Data center troubleshooting method and system, terminal and storage medium

Abstract

The invention relates to a troubleshooting method and system for a data center, a terminal and a storage medium, and belongs to the technical field of data processing.The troubleshooting method comprises the steps that a fault patrol track is constructed according to the number and positions of devices in the data center; controlling a patrol camera to perform periodic patrol on the equipment according to the fault patrol track; acquiring a patrol video stream collected by a patrol camera in a patrol process; judging whether the data center is abnormal or not according to the patrol video stream; and if yes, positioning an abnormal position, and sending alarm information. In addition to monitoring of operation data in the equipment, the inspection camera can be controlled to perform periodic inspection according to a fault inspection track by constructing the fault inspection track, so that whether the data center is abnormal or not is judged according to the obtained inspection video stream, the abnormal position is positioned, and alarm information is sent. Therefore, the monitoring comprehensiveness of the data center is improved, and the fault position can be positioned in time.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a data center troubleshooting method, system, terminal and storage medium. Background Art

[0002] Data centers are facilities used to centrally store, manage, and distribute data, usually including a large number of servers, storage devices, and network devices. They play a vital role in modern information technology architecture and are widely used in cloud computing, website hosting, enterprise data management, and other fields.

[0003] Among the related technologies, the monitoring technology for data centers is becoming more and more mature. Through equipment and service data such as equipment usage data, operating condition data, host and accessory performance data, and accessory replacement data, it can predict equipment failure, service, and accessory requirements, provide technical support for proactive services, extend equipment life, and reduce failure rates.

[0004] However, most of the current monitoring is focused on the internal data of the equipment, and it is not possible to monitor the external data of the equipment. Therefore, if the data center is abnormal due to external factors of the equipment, it is easy to fail to discover it in time, and the fault cannot be repaired in time. Summary of the invention

[0005] In order to improve the comprehensiveness of data center monitoring and locate faults in a timely manner, the present application provides a data center fault troubleshooting method, system, terminal and storage medium.

[0006] In a first aspect, the present application provides a data center troubleshooting method, which adopts the following technical solution: A data center troubleshooting method, comprising: Build a fault inspection track based on the number and location of equipment in the data center; Controlling the patrol camera to perform periodic patrol of the data center according to the fault patrol track; Obtaining the patrol video stream collected by the patrol camera during the patrol process; Determine whether there is an abnormality in the data center according to the patrol video stream; If so, locate the abnormal position and send an alarm message.

[0007] By adopting the above technical solution, in addition to monitoring the operating data in the equipment, it is also possible to construct a fault patrol trajectory to control the patrol camera to perform periodic patrols according to the fault patrol trajectory, so as to determine whether there is an abnormality in the data center based on the acquired patrol video stream, and then locate the abnormal position and send an alarm message to realize external monitoring of the equipment, thereby improving the comprehensiveness of data center monitoring and being able to locate the fault in time.

[0008] Optionally, after sending the alarm information, the following steps are included: Retrieving a pre-established fault influencing factor table, wherein the fault influencing factor table can be updated in real time; Searching for device information associated with the abnormal location from the fault influencing factor table, the device information including device environment data, operating time, and maintenance time; Analyze the cause of the fault based on the device information.

[0009] By adopting the above technical solution, a fault influencing factor table is established, and the environmental data and operating time in the fault influencing factor table are updated in real time. After the maintenance personnel maintain the equipment, the maintenance time is updated; after the equipment fails, the cause of the failure can be analyzed through the environmental data, operating time and maintenance time currently associated with the equipment, thereby improving the efficiency of fault cause analysis.

[0010] Optionally, the troubleshooting method further includes: According to the fault influencing factor table, key locations where faults frequently occur are obtained; Get the real-time location of patrol cameras; Obtaining the area distance between the real-time position and the key position in the advancing direction of the current track of the patrol camera; Determining whether the area spacing is within a spacing threshold; If yes, then reduce the movement speed of the patrol camera; After the patrol camera passes the key position and the area spacing is out of the spacing threshold, the moving speed of the patrol camera is restored.

[0011] By adopting the above technical solution, when the patrol camera approaches the key position, it means that the patrol camera is moving towards the place where faults frequently occur. Therefore, by reducing the moving speed of the patrol camera, the time for the patrol camera to collect the patrol video stream at the place where faults frequently occur can be indirectly extended, thereby improving the accuracy of troubleshooting.

