Alarm system for server equipment

Abstract

The invention relates to the technical field of server equipment monitoring, and particularly discloses an alarm system for server equipment, which comprises a running state monitoring module, a data analysis module, an alarm triggering module and an intelligent optimization module. According to the system, operation parameters are collected in real time through a heat-sensitive sensor and a flow sensor, the sensitivity of the heat-sensitive sensor is adjusted through an intelligent optimization module according to the alarm triggering frequency change rate, and the data sampling period is adjusted according to alarm event time interval distribution; and dynamically adjusting the alarm priority according to the network flow fluctuation amplitude and the server load instantaneous change rate. The method can reduce false alarms, improve alarm real-time performance and enhance system reliability, and is suitable for a complex server operation environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of server security monitoring, and specifically relates to an alarm system for server devices. Background Art

[0002] With the wide application of server devices in modern data centers and cloud computing environments, the real-time monitoring of their operating status and fault alarms have become key links to ensure system stability and efficient operation and maintenance. However, the existing server alarm systems still have significant deficiencies in terms of adaptability under complex working conditions, intelligence of alarm mechanisms, and system integration, which affect the operation and maintenance efficiency and system reliability.

[0003] For example, the patented technology with the publication number CN113765687B obtains the working mode of the server through a monitoring platform, and judges whether the target server is in a maintenance state according to the working mode, and sends an alarm notification when the alarm condition is met. However, this technical solution mainly relies on the judgment of the server's working mode, and does not fully consider the dynamic changes of the server under various complex operating scenarios, which may lead to false alarms or missed alarms. For example, when the server is in the maintenance mode but the key heat dissipation components are abnormal, the alarm may be delayed due to the setting of the mode priority, resulting in a risk of hardware damage. In addition, the setting of the alarm condition in this solution is relatively fixed, lacking the ability of adaptive adjustment, and it is difficult to meet the personalized needs of different server devices. Another example is the patented technology with the publication number CN112037820B, which collects the audio stream and analyzes the target emotion type therein, and sends an alarm message to the server, so as to achieve reliable alarms under conditions of no light or insufficient light. However, this technical solution mainly focuses on the emotion analysis of the audio stream, and does not fully combine the actual operating parameters of the server device (such as temperature, load, network traffic, etc.), which may lead to insufficient accuracy and timeliness of the alarm information. For example, environmental noise may be misjudged as an "urgent emotion" and trigger a false alarm, while the abnormal temperature of the actual server is not monitored. At the same time, the processing of audio data in this solution relies on complex algorithms, which may increase the computational burden of the system and reduce the alarm response speed. The above two pieces of prior art have not achieved multi-source data fusion, lack the intelligent idea of multi-parameter collaborative decision-making, and rely on fixed thresholds or mode rules, and cannot dynamically optimize the alarm strategy according to historical fault data or environmental changes.

[0004] The above problems indicate that there are still certain deficiencies in the existing server alarm systems in terms of dynamic adaptability, multi-dimensional data fusion, and the intelligence of the alarm mechanism. Especially when facing diverse and complex operating environments, it is difficult for the existing technologies to achieve efficient and accurate alarm functions. Therefore, there is an urgent need for a modular alarm system that can comprehensively analyze various operating parameters of the server, optimize the adaptive ability of the alarm mechanism, and improve the accuracy and real-time performance of alarm information, so as to meet the requirements of modern data centers for efficient and intelligent alarm systems. The present invention aims to solve the above technical problems and provide an innovative solution. Summary of the Invention

[0005] The present invention provides an alarm system for server devices to overcome the problems in the prior art that due to the frequent and complex fluctuations of the operating parameters of server devices, the traditional alarm systems are insufficient in dynamic adaptability and multi-dimensional data fusion capabilities.

