Data center automation data scheduling method and system

Abstract

The application is suitable for the technical field of data scheduling, and provides a data center automatic data scheduling method and system. The application determines suspicious machine rooms and multiple normal machine rooms through fault trend analysis on multiple real-time running data; selects an auxiliary machine room; freezes data reception of the suspicious machine room, and migrates data processing queues to the auxiliary machine room; performs data priority analysis and available resource arrangement; and performs balanced scheduling and processing in the multiple normal machine rooms. The application can analyze and compare the fault trends of real-time running data of multiple online machine rooms, determine suspicious machine rooms and multiple normal machine rooms, and then select an auxiliary machine room, complete migration of data processing queues of the suspicious machine room before actual faults occur, and perform balanced scheduling and processing of data in the multiple normal machine rooms, so that the application avoids increasing of business response delay, prolonging of data migration time and degradation of service quality, and thus meets the requirement of high reliability and high real-time business for continuous operation.

Description

Technical Field

[0001] This invention belongs to the field of data scheduling technology, and in particular relates to a method and system for automated data scheduling in data centers. Background Technology

[0002] With the rapid development of cloud computing, big data, artificial intelligence and Internet services, data centers have gradually developed towards large-scale, clustered and multi-room collaboration. A data center usually includes multiple rooms, each of which is equipped with a large number of computing servers, storage devices, network switching devices and virtualization platforms, which together undertake various business tasks such as data storage, computing and processing, business scheduling, model training, and real-time analysis.

[0003] In existing technologies, data scheduling in data centers is mostly handled through real-time fault detection or manual operation and maintenance intervention. Typically, a failover mechanism is only activated and data scheduling is performed after obvious faults such as server offline, network interruption, or equipment failure are detected. Although this can ensure business recovery to a certain extent, since scheduling usually occurs after a fault occurs, a large amount of data awaiting processing in the faulty data center will still accumulate, which can easily lead to increased business response delays, extended data migration times, and decreased service quality, failing to meet the continuous operation requirements of high-reliability and high-real-time services. Summary of the Invention

[0004] The purpose of this invention is to provide an automated data scheduling method and system for data centers, aiming to solve the technical problems existing in the prior art mentioned in the background.

[0005] The embodiments of the present invention are implemented as follows: A data center automated data scheduling method, the method specifically includes the following steps: The system acquires real-time operational data from multiple online computer rooms in the data center, analyzes and compares the fault trends of these real-time operational data, and identifies suspicious computer rooms and multiple normal computer rooms. The communication quality of the suspected data center and multiple normal data centers is analyzed and compared, and an auxiliary data center is selected from the multiple normal data centers. Freeze the data reception of the suspected computer room, read the data processing queue of the suspected computer room, and migrate the data processing queue to the auxiliary computer room; Perform data priority analysis on the data processing queue, record priority information, and sort the available resources of multiple real-time running data, recording resource sorting information; According to the priority information and the resource arrangement information, the data processing queue is balanced and processed in multiple normal data centers.

[0006] As a further limitation of the technical solution of this invention, the step of obtaining real-time operating data of multiple online computer rooms in a data center, analyzing and comparing the fault trends of the multiple real-time operating data, and determining the suspicious computer rooms and multiple normal computer rooms specifically includes the following steps: Identify multiple online server rooms in the data center; Real-time operation monitoring is performed on multiple online data centers to obtain real-time operation data of the multiple online data centers; Fault trend analysis is performed on multiple sets of real-time operational data to calculate fault trend values ​​for multiple online computer rooms; The multiple fault trend values ​​are compared with a preset trend threshold to determine whether there is a potential fault trend; When there is a potential failure trend, identify the suspicious data center and multiple normal data centers.

[0007] As a further limitation of the technical solution of this invention, the analysis and comparison of communication quality between the suspected computer room and multiple normal computer rooms, and the selection of an auxiliary computer room from the multiple normal computer rooms, specifically includes the following steps: Monitor the communication quality between the suspicious data center and multiple normal data centers to obtain multiple communication quality data. Compare multiple communication quality data points and record the communication comparison results; Based on the communication comparison results, an auxiliary computer room is selected from among the multiple normal computer rooms.

