A hardware management method and hardware management platform based on virtualization technology

Abstract

The present application provides a hardware management method based on virtualization technology, which is used for a hardware management platform. The hardware management platform obtains basic information of processors arranged in a computing device, creates a virtual BMC on the BMC, and configures the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system. Thus, the virtual BMC can manage the processors on the computing device one by one. Through the independent connection between the virtual BMC and the operation and maintenance system, when the processor on the computing device changes, the impact on the operation and maintenance of other processors is reduced, thereby improving the stability of the system.

Description

[0001] This application claims the priority of the Chinese patent application filed with the State Intellectual Property Office of China on October 6, 2023, with application number 202311288338.7 and invention name “A hardware platform management method and hardware management platform based on virtualization technology”, the entire contents of which are incorporated by reference in this application. Technical Field

[0002] The invention relates to the field of hardware management, and in particular to a hardware management method based on virtualization technology. Background Art

[0003] With the development of cloud computing technology, the demand for computing power in cloud scenarios has gradually increased. Many computing tasks require large-scale XPU and other hardware clusters to be performed under high-performance networks. All XPU and other processor hardware resources are pooled and connected through high-speed networks to form computing hardware devices. Taking XPU-BOX as an example, physically the same or different types of XPUs are concentrated in a hardware box. As the smallest hardware entity in the cluster network, these entities are collectively referred to as XPU-BOX.

[0004] Each XPU-BOX is equipped with a Baseboard Management Controller (BMC) to monitor and manage the BOX and provide a standard connection interface for the operation and maintenance system. The operation and maintenance system implements operation and maintenance management of all XPU-BOXs through the interface. When the number of XPUs changes during the XPU-BOX update, the device management interface changes, which requires the cloud management system and BMC to be updated repeatedly, resulting in low management efficiency and poor stability of the cloud management system. Summary of the invention

[0005] The embodiment of the present application provides a hardware platform management method and a hardware management platform based on virtualization technology. Through the hardware management platform, a virtual BMC is created, and one virtual BMC manages one corresponding processor. On this basis, when the processor on the computing device is added or removed, it will not affect the management of other processors on the computing device by the operation and maintenance system, which helps to improve the operation and maintenance efficiency and reliability. The technical solution is as follows:

[0006] In the first aspect, the present application provides a hardware management method based on virtualization technology, which is applied to a hardware management platform, the hardware management platform runs on a baseboard management controller BMC, the BMC is connected to a processor set on a computing device, wherein the hardware management platform is connected to an operation and maintenance system, and the operation and maintenance system is used to obtain the status information of the computing device, and the method specifically includes the following steps: obtaining the basic information of the aforementioned processor, the basic information may include the number information, model information, etc. of the processor, the hardware management platform creates a virtual BMC on the BMC according to the basic information, and the virtual BMC is used to manage the processor. The hardware management platform configures the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system, wherein the operation and maintenance system is used to obtain the status information of the first processor through the virtual BMC. When multiple processors are set on the computing device, the hardware management platform obtains the basic information of the multiple processors, and the basic information includes the number information of the multiple processors. It is worth noting that the connection between the BMC and the processor set on the computing device can be a connection established with the BMC when the processor is first connected to the computing device, or the processor can be set on the computing device before connecting to the BMC, and connected to the BMC when powered on or started for the first time.

[0007] In the solution provided in the present application, in order to achieve efficient management of processors in hardware computing devices, the hardware management platform creates virtual BMCs corresponding to these processors in the platform computing device based on the basic information of the processors connected to the current hardware computing device, thereby achieving accurate management of the processors. At the same time, due to the use of the virtual BMC management method, the interface adaptation work of the operation and maintenance system only needs to be adapted to the changed processors, and is not affected by changes in the entire hardware computing device, thereby improving the stability of the interface. By adopting a virtual BMC to manage the corresponding processor, when a single virtual BMC fails, the impact on other virtual BMCs is reduced, thereby improving the stability of the overall operation and maintenance system.

[0008] In combination with the first aspect, in a possible implementation of the first aspect, the basic information also includes management channel information of the processor, and a virtual BMC is created based on the basic information. The method also includes the following specific steps: the hardware management platform creates a virtual BMC for managing the processor through the management channel based on the basic information.