[0006] To achieve the above object, the present invention provides an alarm system for server devices, comprising:

[0007] An operating status monitoring module for real-time collection of the operating parameters of server devices through a sensor group, wherein the sensor group includes a thermosensitive sensor disposed inside the server for detecting temperature changes and a traffic sensor connected to the server for detecting network traffic fluctuations; wherein, the server includes at least one server device;

[0008] A data analysis module connected to the operating status monitoring module for feature extraction and anomaly analysis of the collected operating parameters;

[0009] An alarm trigger module respectively connected to the operating status monitoring module and the data analysis module for sending an alarm signal when the operating parameters exceed the set threshold, wherein the alarm trigger module includes an alarm unit for generating an audible and visual alarm signal and a log storage unit for recording alarm events;

[0010] The intelligent optimization module is respectively connected to the operating status monitoring module, the data analysis module, and the alarm triggering module, and is used to adjust the sensitivity of the thermal sensor according to the change rate of the alarm triggering frequency or adjust the data sampling period of the alarm system according to the time interval distribution of alarm events in the log storage unit, and determine the priority of alarm triggering according to the network traffic fluctuation amplitude and the instantaneous change rate of the server load; wherein, the network traffic fluctuation amplitude is the difference between the maximum value and the minimum value in the network traffic time series collected by the traffic sensor within a preset time period; when the network traffic fluctuation amplitude exceeds the preset fluctuation amplitude threshold, determine whether the data processing capacity of the alarm system meets the requirements according to the instantaneous change rate of the server load. If the instantaneous change rate of the server load is greater than the preset change rate threshold, it is determined that the data processing capacity of the alarm system does not meet the requirements, and the priority of alarm triggering is increased, where the increase amplitude of the priority of alarm triggering is determined by the difference between the instantaneous change rate of the server load and the preset change rate threshold.

[0011] In an alternative manner, the intelligent optimization module is used to determine whether the dynamic adaptability of the alarm system meets the requirements according to the change rate of the alarm triggering frequency. If the change rate of the alarm triggering frequency is greater than the preset first change rate, it is determined that the dynamic adaptability of the alarm system does not meet the requirements.

[0012] In an alternative manner, the intelligent optimization module is used to preliminarily determine that the operating efficiency of the alarm system does not meet the requirements when the change rate of the alarm triggering frequency is greater than the preset first change rate and less than or equal to the preset second change rate, and determine whether the operating efficiency of the alarm system meets the requirements according to the time interval distribution of alarm events in the log storage unit.

[0013] In an alternative manner, the intelligent optimization module is used to reduce the sensitivity of the thermal sensor when the change rate of the alarm triggering frequency is greater than the preset second change rate; wherein, the reduction amplitude of the sensitivity of the thermal sensor is determined by the difference between the change rate of the alarm triggering frequency and the preset second change rate.

[0014] In an alternative manner, the intelligent optimization module is used to determine whether the operating efficiency of the alarm system meets the requirements according to the time interval distribution of alarm events in the log storage unit. If the standard deviation of the time interval distribution of the alarm events is greater than the preset first standard deviation, it is determined that the operating efficiency of the alarm system does not meet the requirements.

[0015] In an alternative manner, the intelligent optimization module is used to preliminarily determine that the data processing capacity of the alarm system does not meet the requirements when the standard deviation of the time interval distribution of the alarm events is greater than the preset second standard deviation, and determine whether the data processing capacity of the alarm system meets the requirements according to the instantaneous change rate of the server load.

[0016] In an alternative manner, the intelligent optimization module is used to shorten the data sampling period of the alarm system when the standard deviation of the time interval distribution of the alarm events is greater than the preset first standard deviation and less than or equal to the preset second standard deviation.

[0017] In an alternative manner, the reduction amplitude of the data sampling period of the alarm system is determined by the difference between the standard deviation of the time interval distribution of the alarm events and the preset first standard deviation.

[0018] In an alternative manner, when the intelligent optimization module shortens the data sampling period, it synchronously monitors the CPU occupancy rate of the server device. When the CPU occupancy rate exceeds the preset load threshold, a reduction coefficient is used to dynamically attenuate the reduction amplitude, where the reduction coefficient is negatively correlated with the percentage by which the CPU occupancy rate exceeds the preset load threshold.

[0019] In an alternative manner, the alarm trigger module further includes a multi-level notification unit. When the priority of the alarm trigger is increased, the multi-level notification unit activates the corresponding alarm channel combination according to the priority level, including a linkage response that simultaneously triggers SMS notifications, email notifications, and visual panel warning signs.

