Method and apparatus for controlling server cluster

Abstract

The application discloses a kind of control method and device of server cluster, it is related to server technical field, method includes: obtaining the target address information of server cluster, wherein, target address information is used to indicate the address of multiple target servers in server cluster in local area network, multiple target servers include multiple servers, controller is in local area network;Obtain the port information of each target server from server cluster by target address information, obtain multiple target port information;Determine the server interface information of server cluster according to multiple target port information and port information set;According to server interface information, control each target server in server cluster responds to the operation request to server cluster, by the present application, it solves the technical problem that the control efficiency of server cluster is lower, reaches the technical effect of improving the control efficiency of server cluster.

Description

Technical Field

[0001] This application relates to the field of server technology, and in particular to a control method and apparatus for a server cluster. Background Technology

[0002] With the continuous advancement of enterprise informatization, server cluster management and control solutions are attracting increasing attention. Currently, for the various types of servers within a server cluster, management tools corresponding to each server type are generally used to manage the respective servers. This fragmented management approach leads to significant difficulties in managing and controlling the server cluster, resulting in low control efficiency. Summary of the Invention

[0003] This application provides a control method and apparatus for a server cluster, which at least solves the problem of low control efficiency of server clusters in related technologies.

[0004] This application provides a control method for a server cluster, applied to a controller of the server cluster. The method includes: obtaining target address information of the server cluster, wherein the target address information indicates the addresses of multiple target servers in the server cluster within a local area network (LAN), the multiple target servers including various types of servers, and the controller being located within the LAN; obtaining port information of each target server from the server cluster using the target address information, resulting in multiple target port information, wherein the target port information indicates the open status of each server port of the target servers; determining server interface information of the server cluster based on the multiple target port information and a set of port information, wherein the server interface information indicates the out-of-band management interfaces used by various servers in the server cluster, and the set of port information records the correspondence between server port information and the out-of-band management interfaces used by the servers; and controlling each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information, wherein the operation request requests the servers in the server cluster to perform operations.

[0005] This application also provides a control device for a server cluster, applied to a controller of a server cluster, the device comprising:

[0006] The first acquisition module is used to acquire the target address information of the server cluster. The target address information is used to indicate the addresses of multiple target servers in the server cluster in the local area network. The multiple target servers include various types of servers, and the controller is in the local area network.

[0007] The second acquisition module is used to obtain the port information of each target server from the server cluster through the target address information, and obtain multiple target port information, wherein the target port information is used to indicate the open status of each server port of the target server.

[0008] The determination module is used to determine the server interface information of the server cluster based on multiple target port information and a set of port information. The server interface information is used to indicate the out-of-band management interface used by various servers in the server cluster, and the set of port information records the correspondence between the server's port information and the out-of-band management interface used by the server.

[0009] The control module is used to control each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information. The operation request is used to request the servers in the server cluster to perform an operation.

[0010] This application also provides an electronic device, including: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of any of the above-described server cluster control methods.

[0011] This application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of any of the above-described server cluster control methods.

[0012] Through this application, the controller of the server cluster obtains the target address information of the server cluster, and obtains the port information of each target server in the server cluster through the target address information, thus obtaining multiple target port information. Based on the multiple target port information and the set of port information, the controller determines the server interface information of the server cluster. That is, the controller of the server cluster can ultimately obtain the server interface information used to indicate the out-of-band management interface used by various servers in the server cluster by obtaining the target address information. The controller controls each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information, so that the management and control of the server cluster can be realized through the controller as a single control tool, reducing the difficulty of controlling the server cluster and improving the control efficiency of the server cluster. Therefore, it can solve the technical problem of low control efficiency of server clusters and achieve the technical effect of improving the control efficiency of server clusters. Attached Figure Description

[0013] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the 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.

[0014] Figure 1 This is a block diagram of the hardware structure for controlling a server cluster according to an embodiment of this application;

[0015] Figure 2 This is a flowchart of a server cluster control method according to an embodiment of this application;

[0016] Figure 3 This is a schematic diagram of the operation of a controller according to an embodiment of this application;

[0017] Figure 4 This is a structural block diagram of a server cluster control device according to an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0019] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0020] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] The specific application environment architecture or specific hardware architecture on which the execution of the control method for the server cluster depends is described here.

[0022] The methods and embodiments provided in this application can be executed on a server device or a similar computing device. Taking running on a server device as an example, Figure 1 This is a hardware structure block diagram of the server cluster control according to an embodiment of this application. For example... Figure 1 As shown, the server device may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The server device may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the server equipment described above. For example, the server equipment may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0023] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the server cluster control method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thus implementing the aforementioned method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the server device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0024] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider for the server device. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0025] The embodiments of this application provide a control method for a server cluster. The method is described in detail below in conjunction with the execution flow of the control method for a server cluster.

[0026] This embodiment provides a method for controlling a server cluster. Figure 2 This is a flowchart of a server cluster control method according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0027] Step S202: Obtain the target address information of the server cluster. The target address information is used to indicate the addresses of multiple target servers in the server cluster in the local area network. The multiple target servers include various types of servers, and the controller is in the local area network.

[0028] Step S204: Obtain the port information of each target server from the server cluster through the target address information to obtain multiple target port information, wherein the target port information is used to indicate the open status of each server port of the target server.

[0029] Step S206: Determine the server interface information of the server cluster based on multiple target port information and port information set. The server interface information is used to indicate the out-of-band management interface used by various servers in the server cluster. The port information set records the correspondence between the server's port information and the out-of-band management interface used by the server.

[0030] Step S208: Control each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information, wherein the operation request is used to request the servers in the server cluster to perform an operation.

