Cloud host scheduling method and device

Abstract

This application relates to a cloud server scheduling method and apparatus. The method includes: a controller in a cloud computing system under a high-density liquid cooling environment determining a target computing node group for a pending service from multiple candidate computing node groups based on the service type of the pending service; each computing node in each candidate computing node group being distributed at different locations of heat dissipation devices in the high-density liquid cooling environment; identifying at least one target computing node in the target computing node group where a cloud server is deployed, and detecting whether the heat dissipation devices corresponding to each target computing node are faulty; and, in the event of a fault in any heat dissipation device, scheduling the cloud server according to other computing nodes in the target computing node group to perform service processing. This method ensures the reliability of cloud server service processing in the cloud computing system.

Description

Technical Field

[0001] This application relates to the field of cloud computing technology, and in particular to a cloud host scheduling method and apparatus. Background Technology

[0002] A cloud computing system is a computing model in which multiple servers act as computing nodes and share computing resources, software and information through a network. Cloud hosts are usually deployed in the computing nodes of a cloud computing system to perform business processing.

[0003] In related technologies, in order to improve the performance and stability of cloud computing systems, cloud computing systems can be built in a high-density liquid-cooled server environment. For example, the blade box containing computing nodes can be deployed on both sides of the liquid-cooled column to reduce the operating temperature of the computing nodes and improve their operating efficiency and reliability.

[0004] However, in related technologies, if the liquid cooling column fails, there will be a technical problem that cannot ensure the reliability of cloud host processing services in the cloud computing system. Summary of the Invention

[0005] Therefore, it is necessary to provide a cloud host scheduling method and apparatus to address the aforementioned technical problems, which can ensure the reliability of cloud host processing services in a cloud computing system.

[0006] In a first aspect, embodiments of this application provide a cloud host scheduling method, applied to a controller in a cloud computing system under a high-density liquid cooling environment, comprising:

[0007] Based on the service type of the service to be processed, the target computing node group for the service to be processed is determined from multiple candidate computing node groups; the computing nodes in each candidate computing node group are distributed and deployed at the locations of different heat dissipation devices in a high-density liquid cooling environment;

[0008] Identify at least one target computing node in the target computing node group that has a cloud host deployed, and detect whether the heat dissipation equipment corresponding to each target computing node is faulty;

[0009] In the event of a malfunction in any of the heat dissipation devices, the cloud host is scheduled based on other computing nodes in the target computing node group to perform business processing.

[0010] In the cloud host scheduling method provided in this application embodiment, the controller in the cloud host system under a high-density liquid cooling environment determines the target computing node group of the service to be processed from multiple candidate computing node groups according to the service type of the service to be processed, and then determines at least one target computing node in the target computing node group that has a cloud host deployed, and detects whether the heat dissipation equipment corresponding to each target computing node is faulty. Then, if the heat dissipation equipment is faulty, the cloud host is scheduled according to other computing nodes in the target computing node group to perform service processing. In this case, each computing node in each candidate computing node group is distributed at the location of different heat dissipation equipment in the high-density liquid cooling environment. In this method, multiple candidate computing nodes are obtained by pre-grouping the computing nodes in the cloud computing system. The computing nodes in each candidate computing node group are distributed at the locations of different heat dissipation devices in a high-density liquid cooling environment. In this way, each candidate computing node group corresponds to multiple heat dissipation devices. Therefore, for any candidate computing node group, when a heat dissipation device in the group fails, the cloud host at the failed heat dissipation device can be scheduled to a computing node at another computing node with a normal heat dissipation device in the same group. Based on this, the target computing node group of the business to be processed can also perform cloud host scheduling. By identifying at least one target computing node with a cloud host deployed in the target computing node group and detecting whether the heat dissipation device corresponding to each target computing node is faulty, if a heat dissipation device fails, the cloud host is scheduled according to the other computing nodes in the target computing node group that are at normal heat dissipation devices. The cloud host at the failed heat dissipation device is scheduled to a computing node at another computing node with a normal heat dissipation device to continue executing the business and ensure the reliability of the cloud host in processing the business.

[0011] In one embodiment, before determining the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed, the method further includes:

[0012] Obtain multiple redundancy levels of computing nodes in a cloud computing system, as well as redundancy grouping factors in a high-density liquid cooling environment; the redundancy grouping factors include at least the location of each heat dissipation device in the high-density liquid cooling environment and the power supply equipment to which each heat dissipation device belongs.

[0013] Based on each redundancy level and redundancy grouping factor, the computing nodes in the cloud computing system are grouped to obtain multiple candidate computing node groups.

[0014] The cloud server scheduling method provided in this application first obtains multiple redundancy levels of computing nodes in the cloud computing system and redundancy grouping factors in a high-density liquid cooling environment. Then, based on each redundancy level and each redundancy grouping factor, the computing nodes in the cloud computing system are grouped to obtain multiple candidate computing node groups. The redundancy grouping factor includes at least the location of each heat dissipation device in the high-density liquid cooling environment and the power supply equipment to which each heat dissipation device belongs. In this method, by grouping the computing nodes in the cloud computing system to obtain multiple candidate computing node groups, and by obtaining multiple redundancy levels and redundancy grouping factors of the computing nodes in the cloud computing system before grouping, the grouping is performed according to redundancy levels, and the redundancy grouping factor is considered during the grouping process. This ensures that the computing nodes in each group are distributed and deployed at the locations of different heat dissipation devices, thereby improving the availability and fault tolerance of the cloud computing system.

[0015] In one embodiment, the computing nodes in the cloud computing system are grouped according to each redundancy level and each redundancy grouping factor to obtain multiple candidate computing node groups, including:

[0016] Based on each redundancy level, obtain the redundancy strategy for each redundancy level; the redundancy strategy represents the number of redundant computing nodes corresponding to each redundancy level.

[0017] For each redundancy level, based on the location of each heat dissipation device and the power supply device to which each device belongs, multiple computing nodes that meet the redundancy strategy of each redundancy level and are located at different heat dissipation devices and different power supply devices are selected from each computing node as the computing nodes corresponding to each redundancy level.

[0018] Each redundancy level corresponds to a computing node as a candidate computing node group, resulting in multiple candidate computing node groups.

[0019] The cloud server scheduling method provided in this application first obtains the redundancy strategy for each redundancy level. Then, for each redundancy level, based on the location of each heat dissipation device and its associated power supply device, multiple computing nodes that conform to the redundancy strategy of each redundancy level and are located at different heat dissipation devices and on different power supply devices are selected from each computing node as the computing nodes corresponding to each redundancy level. Then, the computing nodes corresponding to each redundancy level are grouped as candidate computing node groups, resulting in multiple candidate computing node groups. Here, the redundancy strategy represents the number of redundant computing nodes corresponding to each redundancy level. In this method, different redundancy levels correspond to different redundancy strategies, meaning the number of redundant computing nodes for each redundancy level is different. Based on the redundancy strategy of each redundancy level, a matching number of computing nodes are selected. Furthermore, the selection of these computing nodes also depends on the location of the heat dissipation device and the associated power supply device, ensuring that the selected computing nodes are located at different heat dissipation devices and on different power supply devices. This way, if a heat dissipation device in each group fails, other heat dissipation devices can support business processing, improving the reliability of business processing.

[0020] In one embodiment, before determining the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed, the method further includes:

[0021] Obtain cloud server deployment methods for different business types; cloud server deployment methods include at least master-slave, dual-machine hot standby, and multi-machine hot standby.

[0022] Cloud host deployment is carried out based on the deployment method of each cloud host and multiple candidate computing node groups.

