A daemon process management method and system based on cloud computing

Abstract

The present application is applicable to the field of process management technology, and provides a daemon process management method and system based on cloud computing, the method comprising: obtaining multi-level configuration information of the daemon process, and implementing multi-level management of the daemon process based on the multi-level configuration information; monitoring the operation of the daemon process in real time, generating a worker resource mapping data table, and obtaining the operation status of the daemon process based on the worker resource mapping data table; if the operation status of the daemon process is in an abnormal state, configuring a new worker module for the daemon process and monitoring the configuration process of the new worker module in real time; if the configuration of the new worker module fails, triggering a capacity expansion management mechanism, monitoring the execution result of the capacity expansion operation in real time, generating a capacity expansion monitoring report, and triggering a management reward mechanism based on the capacity expansion monitoring report. The present application significantly improves the reliability, efficiency and security of the daemon process, and also enhances the maintainability and scalability of the cloud computing platform.

Description

Technical Field

[0001] The present application belongs to the field of process management technology, and in particular, relates to a daemon process management method and system based on cloud computing. Background Art

[0002] With the rapid development of cloud computing technology, the demand for large-scale clusters is gradually increasing. The performance of cluster management and scheduling determines the resource utilization of the Infrastructure as a Service (IaaS) layer of large-scale clusters.

[0003] In the prior art, cloud computing business scenarios usually require the use of daemons to perform some data-related processing, such as ETL, summary calculations, etc., and the daemon process will always run in the system background even if there is no task, thereby occupying system resources. However, since the daemon process runs in the system background for a long time, memory leaks may occur. Memory leaks will not only cause the daemon process performance to degrade, but may also cause process crashes or system paralysis, seriously affecting service continuity.

[0004] In summary, the problem of how to monitor and automatically manage the use of daemon resources in real time to ensure the security and efficiency of cloud computing platform business operations needs to be solved urgently. Summary of the invention

[0005] The embodiments of the present application provide a daemon process management method and system based on cloud computing, which can solve at least one of the above-mentioned problems in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a daemon process management method based on cloud computing, including:

[0007] Obtain multi-level configuration information of the daemon process, wherein the multi-level configuration information includes cluster-level configuration information, application-level configuration information, worker-level configuration information, and process-level configuration information;

[0008] Based on the multi-level configuration information, implement multi-level management of the daemon process;

[0009] Monitor the operation of the daemon process in real time, generate a worker resource mapping data table, obtain the operation status of the daemon process based on the worker resource mapping data table, and if the operation status of the daemon process is in an abnormal state, configure a new worker module for the daemon process, wherein one worker module is configured with at least one daemon process, the worker resource mapping data table is a mapping table of resource usage data of the worker module and the daemon process, and the abnormal state is that the operation of the daemon process causes the worker module to be in a high-load state;

[0010] Monitor the configuration process of the new worker module in real time. If the configuration of the new worker module fails, trigger the capacity expansion management mechanism, wherein the capacity expansion management mechanism notifies the administrator to modify the multi-level configuration information and perform the capacity expansion operation;

[0011] Monitor the execution results of the capacity expansion operation in real time, generate a capacity expansion monitoring report, and trigger a management reward mechanism based on the capacity expansion monitoring report, wherein the capacity expansion monitoring report includes administrator information, configuration information modification records, capacity expansion monitoring data, and performance improvement data.

[0012] Furthermore, the multi-level configuration of the daemon process is obtained, and further includes:

[0013] Based on the multi-level configuration information, selecting a configuration mode for the configuration of the worker module, wherein the configuration mode includes a unified configuration and a machine configuration;

[0014] The unified configuration is used to configure each worker module with the same number of daemon processes;

[0015] The machine configuration is used to configure each worker module according to a preset number of daemon processes.

[0016] Further, the multi-level management of the daemon process is implemented based on the multi-level configuration information, including:

[0017] Performing multi-level configuration on the daemon process according to the multi-level configuration information;

[0018] Each level of process management unit corresponds to each layer of operation layer, so that each layer of operation layer performs corresponding operation, and the operation operation includes starting, stopping and restarting;

[0019] Feeding back the execution result of the running operation to the data management unit, and implementing multi-level management of the daemon process based on the execution result;

[0020] Among them, the process management unit includes a cluster management module, an application management module, a worker management module and a process management module. The cluster management module is used to manage the operation of at least one cluster module in the cloud computing platform, the application management module is used to manage the operation of at least one application module in the cluster module, the worker management module is used to manage the operation of at least one worker module in the application module, and the process management module is used to manage the operation of at least one daemon process in the worker module. The operation layer includes a cluster layer, an application layer, a worker layer and a process layer.

[0021] Furthermore, the multi-level configuration of the daemon process also includes:

[0022] Allocate independent CPU, memory resources and disk space to each daemon process;

[0023] Use virtual file system technology to create an independent file system for each daemon process;

[0024] Using network virtualization technology, each daemon process is allocated an independent network space;

[0025] Obtaining security permission data of the daemon process, and determining access rights of the daemon process based on the security permission data;

[0026] Use RabbitMQ to communicate between different daemons;

[0027] The worker module manages communication between different daemons and process-level configuration information.

[0028] Further, the operation status of the daemon process is obtained based on the worker resource mapping data table, and if the operation status of the daemon process is in an abnormal state, a new worker module is configured for the daemon process, including:

[0029] By comparing the resource usage data of each worker module with a preset resource usage threshold in turn, a worker module in a high-load state is obtained, wherein the resource usage data includes CPU usage, memory occupancy, and network I / O usage;

[0030] The expert scoring method is used to score the CPU usage, memory usage, and network I / O usage of each daemon process to obtain a comprehensive score for each daemon process;

[0031] Based on the comprehensive score, obtaining the daemon process that needs to be configured with a new worker module;

[0032] A new worker module is configured for the daemon process, and resource usage data of the new worker module and the daemon process in the new worker module are added to the worker resource mapping data table.