[0012] Optionally, the step of reducing the moving speed of the patrol camera includes: Modeling velocity changes ; Wherein, v represents the moving speed of the patrol camera, d represents the area spacing, d0 represents the spacing threshold, k represents a normal number, representing the maximum possible value of the speed, and a represents a normal number, representing the steepness of the model curve; The moving speed is reduced according to the speed change model.

[0013] Optionally, the step of obtaining key locations where faults frequently occur includes: Determine the number of faults associated with the faulty device through the fault influencing factor table; Devices whose fault times are greater than a preset threshold are screened, and the locations corresponding to the screened devices are the key locations.

[0014] By adopting the above technical solution, the location of the device where the number of failures exceeds the number threshold is the key location where failures frequently occur.

[0015] Optionally, the steps after obtaining the key positions of the places where faults frequently occur include: Determining a troubleshooting level associated with the key location; According to the troubleshooting level, a spacing threshold corresponding to the key position is assigned; the higher the troubleshooting level, the larger the corresponding spacing threshold.

[0016] By adopting the above technical solution, the corresponding spacing threshold can be adaptively adjusted according to the fault troubleshooting level, thereby further improving the accuracy of fault troubleshooting.

[0017] Optionally, the step of determining whether there is an abnormality in the data center according to the patrol video stream includes: Converting the patrol video stream frame by frame into patrol pictures; Eliminate redundant patrol pictures with high similarity to generate a patrol picture set; Inputting the patrol picture set into a pre-trained neural network model to obtain a judgment result; According to the judgment result, it is determined whether the data center is abnormal.

[0018] By adopting the above technical solution, redundant patrol pictures with high similarity are eliminated to achieve picture data noise cleaning, thereby reducing interference with picture judgment and improving judgment efficiency.

[0019] In a second aspect, the present application provides a data center troubleshooting system, which adopts the following technical solutions: A data center fault detection system, comprising: The trajectory building module is used to build fault inspection trajectories based on the data and location of the equipment in the data center; A control module, used to control the patrol camera to perform periodic patrol of the equipment according to the fault patrol track; A video acquisition module is used to acquire the patrol video stream collected by the patrol camera during the patrol process; A judgment module, used for judging whether there is an abnormality in the data center according to the patrol video stream; The abnormal location alarm module is used to locate the abnormal location and send alarm information when the data center is abnormal.

[0020] By adopting the above technical solution, in addition to monitoring the operating data in the equipment, it is also possible to construct a fault patrol trajectory to control the patrol camera to perform periodic patrols according to the fault patrol trajectory, so as to determine whether there is an abnormality in the data center based on the acquired patrol video stream, and then locate the abnormal position and send an alarm message to realize external monitoring of the equipment, thereby improving the comprehensiveness of data center monitoring and being able to locate the fault in time.

[0021] In a third aspect, the present application provides a terminal, which adopts the following technical solution: A terminal, comprising: A memory storing a troubleshooting program of a data center; The processor is used to execute the program stored in the memory to implement the steps of the above-mentioned data center troubleshooting method.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute the above-mentioned data center troubleshooting method.

[0023] In summary, the present application has at least the following beneficial effects: 1. Construct a fault patrol track and control the patrol camera to conduct periodic patrols of the equipment according to the fault patrol track. The purpose is not only to monitor the operating data in the equipment, but also to determine whether there is an abnormality in the data center based on the acquired patrol video stream, and then locate the abnormal location and send an alarm message to realize external monitoring of the equipment, thereby improving the comprehensiveness of data center monitoring and being able to locate the fault in time.

[0024] 2. The purpose of obtaining the regional distance between the real-time position of the patrol camera and the key position of the fault-frequently occurring location in the direction of the current trajectory of the patrol camera, and judging whether the regional distance is within the distance threshold is that when the patrol camera approaches the key position, it indicates that the patrol camera is moving towards the fault-frequently occurring location. Therefore, by reducing the movement speed of the patrol camera, the duration of the patrol camera collecting the patrol video stream at the fault-frequently occurring location can be indirectly extended, thereby improving the accuracy of troubleshooting.

[0025] 3. Determine the troubleshooting level associated with the key location, and assign the spacing threshold corresponding to the key location according to the troubleshooting level. The purpose is to adaptively adjust the spacing threshold corresponding to the fault-frequently occurring location, thereby further improving the accuracy of troubleshooting. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flowchart of an implementation method of Example 1 of the present application; Figure 2 It is a flowchart of a specific step of S140; Figure 3 It is a flowchart of another implementation method of the method embodiment of the present application; Figure 4 It is a structural block diagram of an embodiment of the system of the present application.