[0020] Compared with the prior art, an alarm system for a server device according to the present invention has the following beneficial effects:

[0021] The alarm system of the present invention adjusts the sensitivity of the thermal sensor according to the change rate of the alarm trigger frequency by setting up an operating state monitoring module, a data analysis module, an alarm trigger module, and an intelligent optimization module. Since the ambient temperature fluctuates greatly during the operation of the server, it may cause misjudgment by the thermal sensor. By reducing the sensitivity of the thermal sensor, false alarms caused by small changes in the ambient temperature are reduced, and the accuracy of the alarm is improved; the data sampling period of the alarm system is adjusted according to the time interval distribution of the alarm events in the log storage unit. Since the server operation scenario is complex, the data sampling frequency may be too high or too low. By shortening the data sampling period, abnormal events can be captured more timely, and the real-time performance of the alarm is improved; the priority of the alarm trigger is adjusted according to the instantaneous change rate of the server load. Since the server load fluctuation may affect the processing speed of the alarm information, by increasing the priority of the alarm trigger, it is ensured that critical alarm information can be processed preferentially, thereby enhancing the reliability and intelligence level of the system.

[0022] In addition, the present invention uses the following formula to calculate the alarm trigger priority: ; where represents the adjusted alarm trigger priority, represents the initial priority, represents the priority adjustment coefficient, represents the instantaneous change rate of the server load, represents a preset load change rate threshold. By introducing the ratio of the load change rate to the threshold, this formula dynamically adjusts the alarm priority to ensure that critical alarm information can be processed preferentially in the case of high load fluctuations, while avoiding unnecessary resource waste.

[0023] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0025] Figure 1 is a schematic structural diagram of an alarm system for server devices according to the present invention;

[0026] Figure 2 is a schematic structural diagram of the operation status monitoring module;

[0027] Figure 3 is a schematic overall logic diagram;

[0028] Figure 4 is a schematic diagram of the parameter relationship of the alarm trigger priority calculation formula. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0030] Figure 1 shows a schematic structural diagram of an embodiment of an alarm system for server devices provided by the present invention. As Figure 1 shown, the alarm system includes: an operation status monitoring module 110, a data analysis module 120, an alarm trigger module 130, and an intelligent optimization module 140.

[0031] The operating status monitoring module 110 is used to collect the operating parameters of the server device in real time through a sensor group. The sensor group includes a thermal sensor arranged inside the server to detect temperature changes and a traffic sensor connected to the server to detect network traffic fluctuations. Among them, the server includes at least one server device;

[0032] The data analysis module 120 is connected to the operating status monitoring module 110 and is used to extract features and perform anomaly analysis on the collected operating parameters;

[0033] The alarm trigger module 130 is respectively connected to the operating status monitoring module 110 and the data analysis module 120 and is used to send an alarm signal when the operating parameters exceed the set threshold. The alarm trigger module 130 includes an alarm unit for generating an acoustic and optical alarm signal and a log storage unit for recording alarm events;

[0034] The intelligent optimization module 140 is respectively connected to the operating status monitoring module 110, the data analysis module 120, and the alarm trigger module 130, and is used to adjust the sensitivity of the thermal sensor according to the change rate of the alarm trigger frequency or adjust the data sampling period of the alarm system according to the time interval distribution of the alarm events in the log storage unit, and determine the priority of alarm triggering according to the network traffic fluctuation amplitude and the instantaneous change rate of the server load. Among them, the network traffic fluctuation amplitude is: the difference between the maximum value and the minimum value in the network traffic time series collected by the traffic sensor within a preset time period; when the network traffic fluctuation amplitude exceeds the preset fluctuation amplitude threshold, determine whether the data processing capacity of the alarm system meets the requirements according to the instantaneous change rate of the server load. If the instantaneous change rate of the server load is greater than the preset change rate threshold, it is determined that the data processing capacity of the alarm system does not meet the requirements, and the priority of alarm triggering is increased. The increase amplitude of the priority of alarm triggering is determined by the difference between the instantaneous change rate of the server load and the preset change rate threshold.

[0035] Specifically, when the change rate of the alarm trigger frequency (such as the growth rate of the number of alarms per unit time) exceeds the preset threshold, the system will reduce the sensitivity of the thermal sensor. For example, if the preset change rate is 0.1 times per minute, but the actual change rate reaches 0.15 times per minute, the sensitivity is reduced to reduce false alarms caused by small temperature fluctuations. If the standard deviation of the time interval distribution of alarm events is large (such as the interval time fluctuates violently), the sampling period is shortened to improve real-time performance. For example, when the standard deviation increases from 0.5 seconds to 0.8 seconds, the sampling period is shortened from 1 second to 0.8 seconds. When the instantaneous change rate of the server load exceeds the threshold (such as 10%), the alarm priority is increased. For example, when the load suddenly soars from 30% to 50%, a high-priority alarm (such as SMS + email + acoustic and optical linkage) is triggered.

