounted vehicle testing laboratory.

[0062] Specifically, S5, based on the governance gap results, determines the governance optimization targets, optimizes the governance rules of the governance items based on the governance optimization targets, and re-tests each testing device using the optimized governance items, including: when the governance gap results indicate that all governance items are in a normal state, generating suggestions for adding new governance items; when the governance gap results indicate that there are governance items in a warning state, generating threshold adjustment suggestions for the corresponding governance items based on historical operating data to obtain governance optimization suggestions; based on the governance optimization suggestions, obtaining optimization confirmation results, optimizing the configuration of governance items based on the optimization confirmation results, and re-testing each testing device based on the optimized governance items.

[0063] Specifically, after generating governance gap results, the corresponding governance optimization targets are identified. When the actual test task execution quality consistently falls short of expectations, it indicates that the current health monitoring system has not yet covered new abnormal factors causing task failure, i.e., there are monitoring blind spots. At this time, based on test failure logs, device operation logs, and manual maintenance records, suggestions for adding new governance items are generated, prompting maintenance personnel to add new monitoring indicators or establish new governance items to expand the coverage of health monitoring.

[0064] When the governance gap results indicate the presence of governance items in a warning state, it means that the existing governance items are already able to reflect the abnormal trend, but the corresponding judgment conditions may not be reasonable enough. Therefore, further statistical analysis is conducted on the operation of governance items during normal and fault periods, based on historical operational data, to generate threshold adjustment suggestions for the corresponding governance items. These threshold adjustment suggestions are used to assist operations and maintenance personnel in optimizing warning or unavailability thresholds, enabling governance items to identify equipment anomalies earlier and more accurately. Subsequently, corresponding governance optimization suggestions are formed and submitted to operations and maintenance personnel for manual confirmation. The optimization confirmation results are introduced because governance rules directly affect the scheduling of the entire laboratory testing task. Completely automatic modification may introduce new misjudgments; therefore, the final decision on whether to update the governance item configuration rests with the operations and maintenance personnel.

[0065] After confirmation, the corresponding governance item configurations are updated based on the optimization confirmation results, and the optimized governance items are reused to perform health monitoring on each test device. The health monitoring system can continuously absorb new operation and maintenance experience, improve anomaly identification and rule adaptation capabilities, reduce the workload of manual handling caused by repeated failures, and enable the laboratory's health management capabilities to continuously improve over time.

[0066] In some possible ways of implementation, Figure 2 This is a schematic diagram of the health assessment results of various testing devices provided in the embodiments of this disclosure; as shown below. Figure 2 As shown, single device health assessment: The governance item system is the basic unit of equipment health assessment. Governance items are the fundamental components of equipment health assessment. Each governance item corresponds to a quantitative monitoring rule for a specific observable dimension of the equipment, consisting of five elements: "monitoring indicator + normal range + alarm threshold + unavailability threshold + duration window". Each test device corresponds to a complete list of governance items, covering the following three dimensions: PC (Personal Computer) control node: CPU utilization, I / O input / output response, disk space, driver availability, ZL (test framework core service component) service availability, etc. Worker (worker node): Service process liveness status and task response capability; Bench (Hardware Platform): This includes the availability of in-vehicle infotainment hardware, the completeness of the BOM (Bill of Materials) configuration, and the compliance of parts change records. The list of governance items was jointly determined by a team of technical experts through multi-party consensus.

[0067] Formal definition of governance terms, binary terms, and interval terms: The platform has a total of Each governance item constitutes a set. The weight of each governance item is: (Set by experts according to the importance of the business). Governance items are set. At the assessment time The original monitoring sequence is Its duration window is Define valid measurement values for: ; in: For window aggregation functions that match the type of governance item (such as window mean, maximum value, duration of exceeding threshold, or percentage of time exceeding threshold). This refers to the valid measurement value, not the instantaneous sampled value.

[0068] Type 1: Binary Governance Item, with only two states. For example, relay availability and service process liveness status, with no yellow alert intermediate state: : (Green, normal) or (Red, unavailable); Type 2: Interval Governance Item, with continuously measurable values. The initial warning threshold is set by experts. With unavailable threshold Interval terms are divided into two categories based on the directionality of the indicator: Reverse metrics (the lower the better, such as CPU utilization, I / O latency): (Green) When ; (Yellow) When ; (Red) When ; This refers to the valid measurement value, not the instantaneous sampled value.