[0031] Through the above steps, the server cluster controller obtains the target address information of the server cluster, and then obtains the port information of each target server in the server cluster using the target address information, resulting in multiple target port information. Based on the multiple target port information and the set of port information, the server interface information of the server cluster is determined. That is, the server cluster controller can ultimately obtain the server interface information used to indicate the out-of-band management interface used by various servers in the server cluster by obtaining the target address information. The controller controls each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information, so that the management and control of the server cluster can be achieved through the controller as a single control tool. This reduces the difficulty of controlling the server cluster and improves the control efficiency of the server cluster. Therefore, it can solve the technical problem of low control efficiency of server clusters and achieve the technical effect of improving the control efficiency of server clusters.

[0032] Optionally, in this embodiment, the control method for the server cluster may be, but is not limited to, applied to the controller of the server cluster.

[0033] Optionally, in this embodiment, the server cluster may be used, but is not limited to, to form a data center.

[0034] Optionally, in this embodiment, the controller may be, but is not limited to, a core component of server cluster management. The controller may be, but is not limited to, capable of coordinating and controlling multiple servers within the cluster to achieve functions such as centralized management, automated operation, resource optimization, and fault handling. Specifically, the controller can be a dedicated hardware device, such as a BMC (Baseboard Management Controller). Alternatively, the controller can exist in software form, running on a central server or cloud-based server management software, such as a server management platform.

[0035] In the embodiment provided in step S202, the server cluster may include, but is not limited to, multiple servers, and these multiple servers may belong to, but are not limited to, various types of servers. For example, the server cluster may include, but is not limited to, server 1, server 2, server 3, and server 4, wherein server 1 and server 2 are type A servers and correspond to type A out-of-band management interfaces, server 3 is type B servers and corresponds to type B out-of-band management interfaces, and server 4 is type C servers and corresponds to type C out-of-band management interfaces.

[0036] Optionally, in this embodiment, the server cluster and the controller may, but are not limited to, be connected in a local area network.

[0037] Optionally, in this embodiment, the target address information may, but is not limited to, indicating the addresses of multiple target servers in the server cluster within the local area network (LAN). The target address information may, but is not limited to, include multiple pairs of corresponding network addresses and physical addresses. The network addresses may, but are not limited to, network addresses within the LAN. Specifically, network addresses may, but are not limited to, IP addresses (Internet Protocol Addresses). Physical addresses may, but are not limited to, MAC addresses (Media Access Control Addresses).

[0038] Optionally, in this embodiment, obtaining the target address information of the server cluster may include, but is not limited to: using the ARP (Address Resolution Protocol) protocol to probe devices within the local area network, collecting the IP addresses and MAC addresses of all active devices, and filtering out server devices from them; or using the ICMP (Internet Control Message Protocol) Ping command (a network tool that uses the ICMP protocol to test whether a network connection is reachable) to probe every IP address within the local area network and record the IP address of the responding server.

[0039] In the embodiment provided in step S204, the port information may be used, but is not limited to, to indicate the open status of the device's ports, and the target port information may be used, but is not limited to, to indicate the open status of each server port of the target server.

[0040] Optionally, in this embodiment, the target port information may, but is not limited to, specifically refer to the open status and related attributes of a specific port or all ports on the target server. The target port information can help the controller determine which ports can be used for out-of-band management and which ports provide specific services. The target port information may, but is not limited to, include: port number, port open status (indicating whether the target port is currently receiving and processing data. If the port is open, it means that communication can be established with the server through that port; if the port is closed, it may be due to the service not running, firewall blocking, or network configuration problems), and the port's service name or application type (the specific service or application type running on the port, such as "HTTP (Hypertext Transfer Protocol)," "SSH (Secure Shell)," or "IPMI (Intelligent Platform Management Interface)").

[0041] Optionally, in this embodiment, the port information of each target server in the server cluster can be obtained from the target address information in the following ways, but not limited to: using a network scanning tool to perform port scanning on the IP address of each target server to obtain a list of open ports; or, attempting to connect to a known port that may be bound to an out-of-band management interface (such as the default port 623 of IPMI) through a TCP (Transmission Control Protocol) three-way handshake to determine whether the out-of-band management interface is available; or using a pre-set server model database to query the model of the target server to find out its possible open management ports and verify them one by one.

[0042] In the embodiment provided in step S206, the port information set may, but is not limited to, record the correspondence between the server's port information and the out-of-band management interface used by the server. The port information set may, but is not limited to, be a database or data structure that stores the correspondence between the out-of-band management interface information of a specific server brand and model and their respective default or common ports. The port information set may include, but is not limited to: server brand and model, out-of-band management interface type, default or common port (e.g., port 623 corresponding to out-of-band management interface type A, port 443 corresponding to out-of-band management interface type B, port 80 or 443 corresponding to out-of-band management interface type C, etc.). The construction of the port information set may, but is not limited to, originate from manufacturer manuals, online documentation, community knowledge, or be obtained through statistical analysis after automatically scanning a large number of servers.

[0043] Optionally, in this embodiment, the server interface information may be, but is not limited to, used to indicate the out-of-band management interfaces used by various servers in the server cluster. The server interface information may be, but is not limited to, the type and configuration details of the out-of-band management interface actually used by the server, determined by analyzing the open port information of the target server based on a set of port information. The server interface information may include, but is not limited to, the interface type and port number, and may also include, but is not limited to, the interface version, supported functions, and parameters required to access and control the interface.