[0023] The cloud server scheduling method provided in this application first obtains the cloud server deployment methods for different business types, and then deploys cloud servers according to each cloud server deployment method and multiple candidate computing node groups. The cloud server deployment methods include at least master-slave, dual-machine hot standby, and multi-machine hot standby. In this method, after obtaining each candidate computing node group, corresponding cloud servers can be deployed on each candidate computing node group for business processing. Since the business types corresponding to each candidate computing node group are different, and different business types correspond to different cloud server deployment methods, it is necessary to deploy cloud servers based on the corresponding cloud server deployment method when deploying on each candidate computing node group to maximize the advantages of cloud computing, meet the differentiated needs of businesses in terms of performance and reliability, and thus improve the reliability of cloud server deployment.

[0024] In one embodiment, cloud host deployment is performed based on the deployment method of each cloud host and multiple candidate computing node groups, including:

[0025] For any candidate computing node group, connect each computing node in the candidate computing node group to form a network group of candidate computing nodes.

[0026] According to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to complete the cloud host deployment of the candidate computing node group.

[0027] In the cloud host scheduling method provided in this application embodiment, for any candidate computing node group, each computing node in the candidate computing node group is network-connected to obtain a candidate computing node network group. Then, according to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to complete the cloud host deployment of the candidate computing node group. In this method, before deploying cloud hosts on each computing node in the candidate computing node group, each computing node is first network-connected. This allows cloud hosts on different computing nodes to communicate and interact with data after subsequent deployment. It also facilitates cloud host scheduling in case of heat dissipation equipment failure. Based on this, cloud hosts are then deployed on each computing node according to the cloud host deployment method, thereby improving the reliability of cloud host deployment.

[0028] In one embodiment, cloud hosts are deployed on each computing node in the candidate computing node network group according to the cloud host deployment method, in order to complete the cloud host deployment of the candidate computing node group, including:

[0029] Perform status monitoring on each computing node to obtain its status information;

[0030] If the status information confirms that all computing nodes are in normal condition, then the computing nodes in the candidate computing node network group that match the cloud host deployment method are determined according to the cloud host deployment method, and the cloud hosts in the candidate computing node group are deployed on the computing nodes that match the cloud host deployment method.

[0031] The cloud server scheduling method provided in this application first monitors the status of each computing node to obtain its status information. Then, based on the status information, if all computing nodes are in a normal state, the method determines the computing nodes in the candidate computing node network group that match the cloud server deployment method, and deploys the cloud servers in the candidate computing node group on the matching computing nodes. This method monitors the status of each computing node to ensure it is normal, guaranteeing that each computing node can deploy cloud servers to process services. When all computing nodes are in a normal state, the method selects the computing node from among the computing nodes to deploy the cloud server, and deploys the cloud server on the selected node. This ensures that the computing node used to deploy the cloud server matches the cloud server deployment method, improving deployment accuracy.

[0032] In one embodiment, obtaining cloud host deployment methods for different business types includes:

[0033] Displays the deployment method configuration interface;

[0034] Obtain the cloud server deployment methods edited by users for each business type in the deployment method configuration interface;

[0035] The edited cloud server deployment method is determined to be the cloud server deployment method for different business types.

[0036] The cloud server scheduling method provided in this application embodiment displays a deployment method configuration interface, thereby obtaining the cloud server deployment methods edited by the user for each business type in the deployment method configuration interface. Then, the edited cloud server deployment methods are determined as the cloud server deployment methods for different business types. In this method, by displaying the deployment method configuration interface when obtaining the cloud server deployment methods, users can flexibly configure the cloud server deployment methods for each business type in the deployment method configuration interface. On the one hand, this visual configuration is easy to understand and adjust, lowers the technical threshold, and improves configuration flexibility; on the other hand, this interface-based configuration is traceable and highly maintainable.

[0037] In one embodiment, detecting whether the heat dissipation equipment corresponding to each target computing node is faulty includes:

[0038] For any heat dissipation device, perform fault detection on both the heat dissipation device and the power supply equipment to which it belongs;

[0039] If at least one of the heat dissipation equipment and the power supply equipment is faulty, then the heat dissipation equipment is determined to be faulty.

[0040] If both the heat dissipation equipment and the power supply equipment are functioning correctly, then the heat dissipation equipment is confirmed to be functioning correctly.

[0041] In the cloud server scheduling method provided in this application embodiment, for any heat dissipation device, fault detection is performed on both the heat dissipation device and its associated power supply device. If at least one of the heat dissipation device and the power supply device is faulty, then the heat dissipation device is determined to be faulty; if neither the heat dissipation device nor the power supply device is faulty, then the heat dissipation device is determined not to be faulty. This method, by performing fault detection on both the heat dissipation device and its associated power supply device, determines whether the heat dissipation device is faulty based on the states of both. Essentially, the states of both the heat dissipation device and its associated power supply device affect the determination of the heat dissipation device's fault status. By performing fault detection on both, the accuracy of heat dissipation device fault detection is improved.

[0042] In one embodiment, in the event of a failure in any of the heat dissipation devices, the cloud host is scheduled based on other computing nodes in the target computing node group for business processing, including:

[0043] Determine the number of cloud hosts that need to be scheduled based on the faulty heat dissipation equipment;

[0044] Select any computing node from other computing nodes that matches the number of cloud hosts to be scheduled and is on a properly cooled device, and schedule the cloud hosts to be scheduled to the selected computing node for business processing.

[0045] The cloud server scheduling method provided in this application first determines the number of cloud servers to be scheduled based on the faulty cooling equipment. Then, it arbitrarily selects a computing node from other computing nodes that matches the number of cloud servers to be scheduled and is on a working cooling device. The cloud servers to be scheduled are then scheduled to the selected computing node for business processing. In this method, when scheduling cloud servers, the system determines which cloud servers need to be scheduled based on the faulty cooling equipment. Then, it selects computing nodes from other computing nodes that can support cloud server scheduling and schedules the cloud servers to the selected computing nodes to continue processing business on the cloud servers, thereby improving the reliability of cloud server business processing.

[0046] Secondly, embodiments of this application also provide a cloud host scheduling device, comprising:

[0047] The node group determination module is used to determine the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed; the computing nodes in each candidate computing node group are distributed and deployed at the locations of different heat dissipation devices in a high-density liquid cooling environment;

[0048] The fault detection module is used to identify at least one target computing node in the target computing node group that has a cloud host deployed, and to detect whether the heat dissipation equipment corresponding to each target computing node is faulty.

[0049] The cloud server scheduling module is used to schedule cloud servers for business processing based on other computing nodes in the target computing node group when there is a failure in the heat dissipation equipment.

[0050] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method in any of the embodiments of the first aspect described above.

[0051] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any of the embodiments of the first aspect described above.

[0052] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method in any of the embodiments of the first aspect described above.