[0033] Furthermore, the real-time monitoring of the configuration process of the new worker module and triggering the expansion management mechanism if the configuration of the new worker module fails include:

[0034] Obtain a cluster resource mapping data table, and determine the load status of each cluster node in the cluster based on the cluster resource mapping data table, wherein the cluster resource mapping data table is a basic usage data mapping table of each cluster node in the cluster of the cloud computing platform, and the basic usage data includes the IP address, host name, number of CPU cores, memory capacity, disk space, network bandwidth, load status and network status of each cluster node, and the load status includes a high load status, a normal status and an offline status;

[0035] If there is no cluster node in the cluster whose load status is normal, the capacity expansion management mechanism is triggered;

[0036] If there is a cluster node with a normal load status in the cluster, the cluster node with a normal load status is selected to configure a new worker module;

[0037] After configuring a new worker module, if there is a daemon process that is in an abnormal state, the expansion mechanism is triggered.

[0038] Furthermore, the notifying administrator to modify the multi-level configuration information and perform the capacity expansion operation includes:

[0039] Marking the daemon process in an abnormal state and the worker module corresponding to the daemon process;

[0040] Sending the cluster resource mapping data table and the marking result to an administrator, and notifying the administrator to modify the multi-level configuration information;

[0041] Based on the modified multi-level configuration information and marking results, increase the number of cluster nodes;

[0042] The expansion operation is performed in batches in multiple time intervals, the number of cluster nodes that can be added concurrently in each time interval is obtained, and at least one new cluster node is set for each time interval;

[0043] By analyzing the related tasks being executed in the cluster and the load status of each cluster node, tasks are reallocated to the cluster nodes in the cluster.

[0044] Furthermore, the real-time monitoring of the execution result of the capacity expansion operation generates a capacity expansion monitoring report, and based on the capacity expansion monitoring report, triggers a management reward mechanism, including:

[0045] Based on the cluster after the capacity expansion operation, obtain a worker resource mapping data table and a cluster resource mapping data table, and monitor various data in the worker resource mapping data table and the cluster resource mapping data table;

[0046] If the performance corresponding to each data is improved to the preset baseline, the administrator information, configuration modification information and timestamp information of the administrator who performs the modification task are obtained through the operation log;

[0047] Determine the administrator's reward level based on the significant degree of performance improvement and the preset reward plan;

[0048] The reward level is obtained and a reward notification is generated, wherein the reward notification includes the reward content, the name of the administrator who receives the reward and the specific performance optimization details.

[0049] In a second aspect, an embodiment of the present application provides a daemon process management system based on cloud computing, including:

[0050] Process management background: used to obtain multi-level configuration information of the daemon process, which includes cluster-level configuration information, application-level configuration information, worker-level configuration information, and process-level configuration information;

[0051] A service management unit: used for implementing multi-level management of the daemon process based on the multi-level configuration information;

[0052] A new worker module configuration unit: used for monitoring the operation of the daemon process in real time, generating a worker resource mapping data table, obtaining the operation status of the daemon process based on the worker resource mapping data table, and configuring a new worker module for the daemon process if the operation status of the daemon process is in an abnormal state, wherein one worker module is configured with at least one daemon process, the worker resource mapping data table is a mapping table of resource usage data of the worker module and the daemon process, and the abnormal state is that the operation of the daemon process causes the worker module to be in a high-load state;

[0053] Expansion unit: used to monitor the configuration process of the new worker module in real time. If the configuration of the new worker module fails, the expansion management mechanism is triggered, wherein the expansion management mechanism notifies the administrator to modify the multi-level configuration information and perform the expansion operation;

[0054] Management reward unit: used to monitor the execution results of the capacity expansion operation in real time, generate a capacity expansion monitoring report, and trigger the management reward mechanism based on the capacity expansion monitoring report, wherein the capacity expansion monitoring report includes administrator information, configuration information modification records, capacity expansion monitoring data and performance improvement data.

[0055] In a third aspect, an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned cloud computing-based daemon process management method when executing the computer program.

[0056] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0057] The present invention discloses a daemon process management method based on cloud computing. The method adopts a multi-level management mode for the daemon process, realizes efficient and safe daemon process deployment and management, and detects the running status of each daemon process in real time, promptly discovers the running abnormality of the daemon process, and then triggers the configuration or expansion operation of the worker module, thereby realizing the redistribution of the daemon process to ensure the stable operation of the cloud computing platform. At the same time, a management reward mechanism is designed to automatically record the administrator information and propose corresponding reward measures to motivate the maintenance personnel to continuously optimize the system performance, which significantly improves the reliability, efficiency and security of the daemon process, and also enhances the maintainability and scalability of the cloud computing platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0059] Figure 1 It is a flowchart of a daemon process management method based on cloud computing provided by an embodiment of the present invention;

[0060] Figure 2 It is a schematic diagram of the structure of a daemon process management system based on cloud computing provided by an embodiment of the present invention;

[0061] Figure 3 It is a structural diagram of the terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0062] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0063] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0064] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0065] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0066] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0067] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0068] See also Figure 1 As shown, the present invention is a daemon process management method based on cloud computing, comprising the following steps:

[0069] S100, obtaining multi-level configuration information of a daemon process, wherein the multi-level configuration information includes cluster-level configuration information, application-level configuration information, worker-level configuration information, and process-level configuration information;

[0070] Cloud computing is a technology that provides scalable computing services in the cloud, avoiding the need to estimate resources and make initial investments when using traditional servers. In the process of cloud computing, virtual machine instances need to be created. Different instances usually correspond to different tasks, which can meet application scenarios of different scales and requirements. At the same time, the execution of corresponding tasks usually requires the use of daemons to perform data-related processing such as ETL and summary calculations. For example, CVM Tencent Cloud is used to help users quickly create, deploy and manage virtual machine instances.

[0071] In the present invention, the virtual machine instances created and deployed by the cloud computing platform are encapsulated as cluster modules, and the corresponding tasks in the cluster modules are encapsulated as application modules. There are multiple daemon processes in the application modules. In order to facilitate further management of the daemon processes, one or more of the multiple daemon processes are encapsulated as worker modules according to multi-level configuration information. That is, the number of daemon processes is not in a one-to-one correspondence with the number of worker modules. A worker module can have multiple daemons, which facilitates communication management between different daemons.

[0072] Furthermore, the cloud computing-based daemon process management system of the present application adopts a multi-level management approach to the daemon process, namely cluster level, application level, worker level and process level. Each level is equipped with a corresponding management module, so that different modules at the same level can run independently, thereby avoiding resource conflicts and facilitating management.