[0036] In an alternative manner, the intelligent optimization module 140 is configured to determine whether the dynamic adaptability of the alarm system meets the requirements according to the change rate of the alarm trigger frequency. If the change rate of the alarm trigger frequency is greater than a preset first change rate, it is determined that the dynamic adaptability of the alarm system does not meet the requirements.

[0037] Specifically, the preset first change rate is the upper limit of the allowable alarm frequency fluctuation of the system. The first change rate is default set to 0.05 times / minute. If the actual change rate reaches 0.07 times / minute, it is determined that the dynamic adaptability is insufficient. For example, due to the aging of the cooling fan of a certain server, the temperature fluctuates frequently, triggering multiple alarms. At this time, it is recognized that the alarm frequency increases abnormally, triggering an optimization mechanism (such as reducing the sensitivity or adjusting the sampling period).

[0038] In an alternative manner, when the change rate of the alarm trigger frequency is greater than the preset first change rate and less than or equal to a preset second change rate, the intelligent optimization module 140 is configured to preliminarily determine that the operation efficiency of the alarm system does not meet the requirements, and determine whether the operation efficiency of the alarm system meets the requirements according to the time interval distribution of the alarm events in the log storage unit.

[0039] Specifically, when the change rate of the alarm trigger frequency is within a preset range (such as 0.05 < R ≤ 0.1 times / minute), a secondary verification is required in combination with the standard deviation of the time interval distribution. Assuming that the standard deviation of the interval time of the alarm events is 0.6 seconds (the preset first standard deviation is 0.5 seconds), it is confirmed that the operation efficiency does not meet the requirements, and the data sampling period adjustment is started.

[0040] In an alternative manner, when the change rate of the alarm trigger frequency is greater than the preset second change rate, the intelligent optimization module 140 is configured to reduce the sensitivity of the thermal sensor, and the reduction amplitude of the sensitivity of the thermal sensor is determined by the difference between the change rate of the alarm trigger frequency and the preset second change rate.

[0041] Specifically, , represents the reduction amplitude of the sensitivity, is the sensitivity adjustment coefficient, is the change rate of the alarm trigger frequency, is the preset second change rate. For example, if = 0.1 times / minute, the actual = 0.12 times / minute, = 0.5, then = 0.5×(0.12 - 0.1) = 1%. The sensitivity is adjusted from ±0.1°C to ±0.101°C, reducing false alarms caused by small temperature fluctuations.

[0042] In an optional manner, the intelligent optimization module 140 is used to determine whether the operating efficiency of the alarm system meets the requirements based on the time interval distribution of the alarm events in the log storage unit. If the standard deviation of the time interval distribution of the alarm events is greater than a preset first standard deviation, it is determined that the operating efficiency of the alarm system does not meet the requirements.

[0043] Specifically, the larger the time interval distribution of the alarm event, the more violent the alarm interval fluctuation. For example, if the first standard deviation is preset to 0.5 seconds and the actual standard deviation is 0.7 seconds, it means that the alarm interval is unstable (such as frequent short interval alarms), and the sampling period needs to be shortened to improve the response speed.

[0044] In an optional manner, the intelligent optimization module 140 is used to preliminarily determine that the data processing capacity of the alarm system does not meet the requirements when the standard deviation of the time interval distribution of the alarm event is greater than a preset second standard deviation, and to determine whether the data processing capacity of the alarm system meets the requirements based on the instantaneous change rate of the server load.

[0045] Specifically, when the standard deviation exceeds the threshold, the server load instantaneous change rate is checked to see if it is abnormal. For example, if the load change rate suddenly increases to 15% (the threshold is 10%), it is confirmed that the data processing capacity is insufficient and the alarm priority needs to be increased.