[0027] Explanation of the accompanying drawings: 101, trajectory construction module; 102, control module; 103, video acquisition module; 104, judgment module; 105, abnormal positioning alarm module; 106, video to image module; 107, image rejection module; 108, neural network module; 109, position acquisition module; 110, spacing acquisition module. DETAILED DESCRIPTION

[0028] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the following will be combined with the appended drawings of the embodiments of the present invention. Figure 1 -Attached Figure 4 , the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] The first embodiment of the present application discloses a method for troubleshooting a data center. Figure 1 As an implementation of the troubleshooting method, the troubleshooting method may include S110-S160: S110, constructing a fault inspection trajectory according to the number and location of devices in the data center; S120, controlling the patrol camera to perform periodic patrol of the data center according to the fault patrol track; S130, obtaining a patrol video stream collected by a patrol camera during the patrol process; S140, judging whether there is an abnormality in the data center according to the patrol video stream; S150, if yes, locate the abnormal position and send an alarm message; S160, if not, return to S130.

[0030] Specifically, the number, type and location of the equipment arranged in the data center can be obtained according to the floor plan of the data center, and then the fault patrol trajectory can be planned and constructed according to the corresponding position of the equipment. Then, the patrol camera is driven to move according to the planned patrol trajectory through the pre-arranged driving structure. The driving structure is a conventional structure. As long as the movement and rotation of the patrol camera can be realized, it will not be expanded in detail. According to the patrol video stream collected by the patrol camera during the movement, it is judged whether the data center is abnormal. If it is abnormal, the abnormal position is located and the alarm information is sent to the background. The alarm information includes the abnormal position, the type of abnormal equipment, the possible cause of the abnormality, etc. The patrol camera conducts periodic patrols, and the period can be set by the management personnel through the background, for example, once every 3 hours.

[0031] Reference Figure 2 For S140, a specific step may include S141-144: S141, converting the patrol video stream frame by frame into patrol pictures; S142, removing redundant patrol pictures with high similarity to generate a patrol picture set; S143, inputting the pictures in the patrol picture set into a pre-trained neural network model to obtain a judgment result; S144, determining whether the data center is abnormal based on the judgment result.

[0032] Specifically, the patrol video stream can be continuously screenshotted to convert it into patrol pictures frame by frame. The patrol pictures are initially arranged according to the shooting time sequence, and the adjacent patrol pictures are compared to determine whether they are similar. If they are similar, one is retained and the rest are removed. Then, the pictures in the patrol picture set are arranged again according to the shooting time sequence to generate a patrol picture set. Then, the pictures in the patrol picture set are sequentially input into the pre-trained neural network model according to the shooting time sequence to obtain the judgment result, and then determine whether the data center is abnormal based on the judgment result. When determining that the data center is abnormal, the abnormal position can be located based on the shooting time of the corresponding patrol picture and the location of the patrol camera.

[0033] Furthermore, after sending the alarm information, it is necessary to retrieve the pre-established fault influencing factor table, which can be regarded as an online table that can be updated in real time. The fault influencing factor table includes factors such as equipment location, number, quantity, environmental data, operating time, maintenance time, number of faults, fault time, fault investigation level, etc. After a certain equipment fails, the related equipment information can be quickly found in the fault influencing factor table according to the abnormal location, so as to analyze the cause of the failure according to the equipment information; for example, it is caused by abnormal environmental data or because the equipment has been running for a long time or because the equipment has not been maintained recently.

[0034] Reference Figure 3 When the patrol camera collects the patrol video stream, the troubleshooting method may further include S210-S270: S210, obtaining key locations where faults frequently occur according to the fault influencing factor table; S220, obtaining the real-time position of the patrol camera; S230, obtaining the area distance between the real-time position and the key position in the advancing direction of the current track of the patrol camera; S240, determining whether the area spacing is within a spacing threshold; S250, if yes, reduce the movement speed of the patrol camera; S260, if not, return to S240; S270, after the patrol camera passes the key position and the area spacing is out of the spacing threshold, restore the movement speed of the patrol camera.