[0073] Specifically, if the cloud computing platform creates an instance of big data processing and analysis, there are two tasks in this instance, namely data processing task and data analysis task. For the data processing task, there are multiple daemons. Based on the above information, it can be determined that the number of cluster-level encapsulations in the cluster-level configuration information is 1, and the number of application-level encapsulations is 2. According to the correlation between different daemons, it is determined that multiple daemons can be encapsulated into three worker modules. The number of daemons corresponding to each worker module is 1, 2 and 3 respectively, and the number of worker-level configurations is 3.

[0074] In one embodiment, in order to facilitate the user to input multi-level configuration information, before inputting the multi-level configuration information, the cluster-level configuration information, the application-level configuration information, the worker-level configuration information and the process-level configuration information are generated into a configuration information relationship table according to the configuration requirements. In addition, through the process management backend 201, the user can input the configuration information corresponding to each level. The configuration information of each level includes the ID number, name, IP address, encapsulation quantity, encapsulation type, running quantity and running status. Taking the worker-level configuration information in the above embodiment as an example, the following worker-level configuration information relationship table is established:

[0075] Table 1: Worker-level configuration information relationship table.

[0076]

[0077] In the above worker-level configuration information relationship table, taking the ID number w010101 as an example, 01-01-01 represents the cluster level, application level, and worker level numbers respectively, and "worker-process" indicates that the application module is the worker module under the data processing task.

[0078] In some embodiments, the step S100 further includes:

[0079] Based on the multi-level configuration information, selecting a configuration mode for the configuration of the worker module, wherein the configuration mode includes a unified configuration and a machine configuration;

[0080] The unified configuration is used to configure each worker module with the same number of daemon processes;

[0081] The machine configuration is used to configure each worker module according to a preset number of daemon processes.

[0082] Specifically, since when one or more of the multiple daemon processes are encapsulated as a worker module, the number of daemon processes in different worker modules may be the same or different. Therefore, in this embodiment, a configuration mode selection module is also provided. The configuration mode selection module provides an interface for the user to select the configuration mode. The user can select two different configuration modes for the configuration of the worker module: unified configuration or machine configuration. For the unified configuration, the number of daemon processes of each worker module is the same. For the machine configuration, the number of daemon processes of each worker module is different. It can be understood that each worker module corresponds to an IP address. Specifically, as shown in the above-mentioned worker-level configuration information relationship table, during the configuration process, it is necessary to select the configuration method of the machine configuration. In the worker module named "worker-process", a corresponding number of IP addresses are allocated according to the number of encapsulations, and the corresponding IP addresses are selected to configure 1, 2 and 3 daemons respectively. At this time, each worker module corresponds to an IP address; if in the above-mentioned worker-level configuration information relationship table, the number of daemons in each worker module is 1, a unified configuration method can be selected, and one daemon process is allocated to three worker modules at the same time, thereby improving the configuration speed of the worker module. Therefore, there is no need to use the same number of daemons for each worker module on average. The number of daemons running in each worker module can be determined according to the needs of the tasks to be executed, thereby being able to meet application scenarios of different scales and requirements.

[0083] S200, implementing multi-level management of the daemon process based on the multi-level configuration information;

[0084] In some implementations, the step S200 includes:

[0085] Performing multi-level configuration on the daemon process according to the multi-level configuration information;

[0086] Each level of process management unit corresponds to each layer of operation layer, so that each layer of operation layer performs corresponding operation, and the operation operation includes starting, stopping and restarting;

[0087] Feeding back the execution result of the running operation to the data management unit, and implementing multi-level management of the daemon process based on the execution result;

[0088] Among them, the process management unit includes a cluster management module, an application management module, a worker management module and a process management module. The cluster management module is used to manage the operation of at least one cluster module in the cloud computing platform, the application management module is used to manage the operation of at least one application module in the cluster module, the worker management module is used to manage the operation of at least one worker module in the application module, and the process management module is used to manage the operation of at least one daemon process in the worker module. The operation layer includes a cluster layer, an application layer, a worker layer and a process layer.

[0089] In this embodiment, multi-level management of the daemon process is implemented through the service management unit 202. Specifically, the service management unit 202 includes a configuration layer and an operation layer. The configuration layer obtains multi-level configuration information input by the user, and performs remote configuration and local configuration in sequence based on the multi-level configuration information. An IP address is deployed for each level of module to facilitate communication between different modules. At the same time, the operation layer is provided with a cluster layer, an application layer, a worker layer and a process layer, and the cluster layer, the application layer, the worker layer and the process layer correspond to the cluster management module, the application management module, the worker management module and the process management module respectively. Based on the multi-level configuration information, each operation layer performs corresponding operation operations. Specifically, when performing the corresponding operation operations, based on the configuration information, the CheckingConfig() program for checking the configuration is triggered, and the operation operations in each operation layer are executed accordingly, and the execution results of the operation operations are transmitted to the data management unit. The data management unit feeds back the above execution results to the process management unit. The process management unit implements multi-level management of the daemon process based on the above execution results.

[0090] In some implementations, configuring the daemon process at multiple levels further includes:

[0091] Allocate independent CPU, memory resources and disk space to each daemon process;

[0092] Use virtual file system technology to create an independent file system for each daemon process;

[0093] Using network virtualization technology, each daemon process is allocated an independent network space;

[0094] Obtaining security permission data of the daemon process, and determining access rights of the daemon process based on the security permission data;

[0095] Use RabbitMQ to communicate between different daemons;

[0096] The worker module manages communication between different daemons and process-level configuration information.

[0097] In this embodiment, based on the process level configuration information, the system automatically builds a virtualized environment and instantiates an isolated operating space for each daemon process, thereby avoiding resource conflicts and improving the security of the daemon process. The process level configuration information also includes resource usage data and security permission data.

[0098] Specifically, based on the resource usage data of the daemon process, an independent CPU, memory resources and disk space are allocated to each daemon process. At the same time, the virtual file system technology is used to create an independent file system for each daemon process, and file system isolation is achieved. In addition, the network virtualization technology is used to allocate an independent network space to each daemon process, thereby achieving network isolation; based on the security permission data, the daemon process is set as the smallest management object without further subdivision, and each daemon process can only perform corresponding operations based on the security permission data; for communication between different daemons, data is exchanged based on the RabbitMQ method, and a worker module is set at the same time, so that the process-level configuration information can be centrally managed and dynamically updated between different daemons, thereby achieving unified management of daemons.