[0009] In the solution provided in the present application, in order to realize one-to-one management of the processor by the virtual BMC, the management channel corresponding to the processor is mapped to the corresponding virtual BMC when creating the corresponding virtual BMC, so that each BMC can manage the corresponding processor through an independent management channel to avoid affecting the communication between other virtual BMCs and the processor.

[0010] In combination with the first aspect, in a possible implementation of the first aspect, the hardware management platform is connected to the operation and maintenance system through an Internet Protocol IP address, wherein the IP address is stored on the BMC, and after the virtual BMC is created, the hardware management platform configures the virtual BMC so that the virtual BMC is bound to the IP address, thereby realizing the connection of the virtual BMC to the operation and maintenance system through the IP address. When there are multiple processors and multiple virtual BMCs are created, the hardware management platform configures each of the multiple virtual BMCs so that each virtual BMC is bound to a corresponding IP address.

[0011] In the solution provided by the present application, the connection between the virtual BMC and the operation and maintenance management system is realized by configuring the virtual BMC to bind an IP address. In the case of multiple virtual BMCs, each virtual BMC is bound to an independent IP address, thereby ensuring independent communication between the virtual BMC and the operation and maintenance system, so that when a problem occurs in one of the virtual BMCs, it will not affect other virtual BMCs.

[0012] In combination with the first aspect, in a possible implementation of the first aspect, the basic information of the processor also includes the type of the processor, and a configuration file is also stored in the BMC, and different types of processors load different configuration files. After creating the virtual BMC, the hardware management platform configures the virtual BMC according to the configuration file so that the virtual BMC loads the configuration file, wherein the configuration file is used to instruct the virtual BMC to manage the function of the corresponding type of processor.

[0013] In the solution provided by the present application, the virtual BMC created by the hardware management platform can be a stateless virtual BMC. On the one hand, the stateless virtual BMC can be created by the hardware management platform more quickly and connected to the processor through the hardware management channel. On the other hand, it can also reduce the amount of data stored in the BMC, and only the data required to create the stateless BMC and the configuration files corresponding to each processor type need to be stored. After creating the virtual BMC, the configuration file is loaded according to the type of the processor, thereby realizing the loading of the management function of the processor, so as to facilitate the virtual BMC to manage the processor.

[0014] In combination with the first aspect, in a possible implementation manner of the first aspect, when the processor is removed from the computing device, the hardware management platform obtains information of removing the processor from the computing device, and uninstalls the virtual BMC from the BMC according to the information.

[0015] In the solution provided in the present application, when a processor is removed from a computing device, the virtual BMC that manages it will be uninstalled by the hardware management platform. On the one hand, this saves the storage space of the BMC and provides space for new processors added to subsequent computing devices. On the other hand, when a processor is removed, the hardware management platform needs to notify the operation and maintenance system to remove the virtual BMC from the operation and maintenance system to ensure timely updating of the operation and maintenance information.

[0016] In combination with the first aspect, in a possible implementation of the first aspect, after the hardware management platform configures the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system, the hardware management platform further configures the virtual BMC so that the virtual BMC registers the processor to the operation and maintenance system.

[0017] In the solution provided in the present application, in order to realize real-time update and grasp of processor status information by the operation and maintenance system, after the virtual BMC establishes a connection with the operation and maintenance system, the processor is registered with the operation and maintenance system to realize the operation and maintenance system's management of the processor and obtain the processor's status information in real time.

[0018] In combination with the first aspect, in a possible implementation of the first aspect, the type of processor includes one or more of a central processing unit CPU, a neural network processor NPU, a graphics processor GPU, a tensor processor TPU, a deep learning processor DPU, and a brain processor BPU.

[0019] In the solution provided in the present application, the types of processors set on the computing device can be diverse. To meet different task requirements, the same processor can be set on the same computing device, or different processors can be set to achieve diversified task processing and improve task processing efficiency.

[0020] In combination with the first aspect, in a possible implementation of the first aspect, when a new processor is deployed in a computing device, the hardware management platform can create a virtual BMC to be enabled before the new processor is deployed. When the new processor is connected, the corresponding configuration file is loaded according to the type of the processor obtained, and the connection between the virtual BMC and the processor and the operation and maintenance system can be configured; or when a notification of the new processor access is received, a corresponding virtual BMC can be created according to the corresponding basic information.