[0035] Specifically, the number of faults associated with the faulty device can be determined through the fault influencing factor table, and then the devices with a number of faults greater than the preset number threshold can be screened, and the corresponding position of the screened device is the key position. When the patrol camera moves, the position information will be sent in real time. Therefore, the real-time position of the patrol camera can be obtained, and the movement trajectory of the patrol camera can be determined according to the real-time position, and then the key position of the frequent faults in the movement trajectory can be screened, and then the regional spacing between the real-time position and the key position in the forward direction can be obtained, and it can be judged whether the regional spacing is within the spacing threshold. If so, the movement speed of the patrol camera can be reduced; if not, the judgment continues. When the patrol camera passes the key position, the movement speed of the patrol camera is gradually increased, and after the regional spacing is out of the spacing threshold, the normal movement speed of the patrol camera is restored.

[0036] Before comparing the regional spacing with the spacing threshold, the troubleshooting level associated with the key location can be determined first. The more faults there are, the higher the troubleshooting level is. Then, according to the troubleshooting level, the spacing threshold corresponding to the key location is assigned. The higher the troubleshooting level is, the larger the corresponding spacing threshold is, that is, the patrol camera starts to slow down sooner.

[0037] The step of reducing the moving speed of the patrol camera is to reduce the moving speed according to the constructed speed change model. Speed ​​change model ; Wherein, v represents the moving speed of the patrol camera, d represents the area spacing, d0 represents the spacing threshold, k represents a normal number, representing the maximum possible value of the speed, and a represents a normal number, representing the steepness of the model curve.

[0038] When d0 is much smaller than d, approaches 0, so v approaches k; when d0 is much larger than d, approaches infinity, so v approaches 0.

[0039] The implementation principle of this embodiment is: According to the number and location of equipment in the data center, a fault patrol trajectory is constructed, and then the patrol camera is controlled to perform periodic patrols of the data center according to the fault patrol trajectory. During the patrol, the real-time position of the patrol camera is obtained, and then the regional spacing between the real-time position and the key position in the forward direction of the current trajectory of the patrol camera is obtained, and it is judged whether the regional spacing is within the spacing threshold; if so, the movement speed of the patrol camera is reduced; and after the patrol camera passes the key position and the regional spacing is out of the spacing threshold, the movement speed of the patrol camera is restored; after the patrol is completed, the patrol video stream collected by the patrol camera during the patrol process is obtained, the patrol video stream is converted into patrol pictures frame by frame, and redundant patrol pictures with high similarity are eliminated to generate a patrol picture set, and then the pictures in the patrol picture set are input into the pre-trained neural network model to obtain the judgment result, and according to the judgment result, it is determined whether the data center is abnormal; if so, the abnormal position is located, and an alarm message is sent, and then the pre-established fault influencing factor table is retrieved, and the equipment information associated with the abnormal position is found from the fault influencing factor table, and the cause of the fault is analyzed based on the equipment information.

[0040] Based on the above method embodiment, the second embodiment of the present application discloses a data center fault detection system. Figure 4 As an implementation of the troubleshooting system, the troubleshooting system may include: The trajectory construction module 101 is used to construct a fault inspection trajectory according to the data and location of the equipment in the data center; The control module 102 is used to control the patrol camera to perform periodic patrols on the equipment according to the fault patrol track; The video acquisition module 103 is used to acquire the patrol video stream collected by the patrol camera during the patrol process; A judgment module 104 is used to judge whether there is an abnormality in the data center based on the patrol video stream; The abnormality location alarm module 105 is used to locate the abnormality position and send alarm information when the data center is abnormal.

[0041] The troubleshooting system may further include: The video-to-picture module 106 is used to convert the patrol video stream frame by frame into patrol pictures; The picture elimination module 107 is used to eliminate redundant patrol pictures with high similarity to generate a patrol picture set; The neural network module 108 is used to input the pictures in the patrol picture set into the pre-trained neural network model to obtain a judgment result. The judgment module 104 determines whether the data center is abnormal based on the judgment result.

[0042] Furthermore, the troubleshooting system may also include: The position acquisition module 109 is used to obtain the key positions where faults frequently occur according to the fault influencing factor table, and to obtain the real-time position of the patrol camera; The spacing acquisition module 110 is used to obtain the regional spacing between the real-time position and the key position in the current trajectory direction of the patrol camera. The judgment module 104 is used to judge whether the regional spacing is within the spacing threshold; if so, the control module 102 controls the movement speed of the patrol camera to decrease; after the patrol camera passes the key position, the control module 102 controls the movement speed of the patrol camera to increase, and after the regional spacing deviates from the spacing threshold, the movement speed of the patrol camera is restored.