[0099] Specifically, suppose there are three daemons A, B, and C in a worker module. Set 10% of the CPU, 1GB of memory, and 10GB of disk space for daemon A, 20% of the CPU, 2GB of memory, and 15GB of disk space for daemon B, and 15% of the CPU, 5GB of memory, and 15GB of disk space for daemon C. For independent file systems, set independent file system paths for daemons A, B, and C, for example, set the file system path for daemon A: / var / lib / daemon_a. The file system stores resource usage data and operation logs during the running of the daemon process, and the path is invisible to daemons B and C, thereby ensuring the independence of daemon A. In terms of network isolation, set up independent network namespaces for daemons A, B, and C. For example, daemon A can be assigned to the 19160 / 24 subnet, while daemon B is in the 19160 / 24 subnet and daemon C is in the 19160 / 24 subnet to ensure that there is no conflict in network resources between them. For security permissions, determine each daemon as the smallest management object and grant only necessary permissions to avoid over-authorization, thereby ensuring the security of the daemon. The security permission data includes file access permissions, network port permissions, system call permissions, etc. For permission control, assuming that daemon A is responsible for file processing, the security permission data of daemon A is file access permissions. At this time, daemon A can only perform operations to access specific files, but cannot perform operations corresponding to the security permission data such as network port permissions and system call permissions. In addition, if daemon A and daemon B need to exchange data, a message queue "ABQueue" is configured through RabbitMQ, daemon A publishes data to this message queue, and daemon B consumes data from the message queue.

[0100] In some embodiments, a running log path is also provided in the file system of each daemon process for storing running logs during the running of the daemon process. Users can view the running log file to obtain detailed error information and running status, and use lifecycle management tools such as systemd to handle the starting, stopping and restarting of the daemon process.

[0101] S300, monitor the operation of the daemon process in real time, generate a worker resource mapping data table, obtain the operation status of the daemon process based on the worker resource mapping data table, and if the operation status of the daemon process is in an abnormal state, configure a new worker module for the daemon process, wherein one worker module is configured with at least one daemon process, the worker resource mapping data table is a mapping table of resource usage data of the worker module and the daemon process, and the abnormal state is that the operation of the daemon process causes the worker module to be in a high-load state;

[0102] In some embodiments, the step S300 includes:

[0103] By comparing the resource usage data of each worker module with a preset resource usage threshold in turn, a worker module in a high-load state is obtained, wherein the resource usage data includes CPU usage, memory occupancy, and network I / O usage;

[0104] The expert scoring method is used to score the CPU usage, memory usage, and network I / O usage of each daemon process to obtain a comprehensive score for each daemon process;

[0105] Based on the comprehensive score, obtaining the daemon process that needs to be configured with a new worker module;

[0106] A new worker module is configured for the daemon process, and resource usage data of the new worker module and the daemon process in the new worker module are added to the worker resource mapping data table.

[0107] In this embodiment, the worker modules that are in a high-load state during the operation of the daemon process are first determined, and on this basis, the daemon process that needs to add a new worker module to relieve the load pressure is determined, thereby achieving precise management of the daemon process. The multi-level management method reduces the probability of management errors, thereby ensuring that each task in the cloud computing platform can run stably.

[0108] In this embodiment, the resource usage data of each worker module is obtained, and by comparing the resource usage data of each worker module with a preset resource usage threshold, it is determined whether the worker module is in a high-load state. Specifically, the CPU usage rate, memory occupancy rate, and network I / O usage in the resource usage data of each worker module are compared in turn to see whether they are higher than the preset resource usage threshold. If the CPU usage rate, memory occupancy rate, and network I / O usage in the resource usage data are all higher than the preset resource usage threshold, then the worker module is in a high-load state.

[0109] In one embodiment, resource usage data of each worker module is obtained, and by comparing the resource usage data of each worker module with a preset resource usage threshold, it is determined whether the worker module is in a high-load state, including: obtaining the resource usage data of each worker module in a historical worker resource mapping data table, and performing data analysis on the resource usage data of each worker module in the past week, if the probability that each data in the resource usage data of the worker module in the past week is higher than the preset resource usage threshold is greater than 80%, it means that the worker module is in a high-load state, further ensuring the accuracy of daemon process management.

[0110] Furthermore, the resource usage data of each daemon process in the worker module in a high-load state is obtained, and the new worker module that needs to be created is determined by an expert scoring method to relieve the load pressure of the daemon process, and the resource usage data of the new worker module and the resource usage data of the daemon process in the new worker module are added to the worker resource mapping data table. The problem of high-load operation of the worker module is handled by the above method, thereby ensuring the reasonable allocation and efficient operation of resources in the cloud computing platform.

[0111] Specifically, among the preset resource usage thresholds, the threshold for CPU usage is 80%, the threshold for memory usage is 65%, and the threshold for network I / O usage is 200MB / s; based on the worker-level configuration information relationship table in Table 1, the resource usage data mapping table for each worker module in Table 1 is obtained, as shown in the following table:

[0112] Table 2: Resource usage data mapping table of worker module.

[0113]

[0114] In Table 2 above, the CPU usage, memory occupancy, and network I / O usage of w010103 are all higher than the preset resource usage threshold, indicating that w010103 is processing a large amount of data transmission and w010103 is in a high load state. At this time, the resource usage data mapping table of each daemon process in w010103 is obtained.

[0115] Table 3: Resource usage data mapping table for each daemon process in w010103:

[0116] Daemon process id CPU usage Memory usage Network I / O usage p01010301 45％ 20％ 150MB / s p01010302 50％ 65％ 245MB / p01010303 20％ 10％ 60MB / s

[0117] Based on Table 3 above, the CPU usage, memory usage, and network I / O usage of each daemon process are scored using the expert scoring method to obtain a comprehensive score for each daemon process. Specifically, the CPU usage, memory usage, and network I / O usage of each daemon process are scored using a percentage system. If the comprehensive score of the daemon process is higher than 600 points, it means that there is no need to create a new worker for the daemon process.