[0069] Positive metrics (the higher the better, such as remaining disk space, worker task response rate): (Green) When ; (Yellow) When ; (Red) When ; This refers to the valid measurement value, not the instantaneous sampled value.

[0070] Take "PC CPU utilization" (a reverse indicator) as an example: , The duration window is 20 minutes, meaning if the CPU usage exceeds 80% continuously for more than 20 minutes, this governance item will be marked as red.

[0071] The formula for calculating the health score of a single aircraft is based on the leaky bucket strategy and weighted averaging. The definition is as follows: (If it exists) (If the leaky bucket is triggered, the entire platform turns red); otherwise, the calculation formula is as follows: ; Health level mapping (reference value): Red (Leaky bucket triggered); It is yellow; It is green.

[0072] Figure 3 A schematic diagram illustrating the configuration of optimized governance items provided in embodiments of this disclosure; as shown Figure 3 As shown, the three-layer decision-making strategy and self-driven iteration mechanism are as follows: Strategy 1: Binary Availability Judgment (Triggering Red) This strategy sets a critical unavailability threshold and duration window for each governance item. When the condition is met, the item is marked as red and the bench is automatically taken offline. This is a hard line of defense for equipment health management.

[0073] Strategy Two: Manual Operation and Maintenance Strategy (triggered in yellow, and serves as the entry point for Strategy One). When a governance item is in the warning zone, it is highlighted in yellow, triggering the manual operation and maintenance process. Each manual handling result is recorded in a structured manner, refined, and then injected back into Strategy One, forming a positive evolutionary cycle of "problem discovery, manual intervention, rule refinement, and strategy upgrade".

[0074] Key points of the feedback mechanism: Record the validity of the alarm (real fault / false alarm) when it is first triggered to calibrate the threshold; extract new automation rules when similar problems recur; drive dynamic threshold adjustment after the accumulation of operation and maintenance experience.

[0075] Strategy 3: Objective indicators assist strategy (exposing monitoring blind spots and driving supplementary iterations). Even if all governance items are green, if the actual task execution quality declines, it indicates that there are monitoring blind spots, triggering the supplementation of governance items.

[0076] Laboratory cluster health assessment: The formula for calculating cluster health scores is given, assuming the laboratory has a total of The test benches are categorized into health levels for each test bench. Cluster health score reflects both the "health level of available racks" and the "direct penalty for unavailable racks": ; in, Percentage of unusable racks; The 20th percentile for all non-red table health scores. As a strong penalty, it ensures a significant drop in cluster health score when a large number of racks fail; P20 is more sensitive to the "worst available racks" than the mean or median, and better aligns with the design goal of being "sensitive to localized degradation"; it is defined when all racks are red. The scheduling system senses in real time. Tasks will only be assigned to green or yellow stations, and tasks will not be assigned to red stations.

[0077] Governance gap detection algorithm, defining task execution quality indicators (in terms of time windows) (Statistical granularity) For the first Actual success rate of tasks performed in the window; This represents the average task execution time (normalized ratio relative to the historical baseline).

[0078] Expected success rate Calibration based on historical data (coefficients) Fitting data from historical time series data (initial values ​​provided by expert experience), and using a truncation function to ensure the value range is strictly within a certain range. Inside: ; The governance gap signal is defined as the deviation between expectations and reality: ; Triggering condition: When the gap signal is significant and persistent, the expert review process is triggered. ; After being triggered, the system will handle the situation in two ways: Situation A (all governance items are green): the system will diagnose the situation as a monitoring blind spot and prompt experts to add a new governance item; Situation B (a certain governance item is yellow / close to the red line): the system will diagnose the situation as a threshold being too loose and will initiate a threshold adjustment suggestion algorithm.

[0079] The threshold adjustment suggestion algorithm (expert decision support, not executed automatically) generates threshold adjustment suggestions for experts when scenario B is triggered. The final decision on whether to adjust the threshold rests with the experts; the system does not automatically modify any thresholds.