[0021] In the solution provided in the present application, a virtual BMC to be started is created before a new processor is connected, so that the virtual BMC can be quickly started when the new processor is connected. It only needs to load the configuration file according to the type of the new processor, thereby improving the connection speed between the virtual BMC and the processor and the operation and maintenance system, thereby improving the operation and maintenance efficiency.

[0022] The second aspect or any implementation of the second aspect is an implementation of a device corresponding to the first aspect or any implementation of the first aspect. Specifically, the present application provides a hardware management platform, which runs on a baseboard management controller BMC, and the BMC is connected to a processor set on a computing device, wherein the hardware management platform is connected to an operation and maintenance system, and the operation and maintenance system is used to obtain status information of the computing device. The hardware management platform specifically includes: an information acquisition module, a creation module, and a configuration module, wherein the information acquisition module is used to obtain basic information of the processor, the creation module is used to create a virtual BMC on the BMC according to the basic information, the virtual BMC is used to manage the processor, and the configuration module is used to configure the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system, wherein the operation and maintenance system is used to obtain status information of the first processor through the virtual BMC.

[0023] In combination with the second aspect, in a possible implementation of the second aspect, the basic information also includes management channel information of the processor, and the creation module in the hardware management platform is specifically used to create a virtual BMC for managing the processor through the management channel according to the basic information.

[0024] In combination with the second aspect, in a possible implementation of the second aspect, the hardware management platform is connected to the operation and maintenance system via an Internet Protocol IP address, wherein the BMC stores the IP address, and the configuration module in the hardware management platform is specifically used to configure the virtual BMC so that the virtual BMC is bound to the IP address, and the virtual BMC is also used to connect to the operation and maintenance system via the IP address.

[0025] In combination with the second aspect, in a possible implementation of the second aspect, the basic information also includes the type of processor, the BMC also stores a configuration file, and after the virtual BMC is created, the configuration module is also used to configure the virtual BMC according to the configuration file so that the virtual BMC loads the configuration file, and the configuration file is used to instruct the virtual BMC to manage the function of the corresponding type of processor.

[0026] In combination with the second aspect, in a possible implementation of the second aspect, the acquisition module is further used to obtain information of removing a processor from the computing device, and the hardware management platform also includes an uninstall module, wherein the uninstall module is used to uninstall the virtual BMC from the BMC according to the information.

[0027] In combination with the second aspect, in a possible implementation of the second aspect, after configuring the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connections established between other virtual BMCs and the operation and maintenance system, the configuration module is also used to configure the virtual BMC so that the virtual BMC registers the processor to the operation and maintenance system.

[0028] In a third aspect, the present application provides a BMC cluster, including at least one BMC, each BMC including a BMC processor and a BMC memory; the BMC processor of at least one BMC is used to execute instructions stored in the BMC memory of at least one BMC, so that the BMC cluster executes the above-mentioned first aspect and a method combined with any one of the implementation methods of the above-mentioned first aspect.

[0029] In a fourth aspect, the present application provides a computing device cluster, comprising at least one computing device, each computing device comprising at least one processor and at least one BMC, wherein at least one BMC is used to execute the above-mentioned first aspect and the method combined with any one of the implementation methods in the above-mentioned first aspect, so that the operation and maintenance system obtains the status information of at least one processor in each computing device through the virtual BMC running in at least one BMC.

[0030] In a fifth aspect, the present application provides a computer program product comprising instructions, which, when executed by a computer device cluster, enables the computer device cluster to execute the above-mentioned first aspect and a method combined with any one of the implementation modes of the above-mentioned first aspect.

[0031] In a sixth aspect, the present application provides a computer-readable storage medium, comprising computer program instructions. When the computer program instructions are executed by a computing device cluster, the computing device cluster executes the above-mentioned first aspect and a method combined with any one of the implementation modes of the above-mentioned first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic diagram of a scenario of the operation and maintenance management system provided by an embodiment of the present application;

[0033] Figure 2 This is a schematic diagram of an architecture of a hardware management method based on virtualization technology provided in an embodiment of the present application;

[0034] Figure 3 This is a flow chart of a hardware management method based on virtualization technology provided in an embodiment of the present application;

[0035] Figure 4 This is another flow chart of the hardware management method based on virtualization technology provided in the embodiment of the present application;

[0036] Figure 5 This is another flow chart of the hardware management method based on virtualization technology provided in the embodiment of the present application;

[0037] Figure 6 This is another flow chart of the hardware management method based on virtualization technology provided in the embodiment of the present application;

[0038] Figure 7 It is a structural diagram of the hardware management platform provided in the embodiment of the present application;

[0039] Figure 8 This is a structural diagram of a BMC of a hardware management platform based on virtualization technology provided in an embodiment of the present application;

[0040] Fig. 9 It is a structural diagram of a BMC cluster of a hardware management platform based on virtualization technology provided in an embodiment of the present application.