[0008] As a further limitation of the technical solution of this embodiment of the invention, the steps of freezing the data reception of the suspicious computer room, reading the data processing queue of the suspicious computer room, and migrating the data processing queue to the auxiliary computer room specifically include the following steps: Data reception in the suspected computer room is frozen. Read the data processing queue of the suspected computer room; Based on the data processing queue, construct a data migration list; According to the data migration list, the data processing queue is migrated to the auxiliary computer room.

[0009] As a further limitation of the technical solution of this embodiment of the invention, the step of performing data priority analysis on the data processing queue, recording priority information, and arranging the available resources of multiple real-time running data, and recording resource arrangement information, specifically includes the following steps: According to multiple preset priority factors, the priority of multiple data to be processed in the data processing queue is analyzed and compared to determine the priority of multiple data to be processed. Record the multiple priorities to obtain priority information; From the multiple sets of real-time operational data, extract the available resource data of multiple normal computer rooms; The available resource data are sorted and compared, and the resource sorting information of the normal computer rooms is recorded.

[0010] As a further limitation of the technical solution of this embodiment of the invention, the step of balancing and processing the data processing queue in multiple normal data centers according to the priority information and the resource arrangement information specifically includes the following steps: The priority information and resource arrangement information are analyzed for balanced scheduling, and multiple scheduling directions for the data to be processed are selected from multiple normal data centers. The multiple data to be processed are balanced and scheduled according to the multiple scheduling directions; In multiple normal computer rooms, multiple pending data sets are processed collaboratively.

[0011] A data center automated data scheduling system, comprising a fault trend analysis module, an auxiliary data center selection module, a data migration processing module, a priority ranking analysis module, and a balanced scheduling processing module, wherein: The fault trend analysis module is used to acquire real-time operating data of multiple online computer rooms in the data center, analyze and compare the fault trends of the multiple real-time operating data, and identify suspicious computer rooms and multiple normal computer rooms. The auxiliary computer room selection module is used to analyze and compare the communication quality of a suspicious computer room with multiple normal computer rooms, and select an auxiliary computer room from the multiple normal computer rooms. The data migration processing module is used to freeze the data reception of the suspicious data center, read the data processing queue of the suspicious data center, and migrate the data processing queue to the auxiliary data center; The priority ranking analysis module is used to perform data priority analysis on the data processing queue, record priority information, and rank multiple real-time running data according to available resources, and record resource ranking information. The balanced scheduling processing module is used to perform balanced scheduling and processing of the data processing queue in multiple normal data centers according to the priority information and the resource arrangement information.

[0012] As a further limitation of the technical solution of this embodiment of the invention, the fault trend analysis module specifically includes: The online data center identification unit is used to identify multiple online data centers within a data center. The operation monitoring unit is used to perform real-time operation monitoring on multiple online computer rooms and obtain real-time operation data of multiple online computer rooms; The fault trend analysis unit is used to perform fault trend analysis on multiple real-time operating data and calculate the fault trend values ​​of multiple online computer rooms. The fault trend determination unit is used to compare multiple fault trend values ​​with a preset trend threshold to determine whether there is a potential fault trend. The data center classification unit is used to identify suspicious data centers and multiple normal data centers when there is a potential failure trend.

[0013] As a further limitation of the technical solution of this embodiment of the invention, the data migration processing module specifically includes: A freeze control unit is used to freeze the data reception of the suspected computer room. A data processing queue reading unit is used to read the data processing queue of the suspected computer room; A migration list construction unit is used to construct a data migration list based on the data processing queue; The queue migration unit is used to migrate the data processing queue to the auxiliary computer room according to the data migration list.

[0014] As a further limitation of the technical solution of this embodiment of the invention, the priority ranking analysis module specifically includes: The priority comparison unit is used to analyze and compare the priorities of multiple data to be processed in the data processing queue according to multiple preset priority factors, and to determine the priority of the multiple data to be processed. A priority recording unit is used to record multiple priorities to obtain priority information; Available resource data extraction unit, used to extract available resource data of multiple normal computer rooms from multiple real-time operation data; The resource ranking and comparison unit is used to rank and compare multiple available resource data and record the resource ranking information of multiple normal computer rooms.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention analyzes the fault trends of multiple real-time operational data to identify suspicious data centers and multiple normal data centers; selects an auxiliary data center; freezes data reception in the suspicious data center and migrates the data processing queue to the auxiliary data center; performs data priority analysis and ranks available resources; and performs balanced scheduling and processing across multiple normal data centers. This allows for the analysis and comparison of fault trends in the real-time operational data of multiple online data centers, identifying suspicious data centers and multiple normal data centers, and then selecting an auxiliary data center. Before an actual fault occurs, the migration of the data processing queue from the suspicious data center is completed, and balanced scheduling and processing of data across multiple normal data centers is performed. This avoids increased service response delays, prolonged data migration times, and decreased service quality, thereby meeting the continuous operation requirements of high-reliability, high-real-time services. Attached Figure Description