[0080] Step 1: Calculate the correlation between governance items and task failure—To unify the meaning of the correlation of indicators in different directions, calculate the correlation between each interval-type governance item. First, construct risk-related variables. Reverse indicator Positive indicators (By reversing the direction, the variable corresponding to "increased risk" always increases.) Then calculate. With task failure rate Spearman's rank correlation coefficient (more robust to nonlinear step-failure relationships): ; After the direction is unified, The larger the value, the stronger the association between the increased risk of the governance item and the increased task failure rate. Both types of indicators can be screened using the same threshold. Screening for significantly related governance items ( ,suggestion As a candidate for adjustment, according to Sort in descending order.

[0081] Step 2: Calculate the recommended threshold for significantly relevant governance items. During the historical fault period ( In the sample of measured values, different quantiles are taken based on the directionality of the indicator to capture the "initial critical point" of the fault rather than extreme values: Contrarian indicator (the smaller the better): Take That is, the lower boundary of the fault sample, which corresponds to "the fault starts to occur from here"; Positive indicators (the larger the better): Take This refers to the high-end boundary of the fault sample, which corresponds to "faults begin to occur below this value"; ; List of candidate governance adjustments (by) (descending order), current threshold Compared with the suggested threshold The comparison and adjustment of the impact estimate (how many historical fault periods will be captured in advance after the adjustment) are presented to the experts. The experts manually confirm whether to update the threshold, and the system records each change history.

[0082] Key metrics and operations dashboards: A unified health dashboard is established at the laboratory level, supporting daily operations and maintenance management decisions through the following key metrics: Percentage of healthy (green) equipment: Daily inspection target >60%; if it falls below the threshold, a centralized remediation action will be triggered. Number of unavailable hardware units: Unavailable racks must be repaired as soon as possible in accordance with the MTTR (Mean Time To Repair) principle, while maintaining the integrity of BOM meta tags; Equipment utilization rate: Under the premise of reasonable algorithm allocation, guide the laboratory to add or remove specific types of equipment as needed; Percentage of tasks expected to be completed / not expected to be completed: This measures the overall performance of the scheduling algorithm and identifies scheduling strategies that need to be optimized. Governance gap signal Continuously monitor the deviation between expected and actual task success rates to expose blind spots in governance; Distribution of test case execution failure reasons: Breaking down the failure reasons (non-standard hardware change process / hardware unavailability) to guide the direction of operation and maintenance improvement.

[0083] Figure 4 This is a schematic diagram of the device cluster health monitoring system provided in the embodiments of this disclosure; as shown Figure 4 As shown, this disclosure also provides a device cluster health monitoring system, including: an acquisition module 401, used to acquire multiple governance items of test devices, each governance item including monitoring indicators, duration windows and corresponding governance rules, collect the operation data of each test device, and determine the operation data corresponding to each governance item based on the monitoring indicators of each governance item; an aggregation module 402, used to perform window aggregation based on the operation data and duration windows corresponding to each governance item to obtain the governance value of each governance item, and determine the governance item status corresponding to each governance item based on the governance value and the corresponding governance rules; and an evaluation module 403, used to determine the status of each test device based on the governance item status corresponding to each governance item and the preset governance item weight. The system analyzes the health assessment results of each test device and assigns test tasks to each test device based on these results, obtaining the test task execution results for each test device. The execution module 404 generates a test device cluster health result based on the health assessment results of each test device, determines the expected task execution quality within the corresponding time window based on the cluster health result, determines the actual task execution quality based on the test task execution results, and generates a governance gap result based on the expected and actual task execution quality. The detection module 405 determines the governance optimization object based on the governance gap result, optimizes the governance rules of the governance items based on the governance optimization object, and re-tests each test device using the optimized governance items.

[0084] In some possible implementations, the aggregation module 402 is specifically used to: determine the governance item type based on the data type of the running data, whereby the governance item type includes binary governance items and interval governance items; if the running data is a finite state, then the corresponding governance item is determined as a binary governance item, and the first governance item state is determined based on the availability state of the corresponding detection object; if the running data is a continuous measurement value, then the corresponding governance item is determined as an interval governance item, whereby each interval governance item includes a warning threshold and an unavailability threshold, and the governance value is compared with the warning threshold and unavailability threshold of the corresponding interval governance item respectively to obtain a first comparison result and a second comparison result, and the second governance item state is determined based on the first comparison result and the second comparison result.