[0041] Fig.10 It is a structural diagram of a computing device of a hardware management platform based on virtualization technology provided in an embodiment of the present application;

[0042] Fig.11 It is a structural diagram of a computing device cluster of a hardware management platform based on virtualization technology provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application are described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0044] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0045] References to "one embodiment" or "some embodiments" etc. described in this specification mean that a particular feature, structure or characteristic described in conjunction with the embodiment is included in one or more embodiments of the present application. Thus, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear at different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0046] First, the nouns that appear in this article are explained as follows:

[0047] Baseboard Management Controller (BMC): It is a core component of the server management system defined by the Intelligent Platform Management Interface (IPMI) protocol, i.e., the Intelligent Platform Management Interface. It is a hardware manager integrated in servers, network devices, and other computer systems. Its function is to monitor the hardware status of the device, perform remote management operations, and provide monitoring and control functions for the device. BMC communicates with various components (such as processors, network cards, graphics cards, sensors, etc.) on servers or similar devices through different interfaces, such as the system management bus (SMBUS) l, etc., obtains device information and controls and adjusts it. Management systems for servers and other types of server hardware can run on BMC. Through BMC, remote management of servers can be achieved, including remote power on and off of servers, installation and deployment of operating systems (OS), remote KVM, fault monitoring, etc. BMC provides standard management protocols such as IPMI / Redfish, etc.

[0048] XPU: With the explosive development of AI, 5G, and IoT technologies, the CPU-based architecture can no longer meet the rapidly growing computing power needs in new scenarios. The computing power has been expanded from the central processing unit (CPU) to computing units or processors represented by graphics processing units (GPU), tensor processing units (TPU), neural network processors (NPU), brain processors (Brain Processing Unit, BPU), deep learning processors (DPU), etc. These computing units or processors are collectively referred to as XPU. In the embodiments of the present application, XPU is equivalent to the processor.

[0049] Computing device: A device that usually has multiple processors and corresponding hardware, such as XPU-BOX, servers, data centers, etc. Taking the server as an example, it includes multiple hardware components such as processors, network cards, graphics cards, sensors, etc.

[0050] Infrastructure: The infrastructure includes multiple computing devices, each of which can provide computing, network, and storage resources.

[0051] Operation and maintenance system: obtains the status information of computing devices through BMC to implement operation and maintenance management of computing devices. The operation and maintenance system can run on computing devices or on other devices outside computing devices. For example, when a computing device is equipped with multiple devices such as a processor and a memory, the operation and maintenance system can obtain the status information of the processor and the memory through BMC to implement operation and maintenance management of the computing device.

[0052] XPU-BOX: A physical device that centrally manages multiple XPUs and provides power supply and heat dissipation management for XPUs.

[0053] Hardware management platform: A management platform that runs on the BMC, such as a BMC management system, or other forms of management platforms for managing hardware on servers or hosts.

[0054] Virtual BMC: The BMC system is instantiated through virtualization technology. The virtual BMC usually has the management function of the BMC for the hardware such as the processor set on the computing device. The virtual BMC also includes a stateless virtual BMC. The stateless virtual BMC usually only has the underlying functions of the BMC and requires a specific configuration file to implement the management function of a specific processor.

[0055] Redfish: Redfish is a management standard based on HTTPs services, which uses RESTful (Representational State Transfer) interface to implement device management.

[0056] Simple Network Management Protocol (SNMP): It is a standard network management protocol widely used in TCP / IP networks. This protocol can support network management systems to monitor whether the devices connected to the network have any situations that require management attention. SNMP uses a polling mechanism and provides the most basic function set. It is suitable for use in small, fast, and low-cost environments. In addition, SNMP uses User Datagram Protocol (UDP) messages as a carrier, so it is supported by most devices. At the same time, it ensures that management information is transmitted between any two points, making it easy for administrators to retrieve information at any node on the network and perform troubleshooting.