[0118] In one embodiment, the scoring criteria for daemons are as follows: for CPU usage, a daemon with a CPU usage of less than 20% is scored as 90 points, a daemon with a CPU usage between 20% and 60% is scored as 70 points, and a daemon with a CPU usage above 60% is scored as 50 points; for memory occupancy, a daemon with a memory occupancy of less than 30% is scored as 90 points, a daemon with a memory occupancy between 30% and 60% is scored as 70 points, and a daemon with a memory occupancy above 60% is scored as 50 points. The score is 50 points. For network I / O usage, the score for daemons with network I / O usage below 80MB / s is 90 points, the score for daemons with network I / O usage between 80MB / s and 200MB / s is 70 points, and the score for daemons with network I / O usage above 200MB / s is 50 points. At the same time, weights are set according to the importance of resources in the cloud computing platform, with the weight of CPU usage being 4, the weight of memory usage being 3, and the weight of network I / O usage being 3. According to the above scoring criteria, each daemon process is scored, and the weighted average method is used to perform weighted summation of the scores of each daemon process. The comprehensive score of the daemon process = CPU usage score × 4 + memory occupancy score × 3 + network I / O usage score × 3. Therefore, the comprehensive scores of the three daemons are 760 points, 580 points, and 900 points, respectively. Since 580 points < 600 points, it means that the daemon process p01010302 needs to be configured with a new worker module to share the tasks it executes. After that, the resource usage data of the new worker module configured for the daemon process p01010302 will be mapped to the worker resource data table to ensure data integrity and timeliness of monitoring.

[0119] S400, monitoring the configuration process of the new worker module in real time, and if the configuration of the new worker module fails, triggering the capacity expansion management mechanism, wherein the capacity expansion management mechanism is to notify the administrator to modify the multi-level configuration information and perform the capacity expansion operation;

[0120] In some embodiments, the step S400 includes:

[0121] Obtain a cluster resource mapping data table, and determine the load status of each cluster node in the cluster based on the cluster resource mapping data table, wherein the cluster resource mapping data table is a basic usage data mapping table of each cluster node in the cluster of the cloud computing platform, and the basic usage data includes the IP address, host name, number of CPU cores, memory capacity, disk space, network bandwidth, load status and network status of each cluster node, and the load status includes a high load status, a normal status and an offline status;

[0122] If there is no cluster node in the cluster whose load status is normal, the capacity expansion management mechanism is triggered;

[0123] If there is a cluster node with a normal load status in the cluster, the cluster node with a normal load status is selected to configure a new worker module;

[0124] After configuring a new worker module, if there is a daemon process that is in an abnormal state, the expansion mechanism is triggered.

[0125] In this embodiment, it is necessary to select cluster nodes with normal load status to configure new worker modules, so as to reduce the impact of the configuration of new worker modules on other daemons. Therefore, if it is detected that a new worker module needs to be configured for a daemon process, and there are no cluster nodes in the cluster of the cloud computing platform that are in normal load status, it is necessary to expand the capacity of the cloud computing platform, that is, increase the number of cluster nodes, and redistribute the relevant tasks in the daemon process by adding new cluster nodes to relieve the load pressure of the daemon process. In addition, if after configuring the new worker module, there is still a situation where the daemon process is running and the worker module is in a high load state, it means that the addition of the worker module has failed, and the cluster nodes of the cluster in the cloud computing platform cannot meet the stable operation of the daemon process, so it is necessary to expand the capacity of the cloud computing platform.

[0126] In one embodiment, the hardware configuration of each cluster node is queried through a remote management tool, wherein the hardware configuration includes the number of CPU cores and memory capacity; the disk space and usage of each cluster node are obtained through a hard disk management tool, and the network status and IP address of each cluster node are obtained through a network bandwidth testing tool; the load status of the cluster node is determined by detecting the CPU usage and memory usage of each cluster node; a connectivity test is performed on each cluster node through a network detection tool to obtain the operating status of each cluster node, wherein the operating status includes an offline status and an online status; through the above method, the basic usage data of each cluster node is obtained in turn, and a cluster resource mapping data table is established.

[0127] Specifically, in an instance of big data processing and analysis created for a cloud computing platform, the number of cluster nodes in the cluster is 20. If the CPU usage of a cluster node exceeds 80% or the memory occupancy exceeds 85%, the cluster node is in a high-load state. If a cluster node cannot respond to a Ping command, it means that the cluster node is offline. In the cluster resource mapping data table of 20 cluster nodes established by the above method, 2 cluster nodes are offline and 2 are in a high-load state. All cluster nodes in a high-load state and cluster nodes in an offline state are listed as objects that require priority attention. Cluster nodes in a high-load state need to be expanded or optimized, and cluster nodes in an offline state need to be checked for hardware or network problems.

[0128] In one embodiment, the cluster resource mapping data table is automatically updated every 30 minutes. If the load of a cluster node that was originally in a high-load state drops to a normal level after optimization, its load state is updated to normal. When configuring a new worker module, cluster nodes with a normal load state are preferentially selected.

[0129] In some embodiments, the notifying the administrator to modify the multi-level configuration information and perform the capacity expansion operation includes:

[0130] Marking the daemon process in an abnormal state and the worker module corresponding to the daemon process;

[0131] Sending the cluster resource mapping data table and the marking result to an administrator, and notifying the administrator to modify the multi-level configuration information;

[0132] Based on the modified multi-level configuration information and marking results, increase the number of cluster nodes;

[0133] The expansion operation is performed in batches in multiple time intervals, the number of cluster nodes that can be added concurrently in each time interval is obtained, and at least one new cluster node is set for each time interval;

[0134] By analyzing the tasks in the marked daemons, the daemons in abnormal states are configured to new cluster nodes, and new worker modules are configured based on the correlation between different daemons.

[0135] In this embodiment, for each daemon process in an abnormal state, by analyzing its related configuration files and log files, the worker module corresponding to the daemon process is determined, and the corresponding relationship between the daemon process and the worker module is marked. The daemon process and cluster resource mapping data table marked with abnormalities are sent to the administrator, and the marking results of the abnormal daemon process are attached. The administrator analyzes the causes and impact range of the abnormality according to the cluster resource mapping data table and the marking results of the abnormal daemon process, and determines the number of cluster nodes that need to be increased and which configuration parameters need to be adjusted and optimized. The administrator modifies the relevant multi-level configuration information through the process management background 201, including adjusting the number of abnormal daemons and worker modules, memory limits, timeouts and other parameters, and optimizing the resource allocation strategy of cluster nodes. After the modification is completed, the updated configuration information is synchronized to each cluster node of the cluster, and the affected daemons and worker modules are restarted to ensure that the modified configuration takes effect.