[0085] In some possible implementations, the aggregation module 402 is specifically used to: update the first governance item state or the second governance item state to the first governance item state or the second governance item state when the time for which the first governance item state or the second governance item state continuously satisfies the corresponding governance rule reaches the duration window; otherwise, maintain the current governance item state.

[0086] In some possible implementations, the evaluation module 403 is specifically used to: if the governance item status is unavailable, determine the corresponding test device as an unschedulable device; otherwise, determine the device health score based on the governance item status and preset governance item weights corresponding to each governance item, and determine the test device as a schedulable device and its health level based on the device health score.

[0087] In some possible implementations, execution module 404 is specifically used to: determine the percentage of test devices in an unavailable state in the health assessment results; determine health statistics based on the health scores of test devices in an available state in the health assessment results; and generate test device cluster health results based on the percentage of test devices in an unavailable state and the health statistics.

[0088] In some possible implementations, the execution module 404 is specifically used to: determine the deviation between the expected task success rate and the actual task success rate based on the expected task execution quality and the actual task execution quality; and generate a governance gap result if the deviation exceeds a preset deviation threshold for multiple consecutive time windows.

[0089] In some possible implementations, the detection module 405 is specifically used to: generate suggestions for adding new governance items when the governance gap result indicates that all governance items are in a normal state; when the governance gap result indicates that there are governance items in a warning state, generate threshold adjustment suggestions for the corresponding governance items based on historical operating data to obtain governance optimization suggestions; based on the governance optimization suggestions, obtain optimization confirmation results, optimize the configuration of governance items according to the optimization confirmation results, and re-detect each detection device based on the optimized governance items.

[0090] This disclosure also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of a device cluster health detection method.

[0091] The computer device includes a memory, a processor, and a network interface that are interconnected via a system bus. It should be noted that not all of the shown components are required; more or fewer components may be implemented instead. Those skilled in the art will understand that the computer device described herein is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0092] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0093] The memory includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory may be an internal storage unit of the computer device, such as the hard disk or memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Of course, the memory may include both internal storage units and external storage devices of the computer device. In this embodiment, the memory is typically used to store the operating system and various application software installed on the computer device, such as program code for real-time data processing methods. Furthermore, the memory can also be used to temporarily store various types of data that have been output or will be output.

[0094] In some embodiments, the processor may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor is typically used to control the overall operation of the computer device. In this embodiment, the processor is used to run program code stored in the memory or process data, for example, to run the program code of the method.

[0095] The network interface may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device and other electronic devices.

[0096] This disclosure also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of a device cluster health detection method.

[0097] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0098] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0099] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for health detection of a cluster of devices, characterized in that, include: Multiple governance items of the test equipment are obtained, each governance item includes a monitoring indicator, a duration window and a corresponding governance rule, the operation data of each test equipment is collected, and the operation data corresponding to each governance item is determined according to the monitoring indicator of each governance item; Based on the running data and duration window corresponding to each governance item, window aggregation is performed to obtain the governance value of each governance item, and the governance item status corresponding to each governance item is determined according to the governance value and the corresponding governance rule. Based on the status of each governance item and the preset weight of each governance item, the health assessment result of each test device is determined, and test tasks are assigned to each test device based on the health assessment results, and the test task execution results of each test device are obtained. Based on the health assessment results of each test device, a test device cluster health result is generated. Based on the cluster health result, the expected task execution quality within the corresponding time window is determined. Based on the test task execution result, the actual task execution quality is determined. Based on the expected task execution quality and the actual task execution quality, a governance gap result is generated. Based on the results of the governance gaps, the governance optimization targets are determined, and the governance rules of the governance items are optimized based on the governance optimization targets. The optimized governance items are then used to re-test each testing device.