[0057] Basic input output system (BIOS) configuration: BMC can configure the server's BIOS parameters, such as boot order, memory frequency, and security mode, through the network or local storage media.

[0058] Intelligent Platform Management Interface (IPMI): The management function of BMC is implemented through the IPMI protocol. IPMI is a set of interactive standard management specifications, mainly used for server system cluster autonomy and monitoring the physical health characteristics of the server, such as temperature, voltage, fan working status, power status, etc. At the same time, IPMI is also responsible for recording various hardware information and log records to prompt users and locate subsequent problems. IPMI is an independent on-board component that can run independently of the host system CPU, BIOS, Unified Extensible Firmware Interface (UEFI) and OS, and its core component is BMC. In other words, the interaction between BMC and other components such as BIOS / UEFI, CPU, etc. is completed through IPMI. IPMI can communicate with the operation and maintenance system through different channels, such as local area network (LAN), etc., to transmit IPMI commands and data.

[0059] The management functions of BMC mainly include the following aspects:

[0060] Power control: BMC can control server startup, shutdown, restart, NMI and other operations through IPMI commands or physical buttons, and obtain the power status of the server.

[0061] Monitoring: BMC can monitor the server's physical health characteristics such as temperature, voltage, fan, power supply, etc. through sensors, and issue alarms or adjustments based on preset thresholds and policies.

[0062] Inventory: BMC can record the model, serial number, manufacturer and other information of replaceable parts on the server through Field Replaceable Unit (FRU) information to facilitate maintenance and replacement.

[0063] Log: BMC can record various events that occur on the server through the System Event Log (SEL), such as overtemperature, power failure, memory error, etc., and provide query and clear functions.

[0064] Recovery: BMC can perform corresponding recovery actions based on the events or commands received, such as restarting the server, turning off the fan, reducing the CPU frequency, etc.

[0065] Alarm: BMC can send alarm information to the upper-level network management or users through the SNMP protocol, such as high temperature, power failure, etc.

[0066] Serial port redirection: BMC can use the Serial over LAN (SOL) function to redirect the server's serial port output to the BMC's network interface, making it easier to remotely obtain the server's debugging information or console information.

[0067] Graphics redirection: BMC can redirect the server's graphics output to the BMC's network interface through the KVM over IP (Keyboard, Video and Mouse over IP) function, making it easier to remotely access the server's graphical interface or BIOS interface.

[0068] Firmware update: BMC can update the server firmware, such as BIOS, BMC itself, RAID card, etc., through the network or local storage media.

[0069] To further illustrate the application scenarios of the embodiments of this application, please refer to Figure 1 ,like Figure 1 As shown, Figure 1 This is a schematic diagram of the architecture of the operation and maintenance management system provided in an embodiment of the present application, specifically:

[0070] The operation and maintenance system 10 is connected to the computing devices 30, 31 and 32 in the infrastructure 1, wherein the computing device 30 is provided with BMC301, CPU302, CPU303, CPU304 and CPU305... The computing device 31 is provided with BMC311, GPU312, GPU313, GPU314 and GPU315... The computing device 32 is provided with BMC321, XPU322, XPU323, XPU324 and XPU325... Further, BMC301 is connected to CPU302, CPU303, CPU304 and CPU305 respectively, BMC311 is connected to GPU312, GPU313, GPU314 and GPU315 respectively, and BMC321 is connected to XPU322, XPU323, XPU324 and XPU325 respectively. Exemplarily, the BMC can obtain the status information of each processor through the hardware management channel between each processor. The BMC establishes a connection with the operation and maintenance system through the Redfish interface, and the operation and maintenance system obtains the status information of each processor through the BMC.

[0071] Furthermore, when there are other hardware in the computing device, the BMC can also connect to other hardware through the IPMI interface, obtain the status information of each hardware, establish a connection with the operation and maintenance system through Redfish or SNMP, and report the status information of each hardware in the computing device to the operation and maintenance system. Alternatively, it can receive instructions from the operation and maintenance system to perform specific tasks such as power control, monitoring, inventory query, firmware update, log query, recovery, alarm, etc., to achieve out-of-band management of computing devices.