[0044] Optionally, in this embodiment, determining the server interface information of the server cluster based on multiple target port information and port information sets can be achieved, but is not limited to, through the following methods: First, collect a large number of port usage logs. These logs may include, but are not limited to: IP addresses and their corresponding MAC addresses; port open and closed states at different times; data packet characteristics of communication data packets transmitted through these ports; data packet size, frequency, source port, and destination port, and whether they contain specific authentication or command information; server hardware information, manufacturer, model, and BIOS (Basic Input / Output System) version. The port usage logs can be collected, but are not limited to, using network sniffing tools. Next, perform data preprocessing and feature extraction. Data preprocessing may include, but is not limited to: deleting irrelevant or erroneous records, such as communication records from non-server devices; converting the data into a unified format for easier subsequent processing; and identifying and handling abnormal communication patterns, such as abnormally large data packets or uncommon protocols. Feature extraction involves extracting information from data packets that represents the communication patterns of a specific out-of-band management interface, such as: port number (a specific port, such as IPMI's 623), protocol features (values ​​of specific fields in the packet header), authentication and command information (whether specific authentication information or command requests exist), time series features (changes in the port's open status at different points in time), and the time series of communication data packets. Next, model training is performed: using the collected data and extracted features, an appropriate machine learning algorithm can be selected for training to identify different communication patterns of out-of-band management interfaces. The specific steps of model training include: dataset partitioning: dividing the collected data into training, validation, and test sets; model training: training the model using the training set data and adjusting model parameters to optimize performance; model validation: using the validation set data to conduct preliminary performance tests on the model and adjusting the model to reduce overfitting or underfitting; final testing: using the test set data to evaluate the model's performance on unknown data to ensure the model's generalization ability. Once the model training is complete and its accuracy and stability are verified, the model can be deployed to the controller of the server cluster for real-time or batch prediction of the out-of-band management interface type of new servers. Specific deployment steps may include, but are not limited to: when the controller detects a new server's IP address, immediately collecting the server's port communication data and using a model to predict its possible out-of-band management interface type. Regularly run the prediction model to evaluate the out-of-band management interface types of all servers in the local area network to update the database. Analyze the model's prediction results to check their reasonableness, and manually correct or retrain the model if necessary. Through these steps, a controller capable of automatically identifying and predicting server interface information for a server cluster can be built, reducing manual configuration workload and improving automation and intelligence.

[0045] In the embodiment provided in step S208, the operation request may be, but is not limited to, requesting a server in the server cluster to perform an operation. Specifically, the operation request may, but is not limited to, instructing a specific server to perform an operation, or the operation request may, but is not limited to, instructing any server in the server cluster to perform an operation. For example, the operation request may, but is not limited to, requesting server 1 in the server cluster to restart, or the operation request may, but is not limited to, requesting a server in the server cluster to reduce its operating frequency to save power.

[0046] Optionally, in this embodiment, if the operation request indicates the execution server of the operation, an out-of-band management interface protocol can be matched for the execution server of the operation based on the server interface information, and an operation instruction conforming to the out-of-band management interface protocol can be generated in combination with the operation request. The operation instruction can then be sent to the execution server of the operation through the out-of-band management interface of the execution server of the operation.

[0047] Optionally, in this embodiment, even when the operation request does not specify a particular execution server, the operation entity server can be selected from multiple target servers based on server information. This server information can, but is not limited to, indicate the server's operation execution busy level. After determining the operation entity server, the response method for operation requests that specify the execution server is implemented, controlling the operation entity server to respond to operation requests to the server cluster.

[0048] Optionally, in this embodiment, users can perform one-click operations on each target server in the server cluster through the management interface provided by the controller, such as one-click power on / off and one-click serial port printing.

[0049] Optionally, in this embodiment, the operation request can be generated by the controller, in addition to those input by the user through the management interface. For example, the controller may, but is not limited to, have a mechanism for periodically retrieving serial port logs, and may, but is not limited to, automatically generate an operation request to instruct the retrieval of serial port logs through the out-of-band management interface when the retrieval period arrives. Alternatively, the controller may, but is not limited to, have a business task allocation mechanism, whereby the business tasks received by the server cluster may, but are not limited to, be allocated by the controller to various target servers for execution. Before allocation, the controller may, but is not limited to, generate an operation request to request the acquisition of hardware operating metrics for each target server. The controller combines the information obtained through in-band connections and the hardware operating metrics obtained by controlling the target servers in the server cluster to respond to the operation requests through server interface information to assess the feasibility of each target server receiving the business task.

[0050] As an optional implementation, obtaining target address information of the server cluster includes: broadcasting an address acquisition request within the local area network (LAN), wherein the address acquisition request is used to request devices possessing reference network addresses from a network address set to return the reference physical address corresponding to the reference network address. The network address set includes the network addresses of multiple devices in the LAN, the multiple devices include multiple target servers, and the reference physical address is the physical address of the device; receiving the reference physical addresses returned by each device to obtain multiple pairs of corresponding reference network addresses and reference physical addresses; and filtering multiple target address pairs from the multiple pairs of corresponding reference network addresses and reference physical addresses, wherein the target address pair is a target network address and a target physical address corresponding to a target server, the target network address is the network address of the target server in the LAN, and the target physical address is the physical address of the target server, and the target address information includes multiple target address pairs.

[0051] Optionally, in this embodiment, the reference network address and reference physical address of devices within the local area network (LAN) can be obtained by broadcasting an address acquisition request within the LAN. Specifically, broadcasting an address acquisition request within the LAN may include, but is not limited to, broadcasting an ARP request within the LAN, requesting all devices to return the MAC address corresponding to their IP address.

[0052] Optionally, in this embodiment, the received multiple pairs of corresponding reference network addresses and reference physical addresses may include, but are not limited to, the reference network addresses and reference physical addresses of all devices in the local area network. The target network addresses and target physical addresses corresponding to the target server may be filtered out from these pairs to obtain multiple target address pairs, which are the aforementioned target address information.

[0053] By broadcasting ARP requests and receiving responses, all devices within the local area network (LAN), including servers in a server cluster, can be dynamically discovered. This allows for accurate identification and management of these servers even when their IP or MAC addresses change, improving the flexibility and reliability of server cluster management and control.