[0136] In this embodiment, based on the cluster resource mapping data table, cluster nodes in a high-load state and an offline state are obtained, and the hardware and network problems of the cluster nodes in an offline state are checked. After eliminating the hardware and network problems, if the cluster node is still in an offline state, the cluster node is deleted and a new cluster node is added to replace it. In addition, for the cluster nodes in a high-load state, the number of cluster nodes that need to be added is determined according to the marking results. Based on this, the administrator determines the number of cluster nodes that need to be added.

[0137] In this embodiment, by analyzing the historical access traffic data of the tasks corresponding to each cluster node in the cluster in the cloud computing platform during execution, the average concurrent access volume and peak concurrent access volume in different time periods are obtained. Prediction algorithms such as time series analysis are used, combined with historical data and business growth trends, to predict the concurrent access volume of each time interval in the future. At the same time, according to the tasks executed by the marked daemon process, the maximum concurrent access volume that each newly added cluster node can bear is determined, and then the number of cluster nodes that can be added concurrently in each time interval is obtained. Then, the expansion operation is divided into multiple time intervals in a batch manner, and based on the number of cluster nodes that can be added concurrently in each time interval, a new cluster node is set for each of the time intervals.

[0138] In this embodiment, the task list of the tasks in the marked daemon process is obtained, and the urgency scoring function is used according to the urgency attribute of the task, and the urgency weight of the task is obtained by calculating the urgency score of each task. According to the priority attribute of the task, the priority scoring function is used to calculate the priority score of each task to obtain the priority weight of the task. According to the deadline attribute of the task, the deadline scoring function is used to calculate the deadline score of each task to obtain the deadline weight of the task. The weighted summation method is used to weight the urgency weight, priority weight and deadline weight of the task to obtain the comprehensive weight of each task in the marked daemon process. According to the task attributes of the marked abnormal daemon process, its business relevance and resource occupancy are obtained, and the list of daemons that need to be migrated is determined. By analyzing the correlation between different daemons, the dependency graph of the daemon process is obtained, and the migration order and deployment strategy are determined according to the comprehensive weight and dependency graph of each task. Specifically, the optimal deployment node is determined according to the resource requirements of the daemon process. According to the determined deployment node and the multi-level configuration information modified by the administrator, the container orchestration technology is used to deploy the daemon process to the worker module.

[0139] S500. Monitor the execution result of the capacity expansion operation in real time, generate a capacity expansion monitoring report, and trigger a management reward mechanism based on the capacity expansion monitoring report, wherein the capacity expansion monitoring report includes administrator information, configuration information modification records, capacity expansion monitoring data, and performance improvement data.

[0140] In some embodiments, the step S500 includes:

[0141] Based on the cluster after the capacity expansion operation, obtain a worker resource mapping data table and a cluster resource mapping data table, and monitor various data in the worker resource mapping data table and the cluster resource mapping data table;

[0142] If the performance corresponding to each data is improved to the preset baseline, the administrator information, configuration modification information and timestamp information of the administrator who performs the modification task are obtained through the operation log;

[0143] Determine the administrator's reward level based on the significant degree of performance improvement and the preset reward plan;

[0144] The reward level is obtained and a reward notification is generated, wherein the reward notification includes the reward content, the name of the administrator who receives the reward and the specific performance optimization details.

[0145] In this embodiment, within one hour after the expansion, the resource usage data and key performance data in the worker resource mapping data table are continuously monitored. The key performance includes response time, throughput and CPU usage. If the data in the resource usage data of the worker module is lower than the preset resource usage threshold, it means that the expansion effect is good. At this time, the management reward mechanism is triggered to detect the significant degree of performance improvement in the next month in real time, generate an expansion monitoring report, and reward the corresponding administrator to improve the enthusiasm of employees. If the resources in the resource usage data of the worker module are still higher than the preset resource usage threshold, it means that the expansion effect is not good. At this time, the warning mechanism is triggered to notify the administrator to further modify the multi-level configuration information and system resources, and generate an expansion monitoring result report. At this time, the expansion detection report includes administrator information, configuration information modification records, expansion monitoring data, performance degradation data and optimization suggestions.

[0146] In one embodiment, timestamp and digital signature technology are used to ensure the non-tamperability of the operation log, thereby forming a complete operation audit chain. Specifically, the specific time when the operation occurs is obtained through the system clock, accurate to milliseconds, and then the timestamp is obtained. The administrator's username or account is obtained according to the user login information, and the administrator identification is obtained. At the same time, the administrator's behavior is identified and the administrator's operation type is judged through the multi-level configuration information input by the process management background 201 and the call of the command of the operation operation in the service management unit 202. If the operation is executed normally, the execution result is recorded as success and its detailed information. If the operation fails, the reason for the failure and related information are recorded. The request parameters submitted by the administrator are obtained, wherein the request parameters include the input configuration information data, and the request parameters are recorded in the operation log. The source IP address of the request is obtained through the network layer, and the request source IP address is recorded. The log entry is digitally signed using a hash algorithm combined with a private key to obtain a unique signature for each record to prevent log tampering. After that, the collected key information will be integrated into a complete log record and stored on a secure log server to form an operation audit chain. The key information includes timestamp, administrator ID, operation type, execution object, execution result, request parameters, request source IP, digital signature, etc.

[0147] Specifically, within one hour after the expansion, the resource usage data in the worker resource mapping data table is collected every 5 minutes, and the probability that each data in the resource usage data of the worker module within one hour is higher than the preset resource usage threshold is obtained. If the probability value is less than 20%, it means that the expansion effect is good. At this time, the management reward mechanism is triggered. In the next month, the key performance data is monitored in real time to generate performance improvement data. The performance improvement data is the average value of each key performance data within one month. The performance improvement data is compared with the average value of the key performance data within one week before the expansion to determine the significance of the performance change. If the key performance of one of the key performance data is improved by 10%, the performance improvement reaches the preset baseline, and the corresponding administrator can obtain a first-level reward. If the key performance of one of the key performance data is improved by 20%, the corresponding administrator can obtain a second-level reward. If the key performance of one of the key performance data is improved by 30%, the corresponding administrator can obtain a third-level reward. The reward level of the administrator is determined by analogy. The corresponding relationship between the corresponding performance improvement index and the reward level can be modified and determined by the company leader. For example, in a capacity expansion operation, the response time improves by 10% within one month. Through the operation log, it can be determined that the performance improvement effect was caused by Zhang San modifying the configuration information on October 1, 2023. Then, it can be determined that Zhang San's reward level is a first-level reward, and a reward notification for Zhang San is generated. At the same time, the reward notification is publicized and fed back to Zhang San himself and Zhang San's superiors. The superiors will make the final confirmation and then issue the corresponding reward.