[0016] Figure 1 A flowchart of the data center automated data scheduling method provided in an embodiment of the present invention is shown; Figure 2 The following is an application architecture diagram of the data center automated data scheduling system provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] Understandably, current data scheduling in data centers is mostly handled through real-time fault detection or manual maintenance intervention. Typically, a failover mechanism is only activated and data scheduling is performed after obvious faults such as server offline, network interruption, or equipment failure are detected. While this can ensure business recovery to some extent, since scheduling usually occurs after a fault occurs, a large amount of data awaiting processing in the faulty data center will still accumulate. This can easily lead to increased business response delays, longer data migration times, and decreased service quality, failing to meet the continuous operation requirements of highly reliable and real-time services.

[0019] To address the aforementioned issues, the data center automated data scheduling method and system disclosed in this invention acquires real-time operational data from multiple online server rooms in the data center, analyzes and compares the fault trends of these data to identify suspicious server rooms and multiple normal server rooms; analyzes and compares the communication quality of the suspicious server rooms and the multiple normal server rooms, and selects an auxiliary server room from among the normal server rooms; freezes the data reception of the suspicious server rooms, reads the data processing queue of the suspicious server rooms, and migrates the data processing queue to the auxiliary server room; performs data priority analysis on the data processing queue, records priority information, and ranks the available resources of the multiple real-time operational data, recording resource ranking information; and performs balanced scheduling and processing of the data processing queue across the multiple normal server rooms according to the priority information and resource ranking information. It can analyze and compare the fault trends of real-time operating data from multiple online data centers, identify suspicious data centers and multiple normal data centers, and then select auxiliary data centers. Before an actual fault occurs, it can complete the migration of the data processing queue of the suspicious data center and perform balanced scheduling and processing of data in multiple normal data centers, so as to avoid increased business response delays, extended data migration time and degraded service quality, thereby meeting the requirements of high reliability and high real-time business for continuous operation.

[0020] Specifically, Figure 1 A flowchart of an automated data scheduling method for data centers provided by an embodiment of the present invention is shown.

[0021] In a preferred embodiment of the present invention, a data center automated data scheduling method specifically includes the following steps: Step S101: Obtain real-time operating data from multiple online computer rooms in the data center, analyze and compare the fault trends of the multiple real-time operating data, and identify suspicious computer rooms and multiple normal computer rooms.

[0022] In this embodiment of the invention, multiple online server rooms of a data center are identified, and real-time operation monitoring is performed on these multiple online server rooms to obtain real-time operational data such as CPU utilization, GPU utilization, memory usage, temperature, and queue length. Then, fault trend analysis is performed on the multiple real-time operational data to calculate the fault trend values ​​of the multiple online server rooms. The multiple fault trend values ​​are then compared with a preset trend threshold to determine whether there is a potential fault trend. Specifically, when a fault trend value is greater than the trend threshold, a potential fault trend is determined to exist. At this time, online server rooms with fault trend values ​​greater than the trend threshold are identified as suspicious server rooms, and online server rooms with fault trend values ​​not greater than the trend threshold are identified as normal server rooms. Thus, multiple online server rooms can be divided into suspicious server rooms and multiple normal server rooms.

[0023] Understandably, in fault trend analysis, multiple analyses are performed on real-time operating data to calculate various trend values, such as CPU utilization increase trend, GPU utilization increase trend, memory usage increase trend, temperature increase trend, and queue accumulation trend. These values ​​are then compared with corresponding trend sub-thresholds. When a trend value exceeds the corresponding trend sub-threshold, 1 point is awarded, and the total score is calculated to obtain the corresponding fault trend value for the online data center.