[0054] As an optional implementation, multiple target address pairs are selected from multiple pairs of corresponding reference network addresses and reference physical addresses, including: connecting various devices through multiple pairs of corresponding reference network addresses and reference physical addresses; obtaining device information of each device from each device, wherein the device information is used to indicate the device type of the device; and selecting multiple target address pairs from multiple pairs of corresponding reference network addresses and reference physical addresses based on the device information.

[0055] Optionally, in this embodiment, various devices within the local area network can be connected through multiple pairs of corresponding reference network addresses and reference physical addresses. By connecting these devices, device information can be obtained from each device to determine which devices are server devices and which are not. Then, the corresponding reference network addresses and reference physical addresses of the server devices can be identified as target address pairs.

[0056] Optionally, in this embodiment, the IP address and MAC address information in the ARP response can be used, but is not limited to, to attempt to establish connections with all devices in the network. For example, an attempt can be made to connect to the server's SSH service via the TCP / IP protocol. After establishing a connection, an SNMP GET request (used to collect information about network devices) is sent to obtain the device's sysObjectID (device model) and sysDescr (system description) information, or, after logging in via SSH, commands such as lshw (used to list computer hardware configuration) and dmidecode (used to decode information in the DMI (Desktop Management Interface) table) can be used to obtain a detailed description of the hardware. This information can help determine the device type. Based on the returned device information, it is determined which devices are servers, and their IP addresses and MAC addresses are combined to form target address pairs for subsequent centralized management of the server cluster.

[0057] After filtering out the IP-MAC address pairs of the target servers using the above methods, a centralized device list or database can be built for subsequent unified management and control. This not only simplifies management operations but also enables more granular resource scheduling and fault recovery based on the actual status and location of the servers.

[0058] As an optional implementation, port information of each target server is obtained from the server cluster through target address information, resulting in multiple target port information. This includes: sending standard data packets corresponding to each server port of each target server through multiple target address pairs, wherein each target address pair is a target network address and a target physical address that have a corresponding relationship with the target server, the target network address is the network address of the target server in the local area network, and the target physical address is the physical address of the target server, and the target address information includes multiple target address pairs; receiving response data packets from each target server within a target time period after sending the standard data packets, wherein the response data packets are data packets returned by each target server in response to the received standard data packets; if a response data packet is received within the target time period, determining that the port status of the server port corresponding to the response data packet is open, wherein the multiple target port information includes multiple port statuses; if no response data packet is received within the target time period, determining that the port status of the server port corresponding to the response data packet is closed.

[0059] Optionally, in this embodiment, standard data packets corresponding to each server port can be sent to each server port of each target server through multiple target addresses, such as TCP SYN packets, UDP (User Datagram Protocol) packets, or specific HTTP (Hypertext Transfer Protocol) requests. The purpose of sending standard data packets may be, but is not limited to, testing whether the port responds, thereby determining its status.

[0060] Optionally, in this embodiment, a target time period (such as 1 second, 3 seconds, etc.) can be set after sending the standard data packet to wait for the response data packet from the target server. The response data packet can be a TCP ACK (Acknowledgment) response, a UDP response, or an HTTP status code response, etc., the specific form of which depends on the type of standard data packet sent and the port function.

[0061] Optionally, in this embodiment, if a response data packet is received within the target time period, it can be determined that the current state of the server port is open, which means that the port can receive and respond to external requests and is an effective channel for management operations; if no response data packet is received within the target time period, it can be determined that the current state of the server port is closed, which indicates that the port is not enabled or there is a network problem, and it cannot be used to respond to the current operation request.

[0062] As an optional implementation, the server interface information of the server cluster is determined based on multiple target port information and a set of port information, including: searching for each out-of-band management interface corresponding to each target port status group from port status groups and out-of-band management interfaces with corresponding relationships, wherein each target port status group includes multiple target port states, each target port state is the state of the server port of the target server, and the target port information includes the target port status group; recording the out-of-band management interface corresponding to each target server found, thereby obtaining the server interface information.

[0063] Optionally, in this embodiment, the matching out-of-band management interface type can be searched in a pre-stored port information set based on the port status group (a set of information reflecting the open or closed status of each port of the server) obtained from the target server. Once an out-of-band management interface corresponding to the target port status group is found, the association information between the target server and the out-of-band management interface is recorded, including but not limited to the type of management interface, port number, and access credentials.

[0064] The above information enables accurate identification and matching of the out-of-band management interface of each target server. Precise interface matching ensures that the sent management commands can be correctly parsed and executed, avoiding management failures or anomalies caused by incorrect interface types, and laying the foundation for integrated control of the server cluster.

[0065] Optionally, in this embodiment, determining the server interface information of the server cluster may include, but is not limited to, the following steps: sending a broadcast ARP request to obtain the MAC addresses of all active devices in the local area network; sending a Ping packet to each IP address to filter out responding devices; sending an SNMP GET request to the responding device to obtain sysDescr (system description) and sysObjectID (device model); if the device is a server, matching the manufacturer's preset out-of-band management interface type according to sysObjectID (e.g., a certain model of server corresponds to iDRAC (an out-of-band management controller) port 623); requiring each target server to send a heartbeat packet (containing the current load status) to the controller every 60 seconds; if no heartbeat is received for 3 consecutive times, the target server is marked as "offline" and an alarm is triggered.