[0043] The third embodiment of the present application provides a terminal. As an implementation of the terminal, the terminal may include: a memory and a processor; wherein: The memory is used to store a troubleshooting program of the data center; The processor is used to execute the program stored in the memory to implement the steps of the above-mentioned data center troubleshooting method.

[0044] The memory may be connected to the processor via a communication bus, and the communication bus may be an address bus, a data bus, a control bus, etc.

[0045] In addition, the memory may include a random access memory (RAM) and may also include a non-volatile memory (NVM), such as at least one disk storage.

[0046] And the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0047] The fourth embodiment of the present application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the above-mentioned data center troubleshooting method.

[0048] Computer-readable storage media can be any available media that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives).

[0049] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application in sequence. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A data center troubleshooting method, characterized in that: include: Build a fault inspection track based on the number and location of equipment in the data center; Controlling the patrol camera to perform periodic patrol of the data center according to the fault patrol track; Obtaining the patrol video stream collected by the patrol camera during the patrol process; Determine whether there is an abnormality in the data center according to the patrol video stream; If so, locate the abnormal position and send an alarm message.

2. A data center troubleshooting method according to claim 1, characterized in that: After sending the alarm information, the method includes: Retrieving a pre-established fault influencing factor table, wherein the fault influencing factor table can be updated in real time; Searching for device information associated with the abnormal location from the fault influencing factor table, the device information including device environment data, operating time, and maintenance time; Analyze the cause of the fault based on the device information.

3. A data center troubleshooting method according to claim 2, characterized in that: The troubleshooting method further includes: According to the fault influencing factor table, key locations where faults frequently occur are obtained; Get the real-time location of patrol cameras; Obtaining the area distance between the real-time position and the key position in the advancing direction of the current track of the patrol camera; Determining whether the area spacing is within a spacing threshold; If yes, then reduce the movement speed of the patrol camera; After the patrol camera passes the key position and the area spacing is out of the spacing threshold, the moving speed of the patrol camera is restored.

4. A data center troubleshooting method according to claim 3, characterized in that: The step of reducing the moving speed of the patrol camera comprises: Modeling velocity changes ; Where v represents the moving speed of the patrol camera, d represents the area spacing, d0 represents the spacing threshold, k represents a normal number representing the maximum value of the speed, and a represents a normal number representing the steepness of the model curve; The moving speed is reduced according to the speed change model.

5. A data center troubleshooting method according to claim 3, characterized in that: The step of obtaining the key position of the fault-frequently occurring location comprises: Determine the number of faults associated with the faulty device through the fault influencing factor table; Devices whose fault times are greater than a preset threshold are screened, and the locations corresponding to the screened devices are the key locations.

6. A data center troubleshooting method according to claim 3, characterized in that: The steps after obtaining the key positions of the fault-prone areas include: Determining a troubleshooting level associated with the key location; According to the troubleshooting level, a spacing threshold corresponding to the key position is assigned; the higher the troubleshooting level, the larger the corresponding spacing threshold.

7. A data center troubleshooting method according to claim 1, characterized in that: The step of judging whether there is an abnormality in the data center according to the patrol video stream includes: Converting the patrol video stream frame by frame into patrol pictures; Eliminate redundant patrol pictures with high similarity to generate a patrol picture set; Inputting the patrol picture set into a pre-trained neural network model to obtain a judgment result; According to the judgment result, it is determined whether the data center is abnormal.

8. A data center troubleshooting system, characterized in that: include: A trajectory construction module (101), used to construct a fault inspection trajectory according to the data and location of the equipment in the data center; A control module (102) is used to control the patrol camera to perform periodic patrols on the equipment according to the fault patrol track; A video acquisition module (103), used to acquire the patrol video stream collected by the patrol camera during the patrol process; A judgment module (104), configured to judge whether there is an abnormality in the data center based on the patrol video stream; The abnormality location alarm module (105) is used to locate the abnormal position and send alarm information when the data center is abnormal.

9. A terminal, characterized in that: include: A memory storing a troubleshooting program of a data center; A processor is used to execute the program stored in the memory to implement the steps of the data center troubleshooting method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the data center troubleshooting method as claimed in any one of claims 1 to 7.

Images