[0024] Specifically, in another preferred embodiment provided by the present invention, the step of acquiring real-time operating data from multiple online computer rooms in a data center, analyzing and comparing the fault trends of the multiple real-time operating data, and determining suspicious computer rooms and multiple normal computer rooms specifically includes the following steps: Identify multiple online server rooms in the data center; Real-time operation monitoring is performed on multiple online data centers to obtain real-time operation data of the multiple online data centers; Fault trend analysis is performed on multiple real-time operating data to calculate fault trend values ​​for multiple online data centers; The multiple fault trend values ​​are compared with a preset trend threshold to determine whether there is a potential fault trend; When there is a potential failure trend, identify the suspicious data center and multiple normal data centers.

[0025] Furthermore, the data center automated data scheduling method also includes the following steps: Step S102: Analyze and compare the communication quality of the suspicious computer room with multiple normal computer rooms, and select an auxiliary computer room from the multiple normal computer rooms.

[0026] In this embodiment of the invention, the communication between the suspicious computer room and multiple normal computer rooms is monitored for quality such as network latency, network bandwidth, packet loss rate and jitter. Multiple communication quality data are obtained, and then the multiple communication quality data are compared and the communication comparison results are recorded. Then, according to the communication comparison results, the auxiliary computer room with the highest communication quality with the suspicious computer room is selected from multiple normal computer rooms.

[0027] Specifically, in another preferred embodiment provided by the present invention, the step of analyzing and comparing the communication quality of the suspected data center and multiple normal data centers, and selecting an auxiliary data center from the multiple normal data centers, specifically includes the following steps: Monitor the communication quality between the suspicious data center and multiple normal data centers to obtain multiple communication quality data. Compare multiple communication quality data points and record the communication comparison results; Based on the communication comparison results, an auxiliary computer room is selected from among the multiple normal computer rooms.

[0028] Furthermore, the data center automated data scheduling method also includes the following steps: Step S103: Freeze the data reception of the suspected computer room, read the data processing queue of the suspected computer room, and migrate the data processing queue to the auxiliary computer room.

[0029] In this embodiment of the invention, data reception is frozen for suspicious data centers, so that new business data is no longer sent to the suspicious data centers. The data processing queues of the suspicious data centers, such as data waiting to be processed, data being processed, and data that has been completed but not returned, are read. Based on the data processing queues, a data migration list is constructed, and then all data processing queues are migrated to the auxiliary data center according to the data migration list.

[0030] Understandably, the data migration checklist contains information such as the data number, size, and processing status of data awaiting processing, data being processed, and data that has been completed but not yet returned.

[0031] Specifically, in another preferred embodiment provided by the present invention, freezing the data reception of the suspected data center, reading the data processing queue of the suspected data center, and migrating the data processing queue to the auxiliary data center specifically includes the following steps: Data reception in the suspected computer room is frozen. Read the data processing queue of the suspected computer room; Based on the data processing queue, construct a data migration list; According to the data migration list, the data processing queue is migrated to the auxiliary computer room.

[0032] Furthermore, the data center automated data scheduling method also includes the following steps: Step S104: Perform data priority analysis on the data processing queue, record priority information, and arrange the available resources of multiple real-time running data, recording resource arrangement information.

[0033] In this embodiment of the invention, multiple data to be processed in the data processing queue are analyzed and compared according to multiple preset priority factors to determine the priority of multiple data to be processed and record the priority of multiple data to be processed to obtain priority information. Furthermore, available resource data of multiple normal data centers are extracted from multiple real-time running data, and the multiple available resource data are arranged and compared to record the resource arrangement information of multiple normal data centers.

[0034] It is understood that in the embodiments of the present invention, multiple priority factors, including business type, real-time requirements, user level, and data importance, are considered.

[0035] It is understood that, in the embodiments of the present invention, available resource data includes CPU idle rate, GPU idle rate, remaining memory, current queue count, etc.

[0036] Specifically, in another preferred embodiment provided by the present invention, the step of performing data priority analysis on the data processing queue, recording priority information, and arranging the available resources of multiple real-time running data and recording resource arrangement information specifically includes the following steps: According to multiple preset priority factors, the priority of multiple data to be processed in the data processing queue is analyzed and compared to determine the priority of multiple data to be processed. Record the multiple priorities to obtain priority information; From the multiple sets of real-time operational data, extract the available resource data of multiple normal computer rooms; The available resource data are sorted and compared, and the resource sorting information of the normal computer rooms is recorded.