[0066] Optionally, in this embodiment, determining the server interface information of the server cluster may also include, but is not limited to, the following steps: using the ARP protocol and ICMP Ping to scan all active IP addresses (i.e., target network addresses) in the local area network, and combining the SNMP protocol (Simple Network Management Protocol) to obtain basic device information (such as MAC address (i.e., target physical address) and vendor information (i.e., device information)); identifying server types (e.g., detecting iDRAC default port 623 and IPMI port 623) through TCP / UDP port scanning (i.e., scanning to obtain multiple target port status groups), and matching them with a preset device fingerprint database (i.e., port information set); periodically (e.g., every 5 minutes) scanning for newly added or offline devices, and maintaining the device list through a heartbeat mechanism (devices periodically sending status packets); storing device information (IP, MAC, hardware configuration, service status) in a central database, supporting management by department / function.

[0067] As an optional implementation, controlling each target server in the server cluster to respond to operation requests to the server cluster based on server interface information includes: receiving a reference operation request, wherein the reference operation request is used to request a reference server in the server cluster to perform a reference operation, multiple target servers include a reference server, and the operation request includes the reference operation request; matching corresponding reference interface information for the reference server from the server interface information, wherein the reference interface information is used to indicate the reference management interface used by the reference server, and the out-of-band management interface includes the reference management interface; and controlling the reference server to respond to the reference operation request through the reference interface information.

[0068] Optionally, in this embodiment, when a reference operation request is received to request a reference server in the server cluster to perform a reference operation, corresponding reference interface information may be matched for the reference server, and the reference server may be controlled to respond to the reference operation request through the reference interface information.

[0069] Optionally, in this embodiment, the operation request received by the controller may also be, but is not limited to, a target operation request for requesting the server cluster to perform a target operation. For example, the target operation request may be, but is not limited to, an energy-saving mode switching request, that is, requesting the server cluster to enter or exit energy-saving mode in order to reduce energy consumption and operating costs. Energy-saving modes may include adjusting CPU (Central Processing Unit) frequency, fan speed, power supply policies, etc. When the controller receives a request targeting the entire server cluster, such as an energy-saving mode switching request, it may, but is not limited to, prioritize controlling some target servers to respond to the operation request first, and then control the remaining target servers to respond to the operation request later, based on the different busy levels of the target servers in the server cluster.

[0070] Optionally, in this embodiment, the operation request received by the controller may also be, but is not limited to, requesting any target server in the server cluster to perform the operation to be performed. When the controller receives such an operation request, it may, but is not limited to, select a target server to respond to the operation request based on the operating busyness of each target server in the server cluster.

[0071] Based on the above, the control strategy using interface information can be easily adapted to different types and brands of servers, eliminating the need to write dedicated control logic for each type of server. This significantly reduces the workload of maintaining the server cluster and facilitates the addition of new server types or functions in the future.

[0072] As an optional implementation, controlling the reference server to respond to reference operation requests through reference interface information includes: matching the reference operation request with corresponding reference template information from the instruction template information of the server cluster based on the reference interface information, wherein the instruction template information is used to indicate the instruction format of various out-of-band management interfaces, and the reference template information is used to indicate the instruction format of the reference management interface; generating reference operation instructions based on the reference template information and the reference operation request, wherein the reference operation instructions are used to instruct the reference server to respond to the reference operation request; and sending the reference operation instructions to the reference server through the reference management interface.

[0073] Optionally, in this embodiment, the operation request can be converted into an operation instruction according to the instruction format corresponding to the out-of-band management interface, and the operation instruction can be sent to the corresponding server through the corresponding out-of-band management interface.

[0074] Optionally, in this embodiment, the controller of the server cluster may, but is not limited to, encapsulate differentiated protocols such as IPMI, Redfish (a modern standard for data center and server management), and iDRAC, providing a unified RESTful API (Representational State Transfer Application Programming Interface) for upper-layer calls (e.g., through the ipmitool command-line tool or vendor SDKs). Users can, but are not limited to, select target servers and functions (e.g., set fan silent mode) through the controller's management interface; the controller automatically matches the corresponding out-of-band interface protocol based on the server type (i.e., matches the corresponding reference template information); and send standardized commands (e.g., ipmitool rawxxxx xxxx xxxx xxxx xxxx xxxx set the fan to low speed mode). Optionally, in this embodiment, even when the operating system crashes, operations such as power on / off and log capture can still be performed through the out-of-band interface.

[0075] Optionally, in this embodiment, the reference operation request may include, but is not limited to, requesting the reference server to set to "silent mode," and the controller may, but is not limited to, querying the device database (i.e., server interface information) to determine that the server uses the iDRAC 9.0 interface. The iDRAC Redfish API is invoked to construct an HTTP request:

[0076] POST / redfish / v1 / Chassis / System.Embedded.1 / Thermal / Actions / Thermal.SetFanSpeed

[0077] Body: { "FanSpeed": "Low"}

[0078] If the API (Application Programming Interface) returns an error code (such as 403 Insufficient Permissions), automatically switch to the IPMI backup channel to send commands:

[0079] ipmitool -I lanplus -H<BMC_IP> -U admin -P password raw xxxx xxxxxxxx xxxx xxxx

[0080] Optionally, in this embodiment, in addition to managing the server cluster solely through the out-of-band management interface, the controller can also manage the server cluster through, but is not limited to, a combination of the in-band and out-of-band management interfaces. Figure 3 This is a schematic diagram illustrating the operation of a controller according to an embodiment of this application. Figure 3 As shown, specifically, the controller may include, but is not limited to, a management module, a resource scheduling module, a BIOS management module, a log collection module, and a power-saving and noise reduction module. The management module may, but is not limited to, receiving user commands, parsing the commands, scheduling relevant modules, receiving feedback interfaces from modules, and displaying them to the user. For example, upon receiving a user command (i.e., an operation request) for configuring the server cluster, the management module may send the user command, i.e., the configuration command, to the BIOS management module. Upon receiving a user command to instruct the server cluster to execute server business tasks, the management module may send the user command, i.e., the scheduling command, to the resource scheduling module. The resource scheduling module may, but is not limited to, receiving scheduling commands from the management module, allocating tasks according to the target server load, and feeding back the task allocation results to the management module. The BIOS management module may, but is not limited to, receiving configuration commands from the management module, issuing configurations through out-of-band management interfaces, and feeding back the configuration results to the management module. The log collection module may, but is not limited to, receiving log collection commands (i.e., a type of user command) from the management module, obtaining serial port logs and hardware status information from each target server in the server cluster, storing the logs in a central database, and feeding back the results. The energy-saving and noise-reduction module can, but is not limited to, monitor the target server load and ambient temperature in real time through the out-of-band management interface, automatically adjust the fan speed and power consumption mode, and feed the adjustment results back to the management module.