[0072] Under this architecture, the BMC management strategy is that there is only one BMC system on a single XPU-BOX device, and what is presented to the operation and maintenance system is an XPU-BOX device. Taking the connection to the operation and maintenance system through the Redfish interface as an example, according to the Redfish interface definition, all XPU management interfaces are hung under the Redfish node of the XPU-BOX. The operation and maintenance system first accesses the Redfish interface of the XPU-BOX, obtains the Uniform Resource Identifier (URI) interface of the XPU, and then performs operation and maintenance management on the XPU.

[0073] As computing power requirements grow rapidly, BMC is presented as an XPU-BOX device. Business computing requires large-scale XPU clusters to be performed under high-performance networks, and XPU-BOX is quickly updated and iterated to integrate more XPUs. When the number of XPUs changes, the BMC Redfish interface needs to add corresponding URIs, resulting in changes in the device management interface, increasing operation and maintenance pressure, reducing operation and maintenance efficiency, and reducing the stability of the operation and maintenance system.

[0074] To solve the above problems, see Figure 2 ,like Figure 2 As shown, Figure 2 This is an architectural diagram of a hardware management method based on virtualization technology provided by an embodiment of the present application. The present application solution mainly uses virtualization technology to virtualize the same number of virtual BMCs as XPUs on the BMC. Each virtual BMC manages one XPU. Logically, each XPU has an independent BMC system. XPU changes do not affect operation and maintenance, which improves the stability of the operation and maintenance system. Specifically:

[0075] The embodiments of this application are Figure 1The computing device 32 in the example is used for explanation. In the computing device 32, a BMC 321 and processors XPU 322, XPU 323, XPU 324 and XPU 325 are provided. The BMC 321 is connected to each processor on the computing device. Through a physical management channel or an IPMI interface, the BMC 321 can obtain basic information of each processor, including the number of processors, processor type, processor model, etc. Among them, a hardware management platform runs on the BMC 321. The hardware management platform 20 obtains basic information of the processor through the BMC 321, such as creating virtual BMC 21, virtual BMC 22, virtual BMC 23 and virtual BMC 24 corresponding to the processor, wherein the virtual BMC 21 is used to manage the processor XPU 322, the virtual BMC 22 is used to manage the processor XPU 323, the virtual BMC 23 is used to manage the processor XPU 324, and the virtual BMC 24 is used to manage the processor XPU 325.

[0076] Further, the hardware management platform 20 configures virtual BMC 21, virtual BMC 22, virtual BMC 23, and virtual BMC 24 to establish connections with the operation and maintenance system 10, respectively, wherein the connections established between each virtual BMC and the operation and maintenance system 10 are independent of each other. Exemplarily, the connection established between virtual BMC 21 and the operation and maintenance system 10 is independent of the connection established between virtual BMC 22, virtual BMC 23, and virtual BMC 24 and the operation and maintenance system 10.

[0077] Therefore, each virtual BMC manages the corresponding processor, including but not limited to power control, monitoring, inventory query, firmware update, log query, recovery, alarm, etc. Thus, the virtual BMC can independently manage the corresponding processor and connect to the operation and maintenance system through an independent interface. Therefore, when the processor in the computing device changes, it will not affect other virtual BMCs, thereby improving the stability of the operation and maintenance system and the continuity of operation and maintenance management.

[0078] Please see further Figure 3 ,like Figure 3 As shown, Figure 3 This is a flow chart of a hardware management method based on virtualization technology provided by an embodiment of the present application, specifically:

[0079] Step S301: The hardware management platform obtains basic information of the processor.

[0080] In one example of an embodiment of the present application, a processor is provided on a computing device, and the processor is arranged in the computing device before the computing device is started. The BMC is connected to the processor through an IPMI interface or a system bus, and after the computing device is started, the basic information of the processor is obtained, and the basic information includes the number information, type information, model information, corresponding physical location, etc. of the processor. On the BMC, a hardware management platform is running, and the hardware management platform obtains the basic information of the processor through the BMC.