[0037] Furthermore, the data center automated data scheduling method also includes the following steps: Step S105: According to the priority information and the resource arrangement information, the data processing queue is balanced and processed in multiple normal computer rooms.

[0038] In this embodiment of the invention, by performing balanced scheduling analysis on priority information and resource arrangement information, the scheduling direction of multiple pending data in the data processing queue is planned from multiple normal data centers. Then, according to the multiple scheduling directions, the multiple pending data in the data processing queue are respectively scheduled and transmitted to the corresponding normal data centers. In the multiple normal data centers, the corresponding pending data is processed collaboratively to achieve balanced scheduling and processing of data. This ensures that pending data with higher priority can be scheduled and transmitted to normal data centers with higher available resources for faster processing.

[0039] Specifically, in another preferred embodiment provided by the present invention, the step of balancing and processing the data processing queue across multiple normal data centers according to the priority information and the resource arrangement information specifically includes the following steps: The priority information and resource arrangement information are analyzed for balanced scheduling, and multiple scheduling directions for the data to be processed are selected from multiple normal data centers. The multiple data to be processed are balanced and scheduled according to the multiple scheduling directions; In multiple normal computer rooms, multiple pending data sets are processed collaboratively.

[0040] Furthermore, Figure 2The following is an application architecture diagram of the data center automated data scheduling system provided in an embodiment of the present invention.

[0041] Specifically, in another preferred embodiment provided by the present invention, the data center automated data scheduling system includes: The fault trend analysis module 101 is used to acquire real-time operating data of multiple online computer rooms in the data center, analyze and compare the fault trends of the multiple real-time operating data, and identify suspicious computer rooms and multiple normal computer rooms.

[0042] In this embodiment of the invention, the fault trend analysis module 101 identifies multiple online computer rooms in the data center and performs real-time operation monitoring on these multiple online computer rooms. It acquires real-time operational data such as CPU utilization, GPU utilization, memory usage, temperature, and queue length for each online computer room. Then, it performs fault trend analysis on this real-time operational data to calculate fault trend values ​​for each online computer room. These fault trend values ​​are then compared with a preset trend threshold to determine if a potential fault trend exists. Specifically, if a fault trend value is greater than the trend threshold, a potential fault trend is identified. In this case, online computer rooms with fault trend values ​​greater than the trend threshold are identified as suspicious computer rooms, while those with fault trend values ​​not greater than the trend threshold are identified as normal computer rooms. This allows multiple online computer rooms to be divided into suspicious computer rooms and multiple normal computer rooms.

[0043] Specifically, in another preferred embodiment provided by the present invention, the fault trend analysis module 101 specifically includes: The online data center identification unit is used to identify multiple online data centers within a data center. The operation monitoring unit is used to perform real-time operation monitoring on multiple online computer rooms and obtain real-time operation data of multiple online computer rooms; The fault trend analysis unit is used to perform fault trend analysis on multiple real-time operating data and calculate the fault trend values ​​of multiple online computer rooms. The fault trend determination unit is used to compare multiple fault trend values ​​with a preset trend threshold to determine whether there is a potential fault trend. The data center classification unit is used to identify suspicious data centers and multiple normal data centers when there is a potential failure trend.

[0044] Furthermore, the data center automated data scheduling system also includes: The auxiliary computer room selection module 102 is used to analyze and compare the communication quality of the suspicious computer room and multiple normal computer rooms, and select an auxiliary computer room from the multiple normal computer rooms.

[0045] In this embodiment of the invention, the auxiliary computer room selection module 102 performs quality monitoring on the communication between the suspicious computer room and multiple normal computer rooms, including network latency, network bandwidth, packet loss rate and jitter, obtains multiple communication quality data, compares the multiple communication quality data, records the communication comparison results, and then selects the auxiliary computer room with the highest communication quality with the suspicious computer room from multiple normal computer rooms according to the communication comparison results.

[0046] The data migration processing module 103 is used to freeze the data reception of the suspicious computer room, read the data processing queue of the suspicious computer room, and migrate the data processing queue to the auxiliary computer room.

[0047] In this embodiment of the invention, the data migration processing module 103 performs data reception freeze control on the suspicious data center, so that new business data is no longer sent to the suspicious data center, and reads the data processing queues of the suspicious data center, such as data waiting to be processed, data being processed, and data that has been completed but not returned. Based on the data processing queues, a data migration list is constructed, and then the entire data processing queue is migrated to the auxiliary data center according to the data migration list.