[0053] The cloud host scheduling method and apparatus provided in this application embodiment involve a controller in a cloud host system under a high-density liquid cooling environment determining the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed. The controller then determines at least one target computing node in the target computing node group that is deployed with a cloud host, and detects whether the heat dissipation equipment corresponding to each target computing node is faulty. If any heat dissipation equipment is faulty, the cloud host is scheduled according to other computing nodes in the target computing node group for service processing. Each computing node in each candidate computing node group is distributed at different locations of heat dissipation equipment in the high-density liquid cooling environment. In this method, multiple candidate computing nodes are obtained by pre-grouping the computing nodes in the cloud computing system. The computing nodes in each candidate computing node group are distributed at the locations of different heat dissipation devices in a high-density liquid cooling environment. In this way, each candidate computing node group corresponds to multiple heat dissipation devices. Therefore, for any candidate computing node group, when a heat dissipation device in the group fails, the cloud host at the failed heat dissipation device can be scheduled to a computing node at another computing node with a normal heat dissipation device in the same group. Based on this, the target computing node group of the business to be processed can also perform cloud host scheduling. By identifying at least one target computing node with a cloud host deployed in the target computing node group and detecting whether the heat dissipation device corresponding to each target computing node is faulty, if a heat dissipation device fails, the cloud host is scheduled according to the other computing nodes in the target computing node group that are at normal heat dissipation devices. The cloud host at the failed heat dissipation device is scheduled to a computing node at another computing node with a normal heat dissipation device to continue executing the business and ensure the reliability of the cloud host in processing the business. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is an application environment diagram of a cloud host scheduling method in one embodiment;

[0056] Figure 2 This is a flowchart illustrating a cloud host scheduling method in one embodiment;

[0057] Figure 3 This is a schematic diagram of the process for determining candidate computing node groups in one embodiment;

[0058] Figure 4 This is a schematic diagram of the process for determining candidate computing node groups in another embodiment;

[0059] Figure 5 This is a schematic diagram of a candidate computing node group in one embodiment;

[0060] Figure 6 This is a schematic diagram illustrating the process of deploying a cloud host in one embodiment;

[0061] Figure 7 This is a schematic diagram illustrating the process of deploying a cloud host in another embodiment;

[0062] Figure 8 This is a schematic diagram illustrating the process of deploying a cloud host in another embodiment;

[0063] Figure 9 This is a flowchart illustrating the process of obtaining the cloud host deployment method in one embodiment;

[0064] Figure 10 This is a schematic diagram of a process for detecting whether a heat dissipation device is faulty in one embodiment;

[0065] Figure 11 This is a schematic diagram of the process of scheduling cloud hosts in one embodiment;

[0066] Figure 12 This is a schematic diagram of the structure of a cloud host scheduling device in one embodiment. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0068] The technical background of the embodiments of this application will be described below.

[0069] Cloud computing systems utilize multiple servers as computing nodes, sharing computing resources, software, and information via a network. These computing nodes typically house cloud servers for business processing. Traditional cloud computing systems employ distributed deployment, distributing computing nodes across different server rooms or data centers to achieve high availability and fault tolerance. However, this deployment method requires significant network bandwidth and data synchronization, leading to energy and space waste. To further improve performance and stability, cloud computing systems are often built in high-density liquid-cooled server environments. For example, blade enclosures containing computing nodes are deployed on either side of liquid-cooled columns to reduce the operating temperature of the computing nodes and improve their efficiency and reliability. However, this deployment method is vulnerable to disruptions if the liquid-cooled columns fail, compromising the reliability of cloud server operations within the cloud computing system.

[0070] Based on this, this application provides a cloud server scheduling method. By pre-grouping the computing nodes in the cloud computing system into multiple candidate computing nodes, the computing nodes in each candidate computing node group are distributed across different heat dissipation devices in a high-density liquid cooling environment. Thus, each candidate computing node group corresponds to multiple heat dissipation devices. Therefore, for any candidate computing node group, if a heat dissipation device in that group fails, the cloud server at the failed heat dissipation device can be scheduled to a computing node at another computing node with a normal heat dissipation device in that group. Similarly, cloud server scheduling can be performed on the target computing node group containing the service to be processed. By identifying at least one target computing node in the target computing node group that has a cloud server deployed there, and detecting whether the heat dissipation device corresponding to each target computing node is faulty, if a heat dissipation device fails, cloud server scheduling is performed based on other computing nodes in the target computing node group that are on normal heat dissipation devices. This allows the cloud server at the failed heat dissipation device to be scheduled to a computing node at another computing node with a normal heat dissipation device to continue executing the service, ensuring the reliability of the cloud server processing the service. Of course, the technical solution provided in this application is not limited to solving only the above problems and also has other technical effects, which can be seen in the following embodiments.

[0071] It should be noted that the beneficial effects or technical problems solved by the embodiments of this application are not limited to this one, but may also be other implicit or related problems. For details, please refer to the description of the embodiments below.

[0072] The following describes the application environment of the cloud server scheduling method provided in this application embodiment, which can be applied to computer equipment. This computer equipment can be a controller in a cloud computing system under a high-density liquid cooling environment, and its internal structure diagram can be as follows: Figure 1 As shown. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a cloud host scheduling method. Those skilled in the art will understand that... Figure 1The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0073] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0074] In one exemplary embodiment, such as Figure 2 As shown, a cloud host scheduling method is provided, which can be applied to... Figure 1 The following steps, 201 to 203, are used as an example of computer equipment.

[0075] S201, Based on the service type of the service to be processed, determine the target computing node group of the service to be processed from multiple candidate computing node groups.

[0076] In each candidate computing node group, the computing nodes are distributed at different locations of heat dissipation devices in a high-density liquid cooling environment.

[0077] The business types to be processed include web applications and website hosting, e-commerce and online trading platforms, enterprise office systems, enterprise resource management systems, machine learning and model training, medical and health services, and government and public services. This application embodiment does not specifically limit the business types to be processed.

[0078] Candidate compute node groups are obtained by pre-grouping compute nodes in the cloud computing system based on information such as the location of different heat dissipation devices in a high-density liquid cooling environment. By grouping, the compute nodes in each candidate compute node group are distributed and deployed at the locations of different heat dissipation devices. In this way, each candidate compute node group has compute nodes at different heat dissipation devices. For any candidate compute node group, if a heat dissipation device fails, the compute nodes at other normal heat dissipation devices can still be used. That is, the cloud host at the failed heat dissipation device can be used by these still usable compute nodes to ensure the reliability of the cloud host's business processing.

[0079] In this embodiment of the application, the target computing node group of the service to be processed is first determined according to the service type of the service to be processed.

[0080] In this embodiment, the different candidate computing nodes after grouping are used to process different services. Based on this, the corresponding computing node group can be determined according to the service type. For example, in this embodiment, a correspondence between different candidate computing node groups and different service types is established in advance. Therefore, when determining the target computing node group for the service to be processed, the candidate computing nodes corresponding to the service type of the service to be processed can be matched from the above correspondence between different candidate computing node groups and different service types according to the service type of the service to be processed, and then the matched candidate computing nodes are determined as the target computing node group for the service to be processed.

[0081] S202, identify at least one target computing node in the target computing node group that has a cloud host deployed, and detect whether the heat dissipation equipment corresponding to each target computing node is faulty.

[0082] After identifying the target computing node group for the business to be processed, determine which computing nodes in the target computing node group are deployed with cloud hosts for business processing.

[0083] In one possible implementation, the target computing nodes within the target computing node group that have deployed cloud hosts are determined based on a list of cloud hosts. For example, the list of cloud hosts stores the names or identifiers of candidate computing node groups that have deployed cloud hosts, as well as the computing nodes within those candidate groups that have deployed cloud hosts. Based on this, the name or identifier of the target computing node group can be searched in the list of cloud hosts. Then, the computing nodes corresponding to that target computing node group are the computing nodes with deployed cloud hosts, i.e., the target computing nodes, thus determining the target computing nodes within the target computing node group that have deployed cloud hosts.

[0084] In another implementation, the target compute node with deployed cloud hosts is determined by checking the status of the cloud hosts on each compute node. For example, first, a list of all compute nodes in the target compute node group is obtained. Then, for each compute node, the status of the cloud hosts it hosts is queried, focusing primarily on cloud hosts that are "running" or "deployed". As long as there is at least one cloud host on the node with a running status (or a user-defined "deployed" status, such as created but not started), it is considered a target node.

[0085] Furthermore, based on identifying the target computing nodes where cloud hosts are deployed, the system checks whether the heat dissipation equipment corresponding to each target computing node is faulty. In this embodiment, the heat dissipation equipment in the high-density liquid cooling environment uses liquid cooling columns.