[0081] In another example of an embodiment of the present application, a computing device is provided with a processor, which is hot-inserted into the computing device after the computing device is started. After the computing device detects the insertion of the processor, it notifies a hardware management platform running on the BMC through a hardware signal such as a complex programmable logical device (CPLD) or a general-purpose input / output (GPIO). The hardware management platform obtains basic information of the processor through an IPMI interface or a system bus, and the basic information includes quantity information, type information, model information, corresponding physical location, etc.

[0082] It is worth noting that the number of processors can be one or more, and the processor types can be the same type or different types, which is not limited in the embodiments of the present application.

[0083] By obtaining basic information about the processors on the computing device, the hardware management platform first obtains information about the number, physical location, and corresponding system bus of the processors that need to be managed, making basic preparations for further creating a virtual BMC.

[0084] Step S302: the hardware management platform creates a virtual BMC on the BMC according to the basic information.

[0085] The hardware management platform creates a corresponding virtual BMC on the BMC based on the basic information. The created virtual BMC is used to manage the processors that correspond to it one by one, thereby achieving accurate and independent management and connection of the processors.

[0086] In one embodiment of the present application, the virtual BMC is a virtual BMC capable of executing a processor management function. Specifically, according to the type of processor, the hardware management platform creates a virtual BMC capable of executing a corresponding management function for the processor type. Figure 1 For example, when the processors are GPU312 and GPU313, the number of processors is 2 and the type of processor is GPU. The hardware management platform creates 2 virtual BMCs capable of managing GPU-type processors according to the number of processors and the type of processors.

[0087] In another embodiment of the present application, the virtual BMC is a stateless virtual BMC. The stateless virtual BMC may be a virtual BMC with only basic functions, which is not coupled to the specific XPU hardware, or does not have the function of managing a certain type of processor. Specifically, the BMC stores a virtual system image of the virtual BMC. For example, the configuration file is saved in the read-only data area of ​​the BMC, which saves the implementation method of the BMC management function and management channel for different XPU models. Before creating the virtual BMC, the hardware management platform determines the corresponding configuration file by obtaining the type of processor. After creating the virtual BMC, the virtual BMC is configured to load the configuration file of the corresponding model. The virtual BMC manages the XPU by reading the configuration file data. In combination with Figure 1 For example, when the processor is CPU302 or CPU303, the basic information of the processor obtained by the hardware management platform includes that the type of the processor is CPU. The configuration file is determined according to the type information of the processor, and the created virtual BMC is configured to load the corresponding configuration file to realize normal management of the CPU.

[0088] In this embodiment, the stateless virtual BMC is more conducive to the hardware management platform to quickly create a virtual BMC based on the basic information of the acquired processor, and only the system files of the stateless virtual BMC and the configuration files corresponding to different types of processors need to be stored in the BMC. By decoupling from the processor, the storage space of the BMC can be saved and the space utilization of the BMC can be increased. At the same time, the normal implementation of the management functions of the corresponding virtual BMC for different types of processors is guaranteed.

[0089] In another embodiment of the present application, a virtual BMC system is running on the virtual BMC, and the system is the underlying system of the virtual BMC. In the stateless virtual BMC, only the underlying system is loaded. The underlying system can be dynamically upgraded through the hardware management platform. Specifically, the virtual BMC system can be stored at a remote end, which can be an operation and maintenance system or a memory in a computing device. When the virtual BMC system is updated, the hardware management platform is notified to download remotely. The hardware management platform can close the original virtual BMC process, use the new virtual BMC system, and create a new virtual BMC to continue management.

[0090] Step S303: the hardware management platform configures the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system.

[0091] After the hardware management platform creates the corresponding virtual BMC, the hardware management platform configures the virtual BMC to establish a connection with the operation and maintenance system, and each connection between the virtual BMC and the operation and maintenance system is independent of the connection between other virtual BMCs and the operation and maintenance system, thereby ensuring the independent connection between the virtual BMC and the operation and maintenance system. Figure 2 , XPU322 and XPU323 are provided in the computing device 32, and the hardware management platform creates corresponding virtual BMC21 and virtual BMC22 according to the basic information of the processor, wherein virtual BMC21 is used to manage XPU322, and virtual BMC22 is used to manage XPU323. After the virtual instance BMC21 and virtual BMC22 are created, the hardware management platform configures the virtual instance BMC21 and virtual instance BMC22 to establish connections with the operation and maintenance system 10 respectively, and the virtual BMC21 can establish a connection with the operation and maintenance system 10 through an independent Redfish interface or SNMP protocol, and the virtual BMC22 can establish a connection with the operation and maintenance system 10 through an independent Redfish interface or SNMP protocol. The connection established between the virtual BMC21 and the operation and maintenance system 10 is independent of the connection established between the virtual BMC22 and the operation and maintenance system 10.