[0046] In an optional manner, the intelligent optimization module 140 is used to shorten the data sampling period of the alarm system when the standard deviation of the time interval distribution of the alarm event is greater than the preset first standard deviation and less than or equal to the preset second standard deviation, wherein the shortening extent of the data sampling period of the alarm system is determined by the difference between the standard deviation of the time interval distribution of the alarm event and the preset first standard deviation.

[0047] Specifically, , Indicates the shortening of the data sampling period. is the sampling period adjustment coefficient, is the standard deviation of the alarm event time interval distribution, is the preset first standard deviation. = 0.7 seconds, = 0.5 seconds, =0.2, then =0.2×(0.7-0.5)=0.04 seconds, and the sampling period is shortened from 1 second to 0.96 seconds.

[0048] In an optional manner, the intelligent optimization module 140 is used to determine whether the data processing capacity of the alarm system meets the requirements based on the instantaneous change rate of the server load. If the instantaneous change rate of the server load is greater than a preset change rate threshold, it is determined that the data processing capacity of the alarm system does not meet the requirements, and the priority of the alarm trigger is increased, wherein the increase in the priority of the alarm trigger is determined by the difference between the instantaneous change rate of the server load and the preset change rate threshold.

[0049] Specifically, Calculate; where, Indicates the adjusted alarm trigger priority. Indicates the initial priority, represents the priority adjustment coefficient, Indicates the instantaneous rate of change of server load. Indicates the preset load change rate threshold. =15%, =10%, =0.5, then = +0.5×(1.5-1)= +0.25, triggering a higher level alarm channel (such as starting SMS and phone notifications at the same time).

[0050] In an optional manner, the intelligent optimization module 140 synchronously monitors the CPU occupancy rate of the server device while shortening the data sampling period. When the CPU occupancy rate exceeds a preset load threshold, a reduction coefficient is used to dynamically attenuate the shortening amplitude, wherein the reduction coefficient is negatively correlated with the percentage by which the CPU occupancy rate exceeds the preset load threshold.

[0051] Specifically, when the CPU usage reaches 80% (the preset load threshold is 70%), the reduction factor is negatively correlated with the over-limit ratio. For example, when the over-limit is 10%, the reduction factor is 0.8, and the actual shortening range ΔT is adjusted from 0.04 seconds to 0.04×0.8=0.032 seconds, avoiding increasing the CPU burden.

[0052] In an optional manner, the alarm trigger module 130 also includes a multi-level notification unit. When the priority of the alarm trigger is increased, the multi-level notification unit activates the corresponding alarm channel combination according to the priority level, including simultaneously triggering a linkage response of SMS notification, email notification and visual panel warning sign.

[0053] Among them, the hierarchical alarm strategy is as follows: ① Low priority: only trigger visual panel alarm. ② Medium priority: email notification + panel alarm. ③ High priority: SMS + email + sound and light linkage + phone notification.

[0054] In this embodiment, it should be specifically noted that:

[0055] The operation status monitoring module 110 is the core data acquisition part of the entire alarm system, and its structure is as Figure 2 shown. The sensor group in the operation status monitoring module 110 includes a thermal sensor and a flow sensor, which are respectively used to detect the internal temperature change of the server and the network traffic fluctuation. The thermal sensor is arranged near the main heat-generating components inside the server, such as near the CPU and GPU, to ensure that the temperature changes in the key areas can be captured in real time. The flow sensor is connected to the server network interface to monitor the data traffic fluctuation in and out of the server. In practical applications, the thermal sensor uses a high-precision NTC thermistor, and its sensitivity can be dynamically adjusted according to environmental requirements; the flow sensor selects a hardware module that supports the gigabit Ethernet protocol and can adapt to the requirements of different network bandwidths. The operation status monitoring module 110 obtains temperature and flow data through periodic sampling and transmits these data to the data analysis module 120 for further processing.