[0081] Optionally, the management module may, but is not limited to, be responsible for receiving user instructions and scheduling other modules to execute tasks. The instruction distribution logic of the management module may, but is not limited to, include: processing user requests based on a priority queue (such as high-priority tasks (whether configuration tasks or server business tasks) being processed in the queue); the exception circuit breaker mechanism of the management module may, but is not limited to, automatically marking a server as "faulty" and notifying maintenance if a server fails to operate three times in a row.

[0082] Optionally, the resource scheduling module may, but is not limited to, dynamically allocate tasks and resources based on the load of target servers in the server cluster. It may, but is not limited to, incorporating machine learning models to predict the load of each target server over the next 5 minutes to optimize task allocation, and may, but is not limited to, allowing high-priority tasks to preempt resources from low-priority tasks (based on QoS (Quality of Service) policies). Specifically, it may, but is not limited to, obtaining hardware-level metrics (such as CPU temperature and power consumption) through out-of-band management interfaces and obtaining operating system metrics (such as CPU utilization) through SSH connections. The dynamic weighting formula is defined as:

[0083] ,

[0084] Among them, W i That is, the weight value of the target server i, CPU i This refers to the CPU utilization of target server i, Mem. i This refers to the memory utilization of target server i, BW. free That is, the remaining bandwidth of target server i, BW total This represents the total bandwidth of target server i, with α=0.5, β=0.3, and γ=0.2. The values ​​of α, β, and γ can be flexibly configured. The target servers in the server cluster are sorted in descending order of weight to obtain a server list. The target server with the highest weight is selected, and its remaining resources are checked to see if they meet the task requirements. If they do, the task is assigned and the resource pool is updated; otherwise, the second-best server is selected. When the weights of target servers in the server cluster are similar, a connection-first strategy can be used, but is not limited to. To support overload protection, if a server's CPU utilization consistently exceeds 90% for a specified duration, the target server can be automatically removed from the scheduling pool, but is not limited to...

[0085] Optionally, in this embodiment, the BIOS management module may, but is not limited to, convert user commands into vendor-specific commands (i.e., commands that can be received by each out-of-band management interface) to update the BIOS settings. If the new BIOS settings cause server malfunctions, the system will automatically revert to the most recent stable version.

[0086] Optionally, in this embodiment, the log collection module can, but is not limited to, simultaneously collect serial port logs, out-of-band logs, and operating system logs. It uses the LZ4 algorithm (a lossless data compression algorithm) to compress logs in real time, reducing storage space usage. Specifically, it can, but is not limited to, redirecting server serial port output to a local file via an out-of-band interface. It can, but is not limited to, periodically executing IPMI commands to obtain system event logs. After obtaining the logs, it can, but is not limited to, extracting key events (such as hardware error codes, temperature alarms) using regular expressions. It uses Elasticsearch (a distributed search engine) to build an index, supporting retrieval by time / device / event type. It can, but is not limited to, automatically extracting key information from the logs; for example, it can, but is not limited to, determining target server faults based on preset rules (such as CPU temperature > 85℃, hard disk SMART (Self-Monitoring, Analysis, and Reporting Technology) errors), thereby triggering email or SMS notifications to maintenance personnel to repair the faulty target server. Through scheduled tasks and event-driven mechanisms, full lifecycle management of logs without manual intervention can be achieved.

[0087] Optionally, in this embodiment, the energy-saving and noise-reduction module can, but is not limited to, automatically adjust the fan speed and power consumption mode based on server load and ambient temperature. The energy-saving and noise-reduction module can, but is not limited to, simultaneously optimize fan speed (noise reduction) and CPU power consumption mode (energy saving) by using a multi-objective genetic algorithm to find the Pareto optimal solution. Specifically, the energy-saving and noise-reduction module can, but is not limited to, collect data from the server's internal temperature sensor via an out-of-band management interface and then dynamically adjust the fan speed using a PID (Proportional Integral Differential) algorithm.

[0088] FanSpeed ​​(fan speed) = Kp × (T_current - T_target) + Ki × +Kd×d(T_err) / dt,

[0089] Where T_target is the preset temperature threshold (e.g., 40℃), T_current is the current target server temperature, and T_err = T_current - T_target. The fan mode can be automatically switched to low-power mode during idle periods (e.g., early morning). The power supply strategy can be adjusted in advance based on load prediction (ARIMA model (Autoregressive Integrated Moving Average model)).

[0090] The above controller enables centralized management of all servers under the same local area network, simplifies configuration and operation, realizes intelligent scheduling and sharing of server resources, improves resource utilization and operation and maintenance efficiency, provides one-click operation (such as power on / off, log collection, serial port printing, etc.), reduces operation and maintenance complexity, realizes automated energy saving and noise reduction functions, and reduces energy consumption and noise pollution of server clusters.

[0091] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0092] Embodiments of this application also provide a control device for a server cluster, which is applied to the controller of the server cluster. Figure 4 This is a structural block diagram of a server cluster control device according to an embodiment of this application, such as... Figure 4 As shown, the device includes:

[0093] The first acquisition module 402 is used to acquire target address information of the server cluster. The target address information is used to indicate the addresses of multiple target servers in the server cluster in the local area network. The multiple target servers include various types of servers, and the controller is in the local area network.