[0148] See also Figure 2 As shown, the present invention also provides a daemon process management system based on cloud computing, the system comprising:

[0149] Process management backend 201: used to obtain multi-level configuration information of the daemon process, the multi-level configuration information includes cluster level configuration information, application level configuration information, worker level configuration information and process level configuration information;

[0150] The service management unit 202 is used to implement multi-level management of the daemon process based on the multi-level configuration information;

[0151] A new worker module configuration unit 203 is used to monitor the operation of the daemon process in real time, generate a worker resource mapping data table, obtain the operation status of the daemon process based on the worker resource mapping data table, and configure a new worker module for the daemon process if the operation status of the daemon process is in an abnormal state, wherein one worker module is configured with at least one daemon process, the worker resource mapping data table is a mapping table of resource usage data of the worker module and the daemon process, and the abnormal state is that the operation of the daemon process causes the worker module to be in a high-load state;

[0152] The capacity expansion unit 204 is used to monitor the configuration process of the new worker module in real time. If the configuration of the new worker module fails, the capacity expansion management mechanism is triggered, wherein the capacity expansion management mechanism is to notify the administrator to modify the multi-level configuration information and perform the capacity expansion operation;

[0153] Management reward unit 205: used to monitor the execution results of the capacity expansion operation in real time, generate a capacity expansion monitoring report, and trigger a management reward mechanism based on the capacity expansion monitoring report, wherein the capacity expansion monitoring report includes administrator information, configuration information modification records, capacity expansion monitoring data and performance improvement data.

[0154] It is understandable that if Figure 1 The contents of the cloud computing-based daemon process management method embodiment shown in the figure are applicable to the cloud computing-based daemon process management system embodiment. The functions specifically implemented by the cloud computing-based daemon process management system embodiment are similar to those in the embodiment shown in the figure. Figure 1 The embodiment of the daemon process management method based on cloud computing shown in FIG. 1 is the same as that shown in FIG. 1 , and the beneficial effects achieved are the same as those of FIG. Figure 1 The beneficial effects achieved by the embodiment of the daemon process management method based on cloud computing are also the same.

[0155] It should be noted that the information interaction, execution process and other contents between the above-mentioned systems are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0156] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0157] See also Figure 3 As shown, an embodiment of the present invention further provides a computer device 3, comprising: a memory 302 and a processor 301 and a computer program 303 stored in the memory 302. When the computer program 303 is executed on the processor 301, a cloud computing-based daemon process management method as described in any one of the above methods is implemented.

[0158] The computer device 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that Figure 3 It is only an example of computer device 3 and does not constitute a limitation on computer device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, etc.

[0159] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0160] In some embodiments, the memory 302 may be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 302 may also be an external storage device of the computer device 3, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 3. Further, the memory 302 may also include both an internal storage unit of the computer device 3 and an external storage device. The memory 302 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0161] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A daemon process management method based on cloud computing, characterized in that: include: Obtain multi-level configuration information of the daemon process, wherein the multi-level configuration information includes cluster-level configuration information, application-level configuration information, worker-level configuration information, and process-level configuration information; Based on the multi-level configuration information, implement multi-level management of the daemon process; Monitor the operation of the daemon process in real time, generate a worker resource mapping data table, obtain the operation status of the daemon process based on the worker resource mapping data table, and if the operation status of the daemon process is in an abnormal state, configure a new worker module for the daemon process, wherein one worker module is configured with at least one daemon process, the worker resource mapping data table is a mapping table of resource usage data of the worker module and the daemon process, and the abnormal state is that the operation of the daemon process causes the worker module to be in a high-load state; Monitor the configuration process of the new worker module in real time. If the configuration of the new worker module fails, trigger the capacity expansion management mechanism, wherein the capacity expansion management mechanism notifies the administrator to modify the multi-level configuration information and perform the capacity expansion operation; The notifying administrator to modify multi-level configuration information and perform capacity expansion operations includes: Marking a daemon process in an abnormal state and a worker module corresponding to the daemon process; Sending the cluster resource mapping data table and the marking result to an administrator, and notifying the administrator to modify the multi-level configuration information; Based on the modified multi-level configuration information and marking results, increase the number of cluster nodes; The expansion operation is performed in batches in multiple time intervals, the number of cluster nodes that can be added concurrently in each time interval is obtained, and at least one new cluster node is set for each time interval; By analyzing the tasks in the marked daemons, the daemons in abnormal state are configured to new cluster nodes, and new worker modules are configured based on the correlation between different daemons. Based on the determined deployment nodes and the multi-level configuration information modified by the administrator, the daemon process is deployed to the worker module using container orchestration technology; Monitor the execution results of the capacity expansion operation in real time, generate a capacity expansion monitoring report, and trigger a management reward mechanism based on the capacity expansion monitoring report, wherein the capacity expansion monitoring report includes administrator information, configuration information modification records, capacity expansion monitoring data, and performance improvement data; The real-time monitoring of the execution result of the capacity expansion operation, generating a capacity expansion monitoring report, and triggering a management reward mechanism based on the capacity expansion monitoring report include: Based on the cluster after the capacity expansion operation, obtain a worker resource mapping data table and a cluster resource mapping data table, and monitor various data in the worker resource mapping data table and the cluster resource mapping data table; If the performance corresponding to each data is improved to the preset baseline, the administrator information, configuration modification information and timestamp information of the administrator who performs the modification task are obtained through the operation log; Determine the administrator's reward level based on the significant degree of performance improvement and the preset reward plan; The reward level is obtained and a reward notification is generated, wherein the reward notification includes the reward content, the name of the administrator who receives the reward and the specific performance optimization details.