[0048] Specifically, in another preferred embodiment provided by the present invention, the data migration processing module 103 specifically includes: A freeze control unit is used to freeze the data reception of the suspected computer room. A data processing queue reading unit is used to read the data processing queue of the suspected computer room; A migration list construction unit is used to construct a data migration list based on the data processing queue; The queue migration unit is used to migrate the data processing queue to the auxiliary computer room according to the data migration list.

[0049] Furthermore, the data center automated data scheduling system also includes: The priority ranking analysis module 104 is used to perform data priority analysis on the data processing queue, record priority information, and rank multiple real-time running data according to available resources, and record resource ranking information.

[0050] In this embodiment of the invention, the priority ranking analysis module 104 analyzes and compares the priorities of multiple data to be processed in the data processing queue according to multiple preset priority factors, determines the priority of multiple data to be processed, records the priority of multiple data to be processed, obtains priority information, and extracts available resource data of multiple normal data centers from multiple real-time running data, ranks and compares multiple available resource data, and records the resource ranking information of multiple normal data centers.

[0051] Specifically, in another preferred embodiment provided by the present invention, the priority ranking analysis module 104 specifically includes: The priority comparison unit is used to analyze and compare the priorities of multiple data to be processed in the data processing queue according to multiple preset priority factors, and to determine the priority of the multiple data to be processed. A priority recording unit is used to record multiple priorities to obtain priority information; Available resource data extraction unit, used to extract available resource data of multiple normal computer rooms from multiple real-time operation data; The resource ranking and comparison unit is used to rank and compare multiple available resource data and record the resource ranking information of multiple normal computer rooms.

[0052] Furthermore, the data center automated data scheduling system also includes: The balanced scheduling processing module 105 is used to perform balanced scheduling and processing of the data processing queue in multiple normal computer rooms according to the priority information and the resource arrangement information.

[0053] In this embodiment of the invention, the balanced scheduling processing module 105 performs balanced scheduling analysis on priority information and resource arrangement information, plans the scheduling direction of multiple pending data in the data processing queue from multiple normal data centers, and then schedules and transmits the multiple pending data in the data processing queue to the corresponding normal data centers according to the multiple scheduling directions. In the multiple normal data centers, the corresponding pending data is processed collaboratively to achieve balanced scheduling and processing of data, ensuring that pending data with higher priority can be scheduled and transmitted to normal data centers with higher available resources for faster processing.

[0054] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A data center automated data scheduling method, characterized in that, The method specifically includes the following steps: The system acquires real-time operational data from multiple online computer rooms in the data center, analyzes and compares the fault trends of these real-time operational data, and identifies suspicious computer rooms and multiple normal computer rooms. The communication quality of the suspected data center and multiple normal data centers is analyzed and compared, and an auxiliary data center is selected from the multiple normal data centers. Freeze the data reception of the suspected computer room, read the data processing queue of the suspected computer room, and migrate the data processing queue to the auxiliary computer room; Perform data priority analysis on the data processing queue, record priority information, and sort the available resources of multiple real-time running data, recording resource sorting information; According to the priority information and the resource arrangement information, the data processing queue is balanced and processed in multiple normal data centers.

2. The data center automated data scheduling method according to claim 1, characterized in that, The process of acquiring real-time operational data from multiple online server rooms in a data center, analyzing and comparing the fault trends of these data points, and identifying suspicious server rooms and multiple normal server rooms specifically includes the following steps: Identify multiple online server rooms in the data center; Real-time operation monitoring is performed on multiple online data centers to obtain real-time operation data of the multiple online data centers; Fault trend analysis is performed on multiple sets of real-time operational data to calculate fault trend values ​​for multiple online computer rooms; The multiple fault trend values ​​are compared with a preset trend threshold to determine whether there is a potential fault trend; When there is a potential failure trend, identify the suspicious data center and multiple normal data centers.

3. The data center automated data scheduling method according to claim 1, characterized in that, The analysis and comparison of communication quality between the suspected data center and multiple normal data centers, and the selection of an auxiliary data center from the multiple normal data centers, specifically includes the following steps: Monitor the communication quality between the suspicious data center and multiple normal data centers to obtain multiple communication quality data. Compare multiple communication quality data points and record the communication comparison results; Based on the communication comparison results, an auxiliary computer room is selected from among the multiple normal computer rooms.