[0094] The second acquisition module 404 is used to obtain the port information of each target server from the server cluster through the target address information, and obtain multiple target port information, wherein the target port information is used to indicate the open status of each server port of the target server.

[0095] The determination module 406 is used to determine the server interface information of the server cluster based on multiple target port information and a set of port information. The server interface information is used to indicate the out-of-band management interface used by various servers in the server cluster, and the set of port information records the correspondence between the server's port information and the out-of-band management interface used by the server.

[0096] The control module 408 is used to control each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information. The operation request is used to request the servers in the server cluster to perform an operation.

[0097] Through the above device, the server cluster controller obtains the target address information of the server cluster, and then obtains the port information of each target server in the server cluster through the target address information, resulting in multiple target port information. Based on the multiple target port information and the set of port information, the server interface information of the server cluster is determined. That is, the server cluster controller can ultimately obtain the server interface information used to indicate the out-of-band management interface used by various servers in the server cluster by obtaining the target address information. The controller controls each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information, so that the management and control of the server cluster can be realized through the controller as a single control tool. This reduces the difficulty of controlling the server cluster and improves the control efficiency of the server cluster. Therefore, it can solve the technical problem of low control efficiency of server clusters and achieve the technical effect of improving the control efficiency of server clusters.

[0098] Optionally, the first acquisition module includes: a broadcast unit, used to broadcast an address acquisition request within the local area network (LAN), wherein the address acquisition request is used to request a device possessing a reference network address from a network address set to return the reference physical address corresponding to the reference network address, the network address set including the network addresses of multiple devices in the LAN, the multiple devices including multiple target servers, and the reference physical address being the physical address of the device; a first receiving unit, used to receive the reference physical addresses returned by each device, obtaining multiple pairs of corresponding reference network addresses and reference physical addresses; and a filtering unit, used to filter multiple target address pairs from the multiple pairs of corresponding reference network addresses and reference physical addresses, wherein the target address pair is a target network address and a target physical address corresponding to a target server, the target network address is the network address of the target server in the LAN, the target physical address is the physical address of the target server, and the target address information includes multiple target address pairs.

[0099] Optionally, the filtering unit is also configured to: connect various devices through multiple pairs of corresponding reference network addresses and reference physical addresses; obtain device information of each device from each device, wherein the device information is used to indicate the device type of the device; and filter multiple target address pairs from multiple pairs of corresponding reference network addresses and reference physical addresses based on the device information.

[0100] Optionally, the second acquisition module includes: a sending unit, configured to send standard data packets corresponding to each server port of each target server to each server port of each target server through multiple target address pairs, wherein the target address pair is a target network address and a target physical address corresponding to the target server, the target network address is the network address of the target server in the local area network, the target physical address is the physical address of the target server, and the target address information includes multiple target address pairs; a second receiving unit, configured to receive response data packets from each target server within a target time period after sending the standard data packets, wherein the response data packets are data packets returned by each target server in response to the received standard data packets; a first determining unit, configured to determine that the port status of the server port corresponding to the response data packet is open when a response data packet is received within the target time period, wherein the multiple target port information includes multiple port statuses; and a second determining unit, configured to determine that the port status of the server port corresponding to the response data packet is closed when no response data packet is received within the target time period.

[0101] Optionally, the determining module includes: a lookup unit, used to look up each out-of-band management interface corresponding to each target port status group from the port status groups and out-of-band management interfaces with corresponding relationships, wherein each target port status group includes multiple target port states, each target port state is the state of the server port of the target server, and the target port information includes the target port status group; and a recording unit, used to record the out-of-band management interfaces corresponding to each target server found, thereby obtaining server interface information.

[0102] Optionally, the control module includes: a third receiving unit for receiving a reference operation request, wherein the reference operation request is used to request a reference server in the server cluster to perform a reference operation, and multiple target servers include the reference server, and the operation request includes the reference operation request; a matching unit for matching corresponding reference interface information for the reference server from the server interface information, wherein the reference interface information is used to indicate the reference management interface used by the reference server, and the out-of-band management interface includes the reference management interface; and a control unit for controlling the reference server to respond to the reference operation request through the reference interface information.

[0103] Optionally, the control unit is further configured to: match the corresponding reference template information for the reference operation request from the instruction template information of the server cluster according to the reference interface information, wherein the instruction template information is used to indicate the instruction format of the instructions that can be received by various out-of-band management interfaces, and the reference template information is used to indicate the instruction format of the instructions that can be received by the reference management interface; generate a reference operation instruction according to the reference template information and the reference operation request, wherein the reference operation instruction is used to instruct the reference server to respond to the reference operation request; and send the reference operation instruction to the reference server through the reference management interface.

[0104] For a description of the features in the embodiment corresponding to the control device of the server cluster, please refer to the relevant description in the embodiment corresponding to the control method of the server cluster, which will not be repeated here.

[0105] Embodiments of this application also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above-described server cluster control method embodiments.

[0106] Embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above-described server cluster control method embodiments when running.

[0107] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0108] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described server cluster control method embodiments.

[0109] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described server cluster control method embodiments.