[0086] For example, a liquid cooling column monitoring system in a high-density liquid-cooled environment can be used to detect whether the heat dissipation equipment corresponding to each target computing node is faulty. For instance, the system monitors the liquid cooling column corresponding to each target computing node, acquires monitoring data, and uses this data to determine whether the liquid cooling column is faulty. The monitoring data includes inlet water temperature, outlet water temperature, inlet and return water pressure difference, and actual flow rate. After acquiring the current monitoring data, it is compared with historical monitoring data to determine whether the liquid cooling column is faulty.

[0087] S203: In the event of a failure in any of the heat dissipation devices, the cloud host is scheduled based on other computing nodes in the target computing node group to perform business processing.

[0088] If a fault is detected in any of the cooling devices, the cloud server located at the faulty cooling device needs to be relocated to a computing node at another working cooling device to continue executing services, thus ensuring the reliability of the cloud server's service processing.

[0089] As mentioned earlier, the computing nodes in each candidate computing node group are distributed across different cooling devices in a high-density liquid-cooled environment. Therefore, each candidate computing node group corresponds to multiple cooling devices, and consequently, the target computing node group also corresponds to multiple cooling devices. Thus, when a cooling device failure is detected in the target computing node group, the cloud host is rescheduled based on the other computing nodes located at normally functioning cooling devices within the target computing node group. For example, the cloud host at the failed cooling device is migrated to another computing node at a normally functioning cooling device to continue executing business operations.

[0090] In the cloud host scheduling method provided in this application embodiment, the controller in the cloud host system under a high-density liquid cooling environment determines the target computing node group of the service to be processed from multiple candidate computing node groups according to the service type of the service to be processed, and then determines at least one target computing node in the target computing node group that has a cloud host deployed, and detects whether the heat dissipation equipment corresponding to each target computing node is faulty. Then, if the heat dissipation equipment is faulty, the cloud host is scheduled according to other computing nodes in the target computing node group to perform service processing. In this case, each computing node in each candidate computing node group is distributed at the location of different heat dissipation equipment in the high-density liquid cooling environment. In this method, multiple candidate computing nodes are obtained by pre-grouping the computing nodes in the cloud computing system. The computing nodes in each candidate computing node group are distributed at the locations of different heat dissipation devices in a high-density liquid cooling environment. In this way, each candidate computing node group corresponds to multiple heat dissipation devices. Therefore, for any candidate computing node group, when a heat dissipation device in the group fails, the cloud host at the failed heat dissipation device can be scheduled to a computing node at another computing node with a normal heat dissipation device in the same group. Based on this, the target computing node group of the business to be processed can also perform cloud host scheduling. By identifying at least one target computing node with a cloud host deployed in the target computing node group and detecting whether the heat dissipation device corresponding to each target computing node is faulty, if a heat dissipation device fails, the cloud host is scheduled according to the other computing nodes in the target computing node group that are at normal heat dissipation devices. The cloud host at the failed heat dissipation device is scheduled to a computing node at another computing node with a normal heat dissipation device to continue executing the business and ensure the reliability of the cloud host in processing the business.

[0091] Based on the above embodiments, an embodiment is provided to illustrate the process of determining candidate computing node groups before determining the target computing node group.

[0092] In one exemplary embodiment, such as Figure 3 As shown, before determining the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed, the method further includes:

[0093] S301 obtains multiple redundancy levels of computing nodes in a cloud computing system, as well as redundancy grouping factors in a high-density liquid-cooled environment.

[0094] The redundancy grouping factor includes at least the location of each heat dissipation device in the high-density liquid cooling environment and the power supply equipment to which each heat dissipation device belongs.

[0095] Redundancy levels include at least full redundancy, partial redundancy, and no redundancy. Redundancy grouping factors refer to the factors influencing the grouping of computing nodes in a high-density liquid-cooled environment, including the location of each heat dissipation device, the power supply device to which each heat dissipation device belongs, the location of each computing node, and the service load. Each heat dissipation device can be connected to a separate power supply device, or two heat dissipation devices can be connected to a single power supply device; this embodiment does not limit this.

[0096] For example, redundancy levels and redundancy grouping factors are pre-stored in a database, and computer devices can obtain multiple redundancy levels of computing nodes in the cloud computing system and redundancy grouping factors in a high-density liquid cooling environment from the database.

[0097] For example, redundancy levels and redundancy grouping factors can be obtained through a configuration interface. For instance, a configuration interface is displayed to the user, who then inputs the redundancy level and redundancy grouping factor in the corresponding locations. The user-inputted redundancy level is then determined as multiple redundancy levels for computing nodes in the cloud computing system, and the user-inputted redundancy grouping factor is determined as the redundancy grouping factor for a high-density liquid-cooled environment.

[0098] S302, based on each redundancy level and each redundancy grouping factor, groups the computing nodes in the cloud computing system to obtain multiple candidate computing node groups.

[0099] Based on the redundancy levels and redundancy grouping factors obtained above, the computing nodes in the cloud computing system are grouped to obtain multiple candidate computing node groups.

[0100] In practical applications, the computing nodes in the cloud computing system are grouped according to their redundancy levels. During the grouping process, redundancy grouping factors such as the location of each heat dissipation device, the power supply device to which each heat dissipation device belongs, the location of each computing node, and the business load are considered to ensure that the computing nodes in each group have similar characteristics and performance, thereby determining each group as a candidate computing node group.

[0101] The cloud server scheduling method provided in this application first obtains multiple redundancy levels of computing nodes in the cloud computing system and redundancy grouping factors in a high-density liquid cooling environment. Then, based on each redundancy level and each redundancy grouping factor, the computing nodes in the cloud computing system are grouped to obtain multiple candidate computing node groups. The redundancy grouping factor includes at least the location of each heat dissipation device in the high-density liquid cooling environment and the power supply equipment to which each heat dissipation device belongs. In this method, by grouping the computing nodes in the cloud computing system to obtain multiple candidate computing node groups, and by obtaining multiple redundancy levels and redundancy grouping factors of the computing nodes in the cloud computing system before grouping, the grouping is performed according to redundancy levels, and the redundancy grouping factor is considered during the grouping process. This ensures that the computing nodes in each group are distributed and deployed at the locations of different heat dissipation devices, thereby improving the availability and fault tolerance of the cloud computing system.

[0102] Based on the above embodiments, another embodiment of the process for determining candidate computing node groups is provided for illustration.

[0103] In one exemplary embodiment, such as Figure 4 As shown, based on each redundancy level and redundancy grouping factor, the computing nodes in the cloud computing system are grouped to obtain multiple candidate computing node groups, including:

[0104] S401: Obtain the redundancy strategy for each redundancy level.

[0105] Here, the redundancy strategy represents the number of redundant computing nodes corresponding to each redundancy level.

[0106] In this application embodiment, different redundancy levels correspond to different redundancy strategies, that is, different numbers of redundant computing nodes. For example, in full redundancy, each computing node needs to be configured with an equivalent redundant computing node. In this case, it is necessary to determine the number of primary computing nodes under the full redundancy level, which is the number of redundant computing nodes. If it is partial redundancy, multiple redundant nodes need to be added on the basis of the primary computing nodes. The number of these multiple redundant nodes can be arbitrarily determined, such as one redundant computing node, two redundant computing nodes, etc. If it is no redundancy, the computing nodes have no backups, and the number of redundant computing nodes is zero.

[0107] Redundancy strategies for each redundancy level can be defined by the user, specifying the number of primary compute nodes and redundant compute nodes for each level. Alternatively, the correspondence between redundancy levels and redundancy strategies can be stored in a database, allowing the user to retrieve the redundancy strategy for each level. This redundancy strategy includes not only the number of redundant compute nodes but also the number of primary compute nodes.