4. The data center automated data scheduling method according to claim 1, characterized in that, The steps of freezing the data reception of the suspected data center, reading the data processing queue of the suspected data center, and migrating the data processing queue to the auxiliary data center specifically include the following steps: Data reception in the suspected computer room is frozen. Read the data processing queue of the suspected computer room; Based on the data processing queue, construct a data migration list; According to the data migration list, the data processing queue is migrated to the auxiliary computer room.

5. The data center automated data scheduling method according to claim 1, characterized in that, The steps of performing data priority analysis on the data processing queue, recording priority information, and arranging multiple real-time running data sets according to available resources and recording resource arrangement information specifically include the following steps: According to multiple preset priority factors, the priority of multiple data to be processed in the data processing queue is analyzed and compared to determine the priority of multiple data to be processed. Record the multiple priorities to obtain priority information; From the multiple sets of real-time operational data, extract the available resource data of multiple normal computer rooms; The available resource data are sorted and compared, and the resource sorting information of the normal computer rooms is recorded.

6. The data center automated data scheduling method according to claim 5, characterized in that, The step of balancing and processing the data processing queue across multiple normal data centers according to the priority information and the resource arrangement information specifically includes the following steps: The priority information and resource arrangement information are analyzed for balanced scheduling, and multiple scheduling directions for the data to be processed are selected from multiple normal data centers. The multiple data to be processed are balanced and scheduled according to the multiple scheduling directions; In multiple normal computer rooms, multiple pending data sets are processed collaboratively.

7. A data center automated data scheduling system, characterized in that, The system includes a fault trend analysis module, an auxiliary data center selection module, a data migration processing module, a priority ranking analysis module, and a balanced scheduling processing module, wherein: The fault trend analysis module is used to acquire real-time operating data of multiple online computer rooms in the data center, analyze and compare the fault trends of the multiple real-time operating data, and identify suspicious computer rooms and multiple normal computer rooms. The auxiliary computer room selection module is used to analyze and compare the communication quality of a suspicious computer room with multiple normal computer rooms, and select an auxiliary computer room from the multiple normal computer rooms. The data migration processing module is used to freeze the data reception of the suspicious data center, read the data processing queue of the suspicious data center, and migrate the data processing queue to the auxiliary data center; The priority ranking analysis module is used to perform data priority analysis on the data processing queue, record priority information, and rank multiple real-time running data according to available resources, and record resource ranking information. The balanced scheduling processing module is used to perform balanced scheduling and processing of the data processing queue in multiple normal data centers according to the priority information and the resource arrangement information.

8. The data center automated data scheduling system according to claim 7, characterized in that, The fault trend analysis module specifically includes: The online data center identification unit is used to identify multiple online data centers within a data center. The operation monitoring unit is used to perform real-time operation monitoring on multiple online computer rooms and obtain real-time operation data of multiple online computer rooms; The fault trend analysis unit is used to perform fault trend analysis on multiple real-time operating data and calculate the fault trend values ​​of multiple online computer rooms. The fault trend determination unit is used to compare multiple fault trend values ​​with a preset trend threshold to determine whether there is a potential fault trend. The data center classification unit is used to identify suspicious data centers and multiple normal data centers when there is a potential failure trend.

9. The data center automated data scheduling system according to claim 7, characterized in that, The data migration processing module specifically includes: A freeze control unit is used to freeze the data reception of the suspected computer room. A data processing queue reading unit is used to read the data processing queue of the suspected computer room; A migration list construction unit is used to construct a data migration list based on the data processing queue; The queue migration unit is used to migrate the data processing queue to the auxiliary computer room according to the data migration list.

10. The data center automated data scheduling system according to claim 7, characterized in that, The priority ranking analysis module specifically includes: The priority comparison unit is used to analyze and compare the priorities of multiple data to be processed in the data processing queue according to multiple preset priority factors, and to determine the priority of the multiple data to be processed. A priority recording unit is used to record multiple priorities to obtain priority information; Available resource data extraction unit, used to extract available resource data of multiple normal computer rooms from multiple real-time operation data; The resource ranking and comparison unit is used to rank and compare multiple available resource data and record the resource ranking information of multiple normal computer rooms.

Images