2. The method according to claim 1, characterized in that The multi-level configuration of the acquisition daemon process also includes: Based on the multi-level configuration information, selecting a configuration mode for the configuration of the worker module, wherein the configuration mode includes a unified configuration and a machine configuration; The unified configuration is used to configure each worker module with the same number of daemon processes; The machine configuration is used to configure each worker module according to a preset number of daemon processes.

3. The method according to claim 1, characterized in that The multi-level management of the daemon process based on the multi-level configuration information includes: Performing multi-level configuration on the daemon process according to the multi-level configuration information; Each level of process management unit corresponds to each layer of operation layer, so that each layer of operation layer performs corresponding operation, and the operation operation includes starting, stopping and restarting; Feeding back the execution result of the running operation to the data management unit, and implementing multi-level management of the daemon process based on the execution result; Among them, the process management unit includes a cluster management module, an application management module, a worker management module and a process management module. The cluster management module is used to manage the operation of at least one cluster module in the cloud computing platform, the application management module is used to manage the operation of at least one application module in the cluster module, the worker management module is used to manage the operation of at least one worker module in the application module, and the process management module is used to manage the operation of at least one daemon process in the worker module. The operation layer includes a cluster layer, an application layer, a worker layer and a process layer.

4. The method according to claim 3, characterized in that The multi-level configuration of the daemon process also includes: Allocate independent CPU, memory resources and disk space to each daemon process; Use virtual file system technology to create an independent file system for each daemon process; Using network virtualization technology, each daemon process is allocated an independent network space; Obtaining security permission data of the daemon process, and determining access rights of the daemon process based on the security permission data; Use RabbitMQ to communicate between different daemons; The worker module manages communication between different daemons and process-level configuration information.

5. The method according to claim 1, characterized in that The method of obtaining the running status of the daemon process based on the worker resource mapping data table and configuring a new worker module for the daemon process if the running status of the daemon process is in an abnormal state includes: By comparing the resource usage data of each worker module with a preset resource usage threshold in turn, a worker module in a high-load state is obtained, wherein the resource usage data includes CPU usage, memory occupancy, and network I / O usage; The expert scoring method is used to score the CPU usage, memory usage, and network I / O usage of each daemon process to obtain a comprehensive score for each daemon process; Based on the comprehensive score, obtaining the daemon process that needs to be configured with a new worker module; A new worker module is configured for the daemon process, and resource usage data of the new worker module and the daemon process in the new worker module are added to the worker resource mapping data table.

6. The method according to claim 5, characterized in that The real-time monitoring of the configuration process of the new worker module and triggering the expansion management mechanism if the configuration of the new worker module fails include: Obtain a cluster resource mapping data table, and determine the load status of each cluster node in the cluster based on the cluster resource mapping data table, wherein the cluster resource mapping data table is a basic usage data mapping table of each cluster node in the cluster of the cloud computing platform, and the basic usage data includes the IP address, host name, number of CPU cores, memory capacity, disk space, network bandwidth, load status and network status of each cluster node, and the load status includes a high load status, a normal status and an offline status; If there is no cluster node in the cluster whose load status is normal, the capacity expansion management mechanism is triggered; If there is a cluster node with a normal load status in the cluster, the cluster node with a normal load status is selected to configure a new worker module; After configuring a new worker module, if there is a daemon process that is in an abnormal state, the expansion mechanism is triggered.

7. A daemon process management system based on cloud computing, characterized in that: include: Process management background: used to obtain multi-level configuration information of the daemon process, which includes cluster-level configuration information, application-level configuration information, worker-level configuration information, and process-level configuration information; A service management unit: used for implementing multi-level management of the daemon process based on the multi-level configuration information; A new worker module configuration unit: used for monitoring the operation of the daemon process in real time, generating a worker resource mapping data table, obtaining the operation status of the daemon process based on the worker resource mapping data table, and configuring a new worker module for the daemon process if the operation status of the daemon process is in an abnormal state, wherein one worker module is configured with at least one daemon process, the worker resource mapping data table is a mapping table of resource usage data of the worker module and the daemon process, and the abnormal state is that the operation of the daemon process causes the worker module to be in a high-load state; Expansion unit: used to monitor the configuration process of the new worker module in real time. If the configuration of the new worker module fails, the expansion management mechanism is triggered, wherein the expansion management mechanism notifies the administrator to modify the multi-level configuration information and perform the expansion operation; The notifying administrator to modify multi-level configuration information and perform capacity expansion operations includes: Marking a daemon process in an abnormal state and a worker module corresponding to the daemon process; Sending the cluster resource mapping data table and the marking result to an administrator, and notifying the administrator to modify the multi-level configuration information; Based on the modified multi-level configuration information and marking results, increase the number of cluster nodes; The expansion operation is performed in batches in multiple time intervals, the number of cluster nodes that can be added concurrently in each time interval is obtained, and at least one new cluster node is set for each time interval; By analyzing the tasks in the marked daemons, the daemons in abnormal state are configured to new cluster nodes, and new worker modules are configured based on the correlation between different daemons. Based on the determined deployment nodes and the multi-level configuration information modified by the administrator, the daemon process is deployed to the worker module using container orchestration technology; Management reward unit: used to monitor the execution results of the capacity expansion operation in real time, generate a capacity expansion monitoring report, and trigger a management reward mechanism based on the capacity expansion monitoring report, wherein the capacity expansion monitoring report includes administrator information, configuration information modification records, capacity expansion monitoring data, and performance improvement data; The real-time monitoring of the execution result of the capacity expansion operation, generating a capacity expansion monitoring report, and triggering a management reward mechanism based on the capacity expansion monitoring report include: Based on the cluster after the capacity expansion operation, obtain a worker resource mapping data table and a cluster resource mapping data table, and monitor various data in the worker resource mapping data table and the cluster resource mapping data table; If the performance corresponding to each data is improved to the preset baseline, the administrator information, configuration modification information and timestamp information of the administrator who performs the modification task are obtained through the operation log; Determine the administrator's reward level based on the significant degree of performance improvement and the preset reward plan; The reward level is obtained and a reward notification is generated, wherein the reward notification includes the reward content, the name of the administrator who receives the reward and the specific performance optimization details.

8. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

Images