[0108] S402, for each redundancy level, selects multiple computing nodes that meet the redundancy strategy of each redundancy level from each computing node based on the location of each heat dissipation device and the power supply device to which each device belongs, and which are located at different heat dissipation devices and different power supply devices, as the computing nodes corresponding to each redundancy level.

[0109] Based on the redundancy strategies for each redundancy level obtained above, and combined with the location of each heat dissipation device in the redundancy grouping factor and the power supply equipment to which each heat dissipation device belongs, nodes are grouped.

[0110] In this embodiment, for any redundancy level, based on the location of each heat dissipation device and the power supply device to which each heat dissipation device belongs, multiple computing nodes that conform to the redundancy strategy of the redundancy level and are located at different heat dissipation devices and on different power supply devices are selected from each computing node as the computing nodes corresponding to the redundancy level. The selected multiple computing nodes include not only redundant computing nodes but also the primary computing node.

[0111] For example, with Figure 5 For example, the high-density liquid cooling environment includes three liquid cooling columns, each with a toolbox on both sides, and each toolbox contains multiple computing nodes. In this embodiment, the computing nodes in the cloud computing system are divided into six groups according to the redundancy level and redundancy grouping factor, and the computing nodes in each group are distributed at the locations of different liquid cooling columns.

[0112] S403, each redundancy level corresponds to a computing node as a candidate computing node group, resulting in multiple candidate computing node groups.

[0113] After obtaining the computing nodes corresponding to each redundancy level, each redundancy level's corresponding computing nodes are treated as a candidate computing node group, thus obtaining multiple candidate computing node groups. Figure 5 For example, six candidate computing node groups are obtained.

[0114] The cloud server scheduling method provided in this application first obtains the redundancy strategy for each redundancy level. Then, for each redundancy level, based on the location of each heat dissipation device and its associated power supply device, multiple computing nodes that conform to the redundancy strategy of each redundancy level and are located at different heat dissipation devices and on different power supply devices are selected from each computing node as the computing nodes corresponding to each redundancy level. Then, the computing nodes corresponding to each redundancy level are grouped as candidate computing node groups, resulting in multiple candidate computing node groups. Here, the redundancy strategy represents the number of redundant computing nodes corresponding to each redundancy level. In this method, different redundancy levels correspond to different redundancy strategies, meaning the number of redundant computing nodes for each redundancy level is different. Based on the redundancy strategy of each redundancy level, a matching number of computing nodes are selected. Furthermore, the selection of these computing nodes also depends on the location of the heat dissipation device and the associated power supply device, ensuring that the selected computing nodes are located at different heat dissipation devices and on different power supply devices. This way, if a heat dissipation device in each group fails, other heat dissipation devices can support business processing, improving the reliability of business processing.

[0115] Based on the above embodiments, an embodiment is provided to illustrate the cloud host deployment process before determining the target computing node group.

[0116] In one exemplary embodiment, such as Figure 6 As shown, before determining the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed, the method further includes:

[0117] S501 provides information on cloud server deployment methods for different business types.

[0118] Among them, cloud server deployment methods include at least master-slave, dual-machine hot standby, and multi-machine hot standby.

[0119] Each candidate computing node group corresponds to a business type, and each business type has different cloud host deployment methods, including master-slave, dual-machine hot standby, and multi-machine hot standby.

[0120] When deploying cloud servers for each candidate compute node group, the deployment method must be based on the cloud server deployment method corresponding to the business type of each candidate compute node group. Therefore, it is necessary to first obtain the cloud server deployment method for each business type.

[0121] For example, the cloud server deployment method can be obtained from the configuration file. Different business types correspond to different configuration files, which are stored in a database. The configuration file for each business type can be retrieved from the database based on its identifier. Then, each configuration file can be parsed to extract the cloud server deployment method for that business type.

[0122] S502 deploys cloud hosts based on their deployment methods and multiple candidate computing node groups.

[0123] Based on the cloud server deployment methods for each business type obtained above, cloud servers are deployed in combination with multiple candidate compute node groups. For example, for each candidate compute node group, cloud servers are deployed on the compute nodes in each candidate compute node group according to the cloud server deployment method corresponding to each candidate compute node group.

[0124] The cloud server scheduling method provided in this application first obtains the cloud server deployment methods for different business types, and then deploys cloud servers according to each cloud server deployment method and multiple candidate computing node groups. The cloud server deployment methods include at least master-slave, dual-machine hot standby, and multi-machine hot standby. In this method, after obtaining each candidate computing node group, corresponding cloud servers can be deployed on each candidate computing node group for business processing. Since the business types corresponding to each candidate computing node group are different, and different business types correspond to different cloud server deployment methods, it is necessary to deploy cloud servers based on the corresponding cloud server deployment method when deploying on each candidate computing node group to maximize the advantages of cloud computing, meet the differentiated needs of businesses in terms of performance and reliability, and thus improve the reliability of cloud server deployment.

[0125] Based on the above embodiments, another embodiment of the cloud host deployment process is provided for illustration.

[0126] In one exemplary embodiment, such as Figure 7 As shown, cloud host deployment is performed based on the deployment methods of each cloud host and multiple candidate compute node groups, including:

[0127] S601, For any candidate computing node group, connect each computing node in the candidate computing node group to form a network group of candidate computing nodes.

[0128] In this embodiment, before deploying the cloud host, each computing node in each candidate computing node group needs to be connected via a network to form a highly available cloud computing system. During the network connection process, technologies such as multi-path redundancy and load balancing can be employed to improve network bandwidth and reliability.

[0129] In practical applications, for any candidate computing node group, a matching network connection technology is determined based on the business characteristics corresponding to the candidate computing node group. Then, the matching network connection technology is used to connect each computing node in the candidate computing node group to obtain the candidate computing node network group.

[0130] For example, for services where unacceptable interruptions are possible or where a single point of failure has a wide impact, multi-path redundancy should be prioritized, such as financial transaction systems and real-time communication platforms. For traffic-intensive, high-concurrency services, load balancing should be prioritized, such as web server clusters and big data computing clusters.

[0131] S602, according to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to complete the cloud host deployment of the candidate computing node group.

[0132] After connecting the computing nodes in the candidate computing node group through the network, cloud hosts can be deployed on each computing node so that the cloud hosts on each computing node can communicate and facilitate the scheduling of cloud hosts.

[0133] During the deployment process, the cloud host deployment method corresponding to the candidate computing node group is obtained. Then, according to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to perform business processing on the candidate computing node group.

[0134] In the cloud host scheduling method provided in this application embodiment, for any candidate computing node group, each computing node in the candidate computing node group is network-connected to obtain a candidate computing node network group. Then, according to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to complete the cloud host deployment of the candidate computing node group. In this method, before deploying cloud hosts on each computing node in the candidate computing node group, each computing node is first network-connected. This allows cloud hosts on different computing nodes to communicate and interact with data after subsequent deployment. It also facilitates cloud host scheduling in case of heat dissipation equipment failure. Based on this, cloud hosts are then deployed on each computing node according to the cloud host deployment method, thereby improving the reliability of cloud host deployment.

[0135] Based on the above embodiments, another embodiment of the cloud host deployment process is provided for illustration.

[0136] In one exemplary embodiment, such as Figure 8 As shown, cloud hosts are deployed on each computing node in the candidate computing node network group according to the cloud host deployment method, in order to complete the cloud host deployment of the candidate computing node group, including:

[0137] S701 monitors the status of each computing node and obtains the status information of each computing node.