[0110] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0111] The control method and apparatus for a server cluster provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A control method for a server cluster, characterized in that, The method, applied to the controller of the server cluster, includes: The target address information of the server cluster is obtained, wherein the target address information is used to indicate the address of multiple target servers in the server cluster in the local area network. The multiple target servers include various types of servers. The controller is in the local area network. The various types of servers correspond to various types of out-of-band management interfaces. The target address information is obtained by performing device detection in the local area network to collect device address information and filter out server devices. The port information of each target server in the server cluster is obtained by using the target address information to obtain multiple target port information, wherein the target port information is used to indicate the open status of each server port of the target server; The server interface information of the server cluster is determined based on multiple target port information and port information sets, wherein the server interface information is used to indicate the out-of-band management interface used by various servers in the server cluster, and the port information set records the correspondence between the server port information and the out-of-band management interface used by the server. The server interface information is used to control each target server in the server cluster to respond to operation requests to the server cluster, wherein the operation request is used to request the servers in the server cluster to perform an operation; The step of controlling each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information includes: receiving a reference operation request, wherein the reference operation request is used to request a reference server in the server cluster to perform a reference operation, the plurality of target servers include the reference server, and the operation request includes the reference operation request; matching corresponding reference interface information for the reference server from the server interface information, wherein the reference interface information is used to indicate the reference management interface used by the reference server, and the out-of-band management interface includes the reference management interface; and controlling the reference server to respond to the reference operation request through the reference interface information.

2. The server cluster control method according to claim 1, characterized in that, The step of obtaining the target address information of the server cluster includes: Broadcast an address acquisition request within the local area network, wherein the address acquisition request is used to request a device with a reference network address in the network address set to return the reference physical address corresponding to the reference network address, the network address set including the network addresses of multiple devices in the local area network, the multiple devices including multiple target servers, and the reference physical address being the physical address of the device; Receive the reference physical address returned by each of the devices to obtain multiple pairs of corresponding reference network addresses and reference physical addresses; Multiple target address pairs are selected from multiple pairs of corresponding reference network addresses and reference physical addresses. The target address pair is a target network address and a target physical address that correspond to the target server. The target network address is the network address of the target server in the local area network, and the target physical address is the physical address of the target server. The target address information includes multiple target address pairs.

3. The server cluster control method according to claim 2, characterized in that, The step of selecting multiple target address pairs from multiple pairs of corresponding reference network addresses and reference physical addresses includes: Each device is connected through multiple pairs of corresponding reference network addresses and reference physical addresses; Obtain device information for each of the devices, wherein the device information is used to indicate the device type of the device; Based on the device information, multiple target address pairs are selected from multiple pairs of corresponding reference network addresses and reference physical addresses.

4. The server cluster control method according to claim 1, characterized in that, The step of obtaining port information of each target server from the server cluster using the target address information to obtain multiple target port information includes: Standard data packets corresponding to each of the target servers are sent to each of the server ports of each of the target servers through multiple target address pairs. The target address pair is a target network address and a target physical address that have a corresponding relationship with the target server. The target network address is the network address of the target server in the local area network, and the target physical address is the physical address of the target server. The target address information includes multiple target address pairs. Within a target time period after sending the standard data packet, response data packets are received from each of the target servers, wherein the response data packets are data packets returned by each of the target servers in response to the received standard data packet; If the response data packet is received within the target time period, the port status of the server port corresponding to the response data packet is determined to be open, wherein the multiple target port information includes multiple port statuses; If no response data packet is received within the target time period, the port status of the server port corresponding to the response data packet is determined to be closed.

5. The server cluster control method according to claim 1, characterized in that, The step of determining the server interface information of the server cluster based on multiple sets of target port information and port information includes: Find the out-of-band management interface corresponding to each target port status group from the corresponding port status groups and the out-of-band management interface, wherein each target port status group includes multiple target port statuses, each target port status is the status of the server port of the target server, and the target port information includes the target port status group; Record the out-of-band management interface corresponding to each of the target servers found to obtain the server interface information.

6. The server cluster control method according to claim 1, characterized in that, The step of controlling the reference server to respond to the reference operation request through the reference interface information includes: According to the reference interface information, the corresponding reference template information is matched from the instruction template information of the server cluster for the reference operation request. The instruction template information is used to indicate the instruction format of various out-of-band management interfaces that can be received, and the reference template information is used to indicate the instruction format of the instructions that the reference management interface can receive. A reference operation instruction is generated based on the reference template information and the reference operation request, wherein the reference operation instruction is used to instruct the reference server to respond to the reference operation request; The reference operation command is sent to the reference server through the reference management interface.

7. A control device for a server cluster, characterized in that, A controller applied to the server cluster, the device comprising: The first acquisition module is used to acquire the target address information of the server cluster, wherein the target address information is used to indicate the address of multiple target servers in the server cluster in the local area network, the multiple target servers include various types of servers, the controller is in the local area network, the various types of servers correspond to various types of out-of-band management interfaces, and the target address information is obtained by performing device detection in the local area network to collect device address information and filter out server devices from it; The second acquisition module is used to acquire port information of each target server from the server cluster through the target address information, thereby obtaining multiple target port information, wherein the target port information is used to indicate the open status of each server port of the target server; The determination module is used to determine the server interface information of the server cluster based on multiple target port information and a set of port information, wherein the server interface information is used to indicate the out-of-band management interface used by various servers in the server cluster, and the set of port information records the correspondence between the server's port information and the out-of-band management interface used by the server. The control module is used to control each target server in the server cluster to respond to operation requests to the server cluster based on the server interface information, wherein the operation request is used to request the servers in the server cluster to perform an operation; The control module includes: a third receiving unit for receiving a reference operation request, wherein the reference operation request requests a reference server in the server cluster to perform a reference operation, and the multiple target servers include the reference server, and the operation request includes the reference operation request; a matching unit for matching corresponding reference interface information for the reference server from the server interface information, wherein the reference interface information indicates the reference management interface used by the reference server, and the out-of-band management interface includes the reference management interface; and a control unit for controlling the reference server to respond to the reference operation request through the reference interface information.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the control method for the server cluster as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the control method for the server cluster as described in any one of claims 1 to 6.

Images