[0056] The data analysis module 120 receives the raw data from the operation status monitoring module 110 and performs feature extraction and anomaly analysis on it. In the specific implementation process, the data analysis module 120 first preprocesses the collected temperature and flow data, including filtering and denoising and normalization operations, to eliminate the influence of external interference on the data quality. Subsequently, the data analysis module 120 uses machine learning algorithms to extract features from the data and identify possible abnormal patterns. For example, when the internal temperature of the server fluctuates violently (the temperature change amplitude exceeds ±3°C) within a short period (the default setting is 10s), the data analysis module 120 will mark it as a potential abnormal event; similarly, when the network traffic fluctuation amplitude exceeds the preset threshold, it will also be recorded as an abnormal event. The data analysis module 120 also judges whether the current operation parameters deviate from the normal range through historical data comparison analysis. If an anomaly is found, the relevant information will be transmitted to the alarm trigger module 130. Among them, the preset (fluctuation) threshold can be set according to the actual situation and is not limited here. For example, the fluctuation threshold of the thermal sensor can be set to 5°C, and the fluctuation threshold of the flow sensor can be set to 1Gbps.

[0057] The alarm trigger module 130 generates an alarm signal based on the abnormal information provided by the data analysis module 120 and records the relevant alarm events. The alarm trigger module 130 includes an alarm unit and a log storage unit. The alarm unit is responsible for generating an audible and visual alarm signal. The audible alarm signal is emitted through a buzzer, and the visual alarm signal is achieved by the flashing of an LED light. In practical applications, the design of the alarm unit fully considers the human-computer interaction requirements. For example, the volume of the buzzer is adjustable, and the color of the LED light can be switched to red or yellow according to the alarm level. The log storage unit is used to record the timestamp, abnormal type, and corresponding parameter values of each alarm event, facilitating subsequent fault troubleshooting and system optimization. The working process of the alarm trigger module 130 is as follows: After receiving the abnormal information transmitted by the data analysis module 120, the alarm trigger module 130 first determines whether the current operating parameters exceed the set thresholds (for example, the set threshold of the thermal sensor can be set to 60 °C, and the set threshold of the flow sensor can be set to 2 Gbps). If it exceeds, the audible and visual alarm signal is immediately activated, and the alarm event is written into the log storage unit.

[0058] The intelligent optimization module 140 is one of the key innovations of the present invention. The intelligent optimization module 140 dynamically adjusts various parameters of the alarm system by comprehensively analyzing the change rate of the alarm trigger frequency, the time interval distribution of alarm events, and the instantaneous change rate of the server load, thereby improving the reliability and intelligence level of the system. As Figure 3 shown, in the specific implementation process, the intelligent optimization module 140 first calculates the change rate of the alarm trigger frequency and compares it with the preset first change rate and second change rate (the first change rate is less than the second change rate). If the change rate of the alarm trigger frequency is greater than the first change rate but less than or equal to the second change rate, it is initially determined that the operating efficiency of the alarm system does not meet the requirements, and the time interval distribution of alarm events in the log storage unit is further analyzed. If the change rate of the alarm trigger frequency is greater than the second change rate, the sensitivity of the thermal sensor is reduced, and the reduction amplitude is calculated by the formula where represents the sensitivity reduction amplitude, is the sensitivity adjustment coefficient, is the change rate of the alarm trigger frequency, is the preset second change rate. For example, assuming that the calculated change rate of the alarm trigger frequency is 0.12 times / minute, the first change rate is set to 0.05 times / minute, the second change rate is set to 0.1 times / minute, is defaulted to 0.5, then since , the calculated value according to the formula is , the sensitivity of the thermal sensor decreases by 1%. By the above method, false alarms caused by small changes in the ambient temperature can be effectively reduced, and the accuracy of the alarm can be improved.

[0059] For the analysis of the distribution of alarm event time intervals, the intelligent optimization module 140 calculates its standard deviation and compares it with the preset first standard deviation and second standard deviation. If the standard deviation is greater than the first standard deviation but less than or equal to the second standard deviation, the data sampling period of the alarm system is shortened, and the shortening amplitude is calculated by the formula where represents the shortening amplitude of the data sampling period, is the sampling period adjustment coefficient, is the standard deviation of the alarm event time interval distribution, is the preset first standard deviation. For example, assuming that the standard deviation of the alarm event time interval distribution obtained by statistics is 0.89 s, the first standard deviation is set to 0.5 s, and the second standard deviation is set to 0.5 s, is defaulted to 0.2, then since , then , the shortening amplitude of the data sampling period is 0.078 s. By shortening the data sampling period, abnormal events can be captured more timely, and the real-time performance of the alarm can be improved. If the standard deviation is greater than the second standard deviation, the instantaneous change rate of the server load is further analyzed.