[0138] In this embodiment of the application, before deploying the cloud host, the status of each computing node in the candidate computing node network group is monitored to obtain the status information of each computing node.

[0139] In applications, cloud computing systems are equipped with monitoring and management systems that can monitor the status of each computing node in real time, thereby obtaining status information of each computing node.

[0140] S702, after determining that all computing nodes are in normal condition based on the status information, determine the computing nodes in the candidate computing node network group that match the cloud host deployment method according to the cloud host deployment method, and deploy the cloud hosts of the candidate computing node group on the computing nodes that match the cloud host deployment method.

[0141] Based on the status information of each computing node obtained above, it is determined whether each computing node is in a normal state. When all computing nodes are in a normal state, cloud host deployment is performed according to the corresponding cloud host deployment method.

[0142] In this embodiment, when deploying a cloud host, the first step is to determine the computing nodes in the candidate computing node network group that match the cloud host deployment method, based on the corresponding cloud host deployment method. For example, if the cloud host deployment method is master-slave, then any one computing node is selected from the candidate computing node network group for cloud host deployment, and the other computing node is used as a slave node; if the cloud host deployment method is dual-machine hot standby, then any two computing nodes are selected from the candidate computing node network group for cloud host deployment, meaning that cloud hosts will be deployed on both computing nodes to jointly undertake the business; if the cloud host deployment method is multi-machine hot standby, then multiple computing nodes are selected from the candidate computing node network group for cloud host deployment, meaning that cloud hosts will be deployed on all of these multiple computing nodes, and the multiple cloud hosts will jointly process the business.

[0143] Furthermore, after identifying the computing nodes in the candidate computing node network group that match the cloud host deployment method, the cloud hosts corresponding to the services of the candidate computing node group are deployed on the matching computing nodes for service processing.

[0144] The cloud server scheduling method provided in this application first monitors the status of each computing node to obtain its status information. Then, based on the status information, if all computing nodes are in a normal state, the method determines the computing nodes in the candidate computing node network group that match the cloud server deployment method, and deploys the cloud servers in the candidate computing node group on the matching computing nodes. This method monitors the status of each computing node to ensure it is normal, guaranteeing that each computing node can deploy cloud servers to process services. When all computing nodes are in a normal state, the method selects the computing node from among the computing nodes to deploy the cloud server, and deploys the cloud server on the selected node. This ensures that the computing node used to deploy the cloud server matches the cloud server deployment method, improving deployment accuracy.

[0145] Based on the above embodiments, an embodiment is provided to illustrate the process of obtaining the cloud host deployment method.

[0146] In one exemplary embodiment, such as Figure 9 As shown, the cloud host redundancy mechanism for acquiring pending business includes:

[0147] S801 displays the deployment method configuration interface.

[0148] In this embodiment, different cloud server deployment methods for different business types are obtained through a configuration interface. Specifically, when it is necessary to obtain the cloud server deployment method, a deployment method configuration interface is displayed to the user.

[0149] S802 retrieves the cloud host deployment methods edited by the user for each business type in the deployment method configuration interface.

[0150] In the deployment configuration interface, users can edit the deployment methods for cloud servers of different business types.

[0151] In one embodiment, the user directly enters the cloud server deployment method for each service type in the corresponding location on the deployment method configuration interface. For example, different service types have different input areas, and the user enters the cloud server deployment method for each service type in each input area.

[0152] In another embodiment, the deployment method configuration interface provides an edit box for candidate cloud host deployment methods, allowing users to select the cloud host deployment method for each business type. For example, the deployment method configuration interface provides edit boxes for candidate cloud host deployment methods for different business types, allowing users to select the appropriate cloud host deployment method for each business type from the multiple candidate options in the edit box.

[0153] S803 determines the cloud server deployment method for different business types.

[0154] After the user finishes editing, the cloud server deployment method edited by the user for each business type will be determined as the cloud server deployment method for each business type.

[0155] The cloud server scheduling method provided in this application embodiment displays a deployment method configuration interface, thereby obtaining the cloud server deployment methods edited by the user for each business type in the deployment method configuration interface. Then, the edited cloud server deployment methods are determined as the cloud server deployment methods for different business types. In this method, by displaying the deployment method configuration interface when obtaining the cloud server deployment methods, users can flexibly configure the cloud server deployment methods for each business type in the deployment method configuration interface. On the one hand, this visual configuration is easy to understand and adjust, lowers the technical threshold, and improves configuration flexibility; on the other hand, this interface-based configuration is traceable and highly maintainable.

[0156] Based on the above embodiments, an embodiment of the process for detecting whether the heat dissipation equipment is faulty will be provided for explanation.

[0157] In one exemplary embodiment, such as Figure 10 As shown, the detection of whether the heat dissipation equipment corresponding to each target computing node is faulty includes:

[0158] S901, for any heat dissipation device, perform fault detection on the heat dissipation device and the power supply equipment to which the heat dissipation device belongs.

[0159] In this embodiment of the application, if the heat dissipation device fails or the power supply equipment to which the heat dissipation device belongs fails, the heat dissipation device is considered to be faulty.

[0160] Therefore, when detecting whether a heat dissipation device is faulty, fault detection is performed on both the heat dissipation device and its power supply equipment. This fault detection can be performed using separate detection tools, or it can be performed using a detection module or device within a cloud computing system; this application does not limit the specific method used.

[0161] S902, if at least one of the heat dissipation equipment and the power supply equipment is faulty, then the heat dissipation equipment is determined to be faulty; if neither the heat dissipation equipment nor the power supply equipment is faulty, then the heat dissipation equipment is determined to be faultless.

[0162] After obtaining the fault detection results of the heat dissipation equipment and power supply equipment, determine whether there is a fault in the heat dissipation equipment and power supply equipment.

[0163] If the fault detection results indicate that at least one of the heat dissipation equipment and the power supply equipment is faulty, then the heat dissipation equipment is determined to be faulty; if neither the heat dissipation equipment nor the power supply equipment is faulty, then the heat dissipation equipment is determined to be faultless.

[0164] In the cloud server scheduling method provided in this application embodiment, for any heat dissipation device, fault detection is performed on both the heat dissipation device and its associated power supply device. If at least one of the heat dissipation device and the power supply device is faulty, then the heat dissipation device is determined to be faulty; if neither the heat dissipation device nor the power supply device is faulty, then the heat dissipation device is determined not to be faulty. This method, by performing fault detection on both the heat dissipation device and its associated power supply device, determines whether the heat dissipation device is faulty based on the states of both. Essentially, the states of both the heat dissipation device and its associated power supply device affect the determination of the heat dissipation device's fault status. By performing fault detection on both, the accuracy of heat dissipation device fault detection is improved.

[0165] Based on the above embodiments, an embodiment of the above-described process of scheduling cloud hosts will be provided for illustration.

[0166] In one exemplary embodiment, such as Figure 11 As shown, in the event of a failure in any of the heat dissipation devices, the cloud host is scheduled based on other computing nodes in the target computing node group to perform business processing, including:

[0167] S1001, determine the number of cloud hosts that need to be scheduled based on the faulty heat dissipation equipment.

[0168] The number of faulty heat dissipation devices corresponding to each target computing node may be one or more. When scheduling cloud hosts, it is necessary to determine the number of cloud hosts to be scheduled based on the faulty heat dissipation devices.

[0169] First, identify which cooling devices are malfunctioning, then determine how many cloud hosts are on each malfunctioning cooling device, thus obtaining the number of cloud hosts that need to be scheduled. In other words, the total number of cloud hosts on each malfunctioning cooling device is the number of cloud hosts that need to be scheduled.

[0170] S1002: Select any computing node from other computing nodes that matches the number of cloud hosts to be scheduled and is on a normal heat dissipation device, and schedule the cloud hosts to be scheduled to the selected computing node for business processing.