2. The equipment cluster health detection method according to claim 1, characterized in that, The step of determining the governance item status corresponding to each governance item based on the governance value and the corresponding governance rule includes: The governance item type is determined based on the data type of the operational data, and the governance item type includes binary governance items and interval governance items; If the running data is a finite state, then the corresponding governance item is determined as a binary governance item, and the state of the first governance item is determined according to the available state of the corresponding detection object; If the operating data is a continuous measurement value, the corresponding governance item is determined as an interval governance item. Each interval governance item includes a warning threshold and an unavailability threshold. The governance value is compared with the warning threshold and the unavailability threshold of the corresponding interval governance item to obtain a first comparison result and a second comparison result. The status of the second governance item is determined based on the first comparison result and the second comparison result.

3. The equipment cluster health detection method according to claim 2, characterized in that, The step of determining the governance item status corresponding to each governance item based on the governance value and the corresponding governance rule includes: When the duration of time during which the first governance item state or the second governance item state continuously meets the corresponding governance rule reaches the duration window, the state is updated to the first governance item state or the second governance item state; otherwise, the current governance item state is maintained.

4. The equipment cluster health detection method according to claim 1, characterized in that, The step of determining the health assessment result of each test device based on the governance item status and preset governance item weights for each governance item includes: If the status of the governance item is unavailable, the corresponding test device is determined to be an unschedulable device; otherwise, the device health score is determined according to the status of each governance item and the preset governance item weight, and the test device is determined to be a schedulable device and have a health level based on the device health score.

5. The equipment cluster health detection method according to claim 4, characterized in that, The step of generating a test equipment cluster health result based on the health assessment results of each test equipment includes: Determine the percentage of test devices in the unavailable state in the health assessment results; Health statistics are determined based on the health score of the test equipment in the available status described in the health assessment results; Based on the percentage of test devices in the unavailable state and the health statistics, a health result for the test device cluster is generated.

6. The equipment cluster health detection method according to claim 1, characterized in that, The process of generating governance gap results based on the expected task execution quality and the actual task execution quality includes: Based on the expected task execution quality and the actual task execution quality, determine the deviation between the expected task success rate and the actual task success rate; If the deviation exceeds a preset deviation threshold for multiple consecutive time windows, a governance gap result is generated.

7. The equipment cluster health detection method according to claim 1, characterized in that, The step of determining the governance optimization targets based on the governance gap results, optimizing the governance rules of the governance items based on the governance optimization targets, and re-testing each detection device using the optimized governance items includes: When the governance gap result indicates that all governance items are in a normal state, a suggestion to add new governance items is generated. When the governance gap result indicates the existence of a governance item in a warning state, threshold adjustment suggestions for the corresponding governance item are generated based on historical operating data to obtain governance optimization suggestions; Based on the governance optimization suggestions, obtain optimization confirmation results, optimize the configuration of the governance items according to the optimization confirmation results, and re-test each testing device based on the optimized governance items.

8. A health monitoring system for a cluster of equipment, characterized in that, include: The acquisition module is used to acquire multiple governance items of the test equipment. Each governance item includes a monitoring indicator, a duration window and a corresponding governance rule. The module collects the operating data of each test equipment and determines the operating data corresponding to each governance item based on the monitoring indicator of each governance item. The aggregation module is used to perform window aggregation based on the running data and duration window corresponding to each governance item to obtain the governance value of each governance item, and to determine the governance item status corresponding to each governance item according to the governance value and the corresponding governance rule; The evaluation module is used to determine the health evaluation result of each test device based on the status of each governance item and the preset governance item weight, and to allocate test tasks to each test device based on the health evaluation result, and obtain the test task execution result of each test device. The execution module is used to generate a test device cluster health result based on the health assessment results of each test device, determine the expected task execution quality within the corresponding time window based on the cluster health result, determine the actual task execution quality based on the test task execution result, and generate a governance gap result based on the expected task execution quality and the actual task execution quality. The detection module is used to determine the objects to be optimized for governance based on the results of the governance gaps, optimize the governance rules of the governance items based on the objects to be optimized for governance, and re-detect each detection device through the optimized governance items.

9. An electronic device, characterized in that, include: processor; A memory for storing executable instructions; wherein the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the device cluster health detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The storage medium stores a computer program, which, when executed by a processor, causes the processor to implement the device cluster health detection method according to any one of claims 1 to 7.

Images