[0060] The analysis of the instantaneous change rate of the server load is another important function of the intelligent optimization module 140. When the instantaneous change rate of the server load exceeds the preset change rate threshold, the intelligent optimization module 140 will increase the priority of alarm triggering, and the increase amplitude is calculated by the formula where; represents the adjusted alarm triggering priority, represents the initial priority, represents the priority adjustment coefficient, represents the instantaneous change rate of the server load (such as 15%), represents the preset load change rate threshold (such as 10%). The physical meaning of this formula is that when the server load fluctuates greatly, by dynamically adjusting the alarm priority, it is ensured that key alarm information can be processed first, and unnecessary resource waste is avoided. As Figure 4 shows, the relationship between the variables in the formula clearly demonstrates the non-linear mapping relationship between the load change rate and the alarm priority. It should be noted that different priorities correspond to different alarm methods. For example, high-priority alarms are notified through multiple channels, and the operation and maintenance personnel respond within 5 seconds to avoid the risk of downtime.

[0061] In actual application scenarios, the alarm system of the present invention can be deployed in the server cluster of a data center to monitor the operating status of multiple servers. For example, in a large-scale cloud computing center, the number of server devices is huge and the operating environment is complex, and traditional alarm systems are difficult to meet the requirements of dynamic adaptability and multi-dimensional data fusion. By deploying the alarm system of the present invention, the operating status monitoring module 110 can collect the temperature and traffic data of each server in real time, and the data analysis module 120 deeply mines these data to identify potential anomalies. When the temperature of a certain server suddenly rises or the network traffic fluctuates abnormally, the alarm trigger module 130 will immediately issue an audible and visual alarm signal and record the relevant information in the log storage unit. At the same time, the intelligent optimization module 140 dynamically adjusts the sensitivity of the thermal sensor, the data sampling period, and the alarm trigger priority according to the change rate of the alarm trigger frequency, the distribution of the alarm event time interval, and the instantaneous change rate of the server load, so as to ensure that the alarm system can operate efficiently in a complex environment.

[0062] In addition, the alarm system of the present invention also has good scalability and compatibility. For example, the operating status monitoring module 110 can add other types of sensors according to actual needs, such as humidity sensors or voltage sensors, to achieve the monitoring of more operating parameters. The data analysis module 120 supports the integration of multiple machine learning algorithms and can select the optimal feature extraction and anomaly analysis methods according to different application scenarios. The audible and visual alarm signal design of the alarm trigger module 130 is flexible and can be customized with different alarm modes according to user needs. The parameter adjustment strategy of the intelligent optimization module 140 can also be optimized according to specific application scenarios. For example, in low-load scenarios, the sensitivity adjustment conditions are appropriately relaxed, while in high-load scenarios, more attention is paid to the dynamic adjustment of the alarm priority.

[0063] The alarm system of this embodiment realizes the comprehensive monitoring and intelligent management of the operating status of server devices through the collaborative work of the operating status monitoring module 110, the data analysis module 120, the alarm trigger module 130, and the intelligent optimization module 140. Its core advantage lies in being able to dynamically adjust the sensitivity of the thermal sensor, the data sampling period, and the alarm trigger priority according to the change rate of the alarm trigger frequency, the distribution of the alarm event time interval, and the instantaneous change rate of the server load, thereby significantly improving the accuracy, real-time performance, and reliability of the alarm. In actual applications, this system can not only effectively reduce false alarms and missed alarms, but also ensure that key alarm information can be processed preferentially, providing a strong guarantee for the safe and stable operation of server devices.

[0064] The above description is only a preferred embodiment of the present invention and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present invention.

[0065] It should be noted that the terms "first", "second", etc. in the specification and claims of this application are used to distinguish similar objects, and represent a limitation on a specific order or sequence. Under appropriate circumstances, the order of use of similar objects can be interchanged so that the embodiments of this application described here can be implemented in an order other than the illustrated or described order.