[0171] After determining the number of cloud hosts to be scheduled, select any computing node from the other computing nodes in the target computing node group that matches the number of cloud hosts to be scheduled, and perform cloud host scheduling. The matching computing node must be a computing node on a normal heat dissipation device so that cloud hosts can be scheduled to process services on these computing nodes.

[0172] It should be noted that the monitoring and management system in this application embodiment can also monitor the status of each deployed cloud host in real time. If a cloud host is detected to be faulty or in an abnormal state, the faulty cloud host can be redeployed on other computing nodes in the group to continue executing the business that the faulty cloud host has not completed, thereby improving the reliability of the cloud host.

[0173] The cloud server scheduling method provided in this application first determines the number of cloud servers to be scheduled based on the faulty cooling equipment. Then, it arbitrarily selects a computing node from other computing nodes that matches the number of cloud servers to be scheduled and is on a working cooling device. The cloud servers to be scheduled are then scheduled to the selected computing node for business processing. In this method, when scheduling cloud servers, the system determines which cloud servers need to be scheduled based on the faulty cooling equipment. Then, it selects computing nodes from other computing nodes that can support cloud server scheduling and schedules the cloud servers to the selected computing nodes to continue processing business on the cloud servers, thereby improving the reliability of cloud server business processing.

[0174] In addition, in an exemplary embodiment, the process of cloud host deployment and scheduling in a high-density liquid-cooled server environment in this application embodiment will be described.

[0175] (1) Cloud deployment module: A cloud computing system is established in a high-density liquid-cooled server environment. High-density liquid cooling technology can effectively reduce the operating temperature of the server and improve its operating efficiency and reliability. Therefore, this embodiment establishes a cloud computing system in a high-density liquid-cooled server environment to improve the system's performance and stability.

[0176] (2) Redundancy grouping module: Based on the requirements of the cloud computing system, the computing nodes are grouped according to their redundancy level. During the grouping process, factors such as node location, heat dissipation equipment, power supply, and business load can be considered to ensure that the computing nodes in each group have similar characteristics and performance.

[0177] (3) Redundancy strategy module: In each group, multiple computing nodes are deployed redundantly to improve the availability and reliability of the system. Specifically, based on the different characteristics of the business system, redundancy mechanisms such as master-slave, dual-machine hot standby, or multi-machine hot standby can be adopted according to the redundancy group to ensure that if a computing node fails, the system can automatically switch to other nodes, thereby avoiding system downtime.

[0178] (4) Network connection module: This module connects multiple packets through the network to form a highly available cloud computing system. During the connection process, technologies such as multi-path redundancy and load balancing can be used to improve network bandwidth and reliability. At the same time, virtualization technology can be used to virtualize different computing nodes into a single logical node for easier management and maintenance.

[0179] (5) Monitoring and management module: Through the monitoring and management system, the status of servers and cloud hosts is monitored in real time, faults are detected and dealt with in a timely manner, and the stability and reliability of services are guaranteed.

[0180] (6) System scheduling module: Based on the above (5) steps, the system schedules the task execution order in a reasonable manner according to the redundancy strategy mechanism, minimizes the waste of computing resources, and automatically schedules virtual machines (cloud hosts) to be distributed to different redundancy groups and redundancy nodes.

[0181] In this embodiment, high-density liquid cooling technology can effectively improve the operating efficiency and reliability of the server; grouping and redundant deployment can improve the availability and fault tolerance of the system; and connecting multiple groups through a network to form a highly available cloud computing system can improve the reliability and performance of the system.

[0182] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0183] Based on the same inventive concept, this application also provides a cloud server scheduling apparatus for implementing the cloud server scheduling method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more cloud server scheduling apparatus embodiments provided below can be found in the limitations of the cloud server scheduling method described above, and will not be repeated here.

[0184] In one exemplary embodiment, such as Figure 12 As shown, a cloud server scheduling device 1 is provided, comprising: a node group determination module 10, a fault detection module 20, and a cloud server scheduling module 30, wherein:

[0185] The node group determination module 10 is used to determine the target computing node group of the service to be processed from multiple candidate computing node groups according to the service type of the service to be processed; each computing node in each candidate computing node group is distributed at the location of different heat dissipation devices in a high-density liquid cooling environment;

[0186] The fault detection module 20 is used to determine at least one target computing node in the target computing node group that has a cloud host deployed, and to detect whether the heat dissipation equipment corresponding to each target computing node is faulty.

[0187] The cloud host scheduling module 30 is used to schedule cloud hosts for business processing based on other computing nodes in the target computing node group when each heat dissipation device fails.

[0188] In one embodiment, the cloud host scheduling device 1 further includes:

[0189] The grouping information acquisition module is used to acquire multiple redundancy levels of computing nodes in the cloud computing system, as well as the redundancy grouping factor in a high-density liquid cooling environment. The redundancy grouping factor includes at least the location of each heat dissipation device in the high-density liquid cooling environment and the power supply equipment to which each heat dissipation device belongs.

[0190] The node grouping module is used to group the computing nodes in the cloud computing system according to each redundancy level and each redundancy grouping factor to obtain multiple candidate computing node groups.

[0191] In one embodiment, the node grouping module is further configured to:

[0192] Based on each redundancy level, obtain the redundancy strategy for each redundancy level; the redundancy strategy represents the number of redundant computing nodes corresponding to each redundancy level.

[0193] For each redundancy level, based on the location of each heat dissipation device and the power supply device to which each device belongs, multiple computing nodes that meet the redundancy strategy of each redundancy level and are located at different heat dissipation devices and different power supply devices are selected from each computing node as the computing nodes corresponding to each redundancy level.

[0194] Each redundancy level corresponds to a computing node as a candidate computing node group, resulting in multiple candidate computing node groups.

[0195] In one embodiment, the cloud host scheduling device 1 further includes:

[0196] The deployment method acquisition module is used to obtain the cloud server deployment methods for different business types; cloud server deployment methods include at least master-slave, dual-machine hot standby, and multi-machine hot standby.

[0197] The cloud server deployment module is used to deploy cloud servers according to the deployment methods of each cloud server and multiple candidate computing node groups.

[0198] In one embodiment, the cloud host deployment module is further configured to:

[0199] For any candidate computing node group, connect each computing node in the candidate computing node group to form a network group of candidate computing nodes.

[0200] According to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to complete the cloud host deployment of the candidate computing node group.

[0201] In one embodiment, the cloud host deployment module is further configured to:

[0202] The status of each computing node is monitored to obtain its status information. If the status information determines that each computing node is in a normal state, the computing nodes in the candidate computing node network group that match the cloud host deployment method are determined according to the cloud host deployment method. The cloud hosts in the candidate computing node group are then deployed on the computing nodes that match the cloud host deployment method.

[0203] In one embodiment, the deployment method acquisition module is further used for:

[0204] Displays the deployment method configuration interface;

[0205] Obtain the cloud server deployment methods edited by users for each business type in the deployment method configuration interface;

[0206] The edited cloud server deployment method is determined to be the cloud server deployment method for different business types.

[0207] In one embodiment, the fault detection module 20 is further configured to:

[0208] For any heat dissipation device, perform fault detection on both the heat dissipation device and the power supply equipment to which it belongs;

[0209] If at least one of the heat dissipation equipment and the power supply equipment is faulty, then the heat dissipation equipment is determined to be faulty.

[0210] If both the heat dissipation equipment and the power supply equipment are functioning correctly, then the heat dissipation equipment is confirmed to be functioning correctly.