[0066] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. An alarm system for server equipment, characterized in that: include: An operation status monitoring module, for collecting operation parameters of a server device in real time through a sensor group, wherein the sensor group includes a thermistor disposed inside the server for detecting temperature changes and a flow sensor connected to the server for detecting fluctuations in network flow; wherein the server includes at least one server device; A data analysis module, connected to the operation status monitoring module, for performing feature extraction and abnormality analysis on the collected operation parameters; An alarm trigger module, connected to the operation status monitoring module and the data analysis module respectively, for sending an alarm signal when the operation parameter exceeds a set threshold value, the alarm trigger module includes an alarm unit for generating an audible and visual alarm signal and a log storage unit for recording alarm events; The intelligent optimization module is respectively connected to the operation status monitoring module, the data analysis module and the alarm triggering module, and is used to adjust the sensitivity of the thermistor according to the change rate of the alarm triggering frequency or adjust the data sampling period of the alarm system according to the time interval distribution of the alarm events in the log storage unit, and determine the priority of the alarm triggering according to the network traffic fluctuation amplitude and the instantaneous change rate of the server load; wherein the network traffic fluctuation amplitude is: the difference between the maximum value and the minimum value in the network traffic time series collected by the traffic sensor within a preset time period; when the network traffic fluctuation amplitude exceeds the preset fluctuation amplitude threshold, determine whether the data processing capacity of the alarm system meets the requirements according to the instantaneous change rate of the server load, if the instantaneous change rate of the server load is greater than the preset change rate threshold, determine that the data processing capacity of the alarm system does not meet the requirements, and increase the priority of the alarm triggering, wherein the increase in the priority of the alarm triggering is determined by the difference between the instantaneous change rate of the server load and the preset change rate threshold.

2. The alarm system for server equipment according to claim 1, characterized in that: The intelligent optimization module is used to determine whether the dynamic adaptability of the alarm system meets the requirements according to the change rate of the alarm trigger frequency. If the change rate of the alarm trigger frequency is greater than a preset first change rate, it is determined that the dynamic adaptability of the alarm system does not meet the requirements.

3. The alarm system for server equipment according to claim 2, characterized in that: The intelligent optimization module is used to preliminarily determine that the operating efficiency of the alarm system does not meet the requirements when the change rate of the alarm trigger frequency is greater than the preset first change rate and less than or equal to the preset second change rate, and to determine whether the operating efficiency of the alarm system meets the requirements based on the time interval distribution of the alarm events in the log storage unit.

4. The alarm system for server equipment according to claim 3, characterized in that: The intelligent optimization module is used to reduce the sensitivity of the thermistor when the change rate of the alarm trigger frequency is greater than the preset second change rate, wherein the reduction in the sensitivity of the thermistor is determined by the difference between the change rate of the alarm trigger frequency and the preset second change rate.

5. The alarm system for server equipment according to claim 1, characterized in that: The intelligent optimization module is used to determine whether the operating efficiency of the alarm system meets the requirements based on the time interval distribution of the alarm events in the log storage unit. If the standard deviation of the time interval distribution of the alarm events is greater than a preset first standard deviation, it is determined that the operating efficiency of the alarm system does not meet the requirements.

6. The alarm system for server equipment according to claim 5, characterized in that: The intelligent optimization module is used to preliminarily determine that the data processing capacity of the alarm system does not meet the requirements when the standard deviation of the time interval distribution of the alarm event is greater than a preset second standard deviation, and to determine whether the data processing capacity of the alarm system meets the requirements based on the instantaneous change rate of the server load.

7. The alarm system for server equipment according to claim 6, characterized in that: The intelligent optimization module is used to shorten the data sampling period of the alarm system when the standard deviation of the time interval distribution of the alarm event is greater than the preset first standard deviation and less than or equal to the preset second standard deviation; wherein the extent of shortening of the data sampling period of the alarm system is determined by the difference between the standard deviation of the time interval distribution of the alarm event and the preset first standard deviation.

8. The alarm system for server equipment according to claim 7, characterized in that: The intelligent optimization module synchronously monitors the CPU occupancy rate of the server device when shortening the data sampling period. When the CPU occupancy rate exceeds a preset load threshold, a reduction coefficient is used to dynamically attenuate the shortening amplitude, wherein the reduction coefficient is negatively correlated with the percentage by which the CPU occupancy rate exceeds the preset load threshold.

9. The alarm system for server equipment according to claim 1, characterized in that: The alarm trigger module also includes a multi-level notification unit. When the priority of the alarm trigger is increased, the multi-level notification unit activates the corresponding alarm channel combination according to the priority level, including simultaneously triggering a linkage response of SMS notification, email notification and visual panel warning mark.

Images