[0211] In one embodiment, the cloud host scheduling module 30 is further configured to:

[0212] Based on the faulty cooling equipment, determine the number of cloud hosts that need to be scheduled; randomly select a computing node from other computing nodes that matches the number of cloud hosts that need to be scheduled and is on a normal cooling equipment, and schedule the cloud hosts that need to be scheduled to the selected computing node for business processing.

[0213] Each module in the aforementioned cloud server scheduling device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0214] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0215] Based on the service type of the service to be processed, the target computing node group for the service to be processed is determined from multiple candidate computing node groups; the computing nodes in each candidate computing node group are distributed and deployed at the locations of different heat dissipation devices in a high-density liquid cooling environment;

[0216] Identify at least one target computing node in the target computing node group that has a cloud host deployed, and detect whether the heat dissipation equipment corresponding to each target computing node is faulty;

[0217] In the event of a malfunction in any of the heat dissipation devices, the cloud host is scheduled to perform business processing based on other computing nodes in the target computing node group.

[0218] The implementation principles and technical effects of each step in the processor embodiment of this application are similar to those of the cloud host scheduling method described above, and will not be repeated here.

[0219] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0220] Based on the service type of the service to be processed, the target computing node group for the service to be processed is determined from multiple candidate computing node groups; the computing nodes in each candidate computing node group are distributed and deployed at the locations of different heat dissipation devices in a high-density liquid cooling environment;

[0221] Identify at least one target computing node in the target computing node group that has a cloud host deployed, and detect whether the heat dissipation equipment corresponding to each target computing node is faulty;

[0222] In the event of a malfunction in any of the heat dissipation devices, the cloud host is scheduled to perform business processing based on other computing nodes in the target computing node group.

[0223] The implementation principles and technical effects of each step carried out by the computer program when it is executed by the processor in this embodiment are similar to those of the cloud host scheduling method described above, and will not be repeated here.

[0224] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0225] Based on the service type of the service to be processed, the target computing node group for the service to be processed is determined from multiple candidate computing node groups; the computing nodes in each candidate computing node group are distributed and deployed at the locations of different heat dissipation devices in a high-density liquid cooling environment;

[0226] Identify at least one target computing node in the target computing node group that has a cloud host deployed, and detect whether the heat dissipation equipment corresponding to each target computing node is faulty;

[0227] In the event of a malfunction in any of the heat dissipation devices, the cloud host is scheduled to perform business processing based on other computing nodes in the target computing node group.

[0228] The implementation principles and technical effects of each step carried out by the computer program when it is executed by the processor in this embodiment are similar to those of the cloud host scheduling method described above, and will not be repeated here.

[0229] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0230] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0231] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0232] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A cloud host scheduling method, characterized in that, A controller for a cloud computing system operating in a high-density liquid-cooled environment, the method comprising: Based on the service type of the service to be processed, the target computing node group for the service to be processed is determined from multiple candidate computing node groups; each computing node in each candidate computing node group is distributed and deployed at the location of different heat dissipation devices in the high-density liquid cooling environment; Identify at least one target computing node in the target computing node group that has a cloud host deployed, and detect whether the heat dissipation equipment corresponding to each target computing node is faulty; In the event of a malfunction in any of the aforementioned heat dissipation devices, the cloud host is scheduled to perform business processing based on other computing nodes in the target computing node group.

2. The method of claim 1, wherein, Before determining the target computing node group for the pending service from multiple candidate computing node groups based on the service type of the pending service, the method further includes: Obtain multiple redundancy levels of computing nodes in the cloud computing system, as well as redundancy grouping factors in the high-density liquid cooling environment; the redundancy grouping factors include at least the location of each heat dissipation device in the high-density liquid cooling environment and the power supply device to which each heat dissipation device belongs; Based on the redundancy levels and redundancy grouping factors, the computing nodes in the cloud computing system are grouped to obtain the multiple candidate computing node groups.

3. The method according to claim 2, characterized in that, The step of grouping the computing nodes in the cloud computing system according to each of the redundancy levels and each of the redundancy grouping factors to obtain the plurality of candidate computing node groups includes: Based on each of the redundancy levels, a redundancy strategy for each redundancy level is obtained; the redundancy strategy represents the number of redundant computing nodes corresponding to each redundancy level. For each redundancy level, based on the location of each heat dissipation device and the power supply device to which each device belongs, multiple computing nodes that conform to the redundancy strategy of each redundancy level and are located at different heat dissipation devices and different power supply devices are selected from each computing node as the computing nodes corresponding to each redundancy level. Each redundancy level corresponds to a computing node as a candidate computing node group, resulting in the multiple candidate computing node groups.

4. The method according to any one of claims 1-3, characterized in that, Before determining the target computing node group for the service to be processed from multiple candidate computing node groups based on the service type of the service to be processed, the method further includes: Obtain cloud server deployment methods for different business types; the cloud server deployment methods include at least master-slave, dual-machine hot standby, and multi-machine hot standby. Cloud host deployment is performed according to the respective cloud host deployment methods and the multiple candidate computing node groups.

5. The method according to claim 4, characterized in that, The step of deploying cloud hosts according to each of the cloud host deployment methods and the multiple candidate computing node groups includes: For any candidate computing node group, each computing node in the candidate computing node group is connected to the network to obtain a candidate computing node network group. According to the cloud host deployment method, cloud hosts are deployed on each computing node in the candidate computing node network group to complete the cloud host deployment of the candidate computing node group.

6. The method according to claim 5, characterized in that, The step of deploying cloud hosts on each computing node in the candidate computing node network group according to the cloud host deployment method, to complete the cloud host deployment of the candidate computing node group, includes: The status of each computing node is monitored to obtain the status information of each computing node; If it is determined that all computing nodes are in a normal state based on the status information, the computing nodes in the candidate computing node network group that match the cloud host deployment method are determined according to the cloud host deployment method, and the cloud hosts of the candidate computing node group are deployed on the computing nodes that match the cloud host deployment method.

7. The method according to claim 4, characterized in that, The methods for obtaining cloud server deployment methods for different business types include: Displays the deployment method configuration interface; Obtain the cloud server deployment method edited by the user for each of the aforementioned business types in the deployment method configuration interface; The edited cloud server deployment method is determined as the cloud server deployment method for the different business types.

8. The method according to any one of claims 1-3, characterized in that, The detection of whether the heat dissipation equipment corresponding to each of the target computing nodes is faulty includes: For any heat dissipation device, fault detection is performed on both the heat dissipation device and the power supply equipment to which the heat dissipation device belongs; If at least one of the heat dissipation device and the power supply device is faulty, then the heat dissipation device is determined to be faulty. If neither the heat dissipation device nor the power supply device is faulty, then it is determined that the heat dissipation device is not faulty.

9. The method according to any one of claims 1-3, characterized in that, In the event of a malfunction in any of the aforementioned heat dissipation devices, the cloud host is scheduled based on other computing nodes in the target computing node group for business processing, including: Determine the number of cloud hosts that need to be scheduled based on the faulty heat dissipation equipment; Select any computing node from the other computing nodes that matches the number of cloud hosts to be scheduled and is on a normal cooling device, and schedule the cloud hosts to be scheduled to the selected computing node for business processing.

10. A cloud server scheduling device, characterized in that, The device includes: The node group determination module is used to determine the target computing node group of the service to be processed from multiple candidate computing node groups according to the service type of the service to be processed; each computing node in each candidate computing node group is distributed at the location of different heat dissipation devices in a high-density liquid cooling environment; The fault detection module is used to determine at least one target computing node in the target computing node group that has a cloud host deployed, and to detect whether the heat dissipation equipment corresponding to each target computing node is faulty. The cloud server scheduling module is used to schedule the cloud server according to other computing nodes in the target computing node group to perform business processing when each of the heat dissipation devices fails.

Images