[0092] By configuring the virtual BMC so that the connection between the virtual BMC and the operation and maintenance system is independent of the connections between other virtual BMCs and the operation and maintenance system, the operation and maintenance system can implement independent management functions for any processor. Even if an individual virtual BMC fails or an individual processor changes, it will not affect the operation and maintenance system's ability to manage other processors, thereby improving the stability and reliability of the system.

[0093] Further, to clearly illustrate the specific steps of the embodiment of the present application, please refer to Figure 4 ,like Figure 4 As shown, Figure 4 This is a flow chart of a hardware management method based on virtualization technology provided by an embodiment of the present application, specifically:

[0094] S401 , the hardware management platform 20 obtains basic information of the processor 322 .

[0095] The hardware management platform obtains basic information of the processor 322, which includes quantity information, type information, model information, corresponding physical location, etc. When the processor 322 is a GPU, the basic information obtained by the hardware management platform includes the type information of the processor 322 as a GPU, and the basic information also includes the hardware management channel corresponding to the processor 322. For example, if the management channels for XPU322, XPU323, and XPU324 on the computing device 32 are Channel-1, Channel-2, and Channel-3, respectively, in order to correctly associate the virtual BMC with each processor, the hardware management platform completes the following mappings between the virtual BMC and the hardware management channel according to the hardware management channel:

[0096] Channel-1---->Virtual BMC1:channel-1;

[0097] Channel-2---->Virtual BMC2:channel-1;

[0098] Channel-3---->Virtual BMC3: channel-1.

[0099] S402: The hardware management platform 20 creates a virtual BMC 21 according to basic information.

[0100] The hardware management platform 20 creates a virtual BMC 21 on the BMC according to the basic information of the processor 322 obtained. The BMC 21 may be a stateless virtual BMC, which is decoupled from the processor 322. Furthermore, during the creation process, the hardware management channel of the processor corresponding to the virtual BMC 21 is synchronously configured into the virtual BMC 21.

[0101] S403: The hardware management platform 20 determines a configuration file according to the basic information.

[0102] The hardware management platform 20 determines the configuration file of the processor 322 according to the type information in the basic information of the processor 322 obtained. The BMC 321 may pre-store configuration files corresponding to different types of processors. When the hardware management platform 20 obtains the basic information of the processor 322, it determines the configuration file corresponding to the processor 322 from the BMC 321.

[0103] For example, different types of processors may have different interfaces, such as the platform environment control interface (PECI), the system bus SMBUS, etc. The BMC needs to communicate with the processor through these interfaces to obtain the processor's status information and control and adjust according to the preset strategy; different types of processors may have different sensors, such as temperature, voltage, power consumption, etc. The BMC needs to store the information of these sensors and record the events of these sensors according to the SEL; different types of processors may have different firmware, such as BIOS, microcode, etc. The BMC needs to update these firmware through IPMI commands, or store the information of these firmware through FRU; different types of processors may have different BIOS configurations, such as boot order, memory frequency, security mode, etc. The BMC needs to configure these BIOS parameters through IPMI commands, or store the information of these BIOS parameters through FRU. Therefore, different types of processors need to configure the virtual BMC to load the corresponding configuration file to achieve the function of correctly managing the corresponding processor.

[0104] In one embodiment of the present application, the configuration file can also be saved remotely, can be saved in the memory of the computing device, and can also be saved in the operation and maintenance system. When the hardware management platform obtains the type information of the processor, the configuration file can be obtained remotely to save the storage space of the local BMC.

[0105] S404, the hardware management platform 20 configures the virtual BMC 21 to load a configuration file.

[0106] After determining the configuration file, the hardware management platform 20 configures the virtual BMC 21 to load the configuration file. Exemplarily, the hardware management platform 20 may determine the corresponding configuration file through the BMC file system and send it to the virtual BMC 21, and the virtual BMC 21 loads the configuration file into the virtual BMC 21 after obtaining the configuration file; or the hardware management platform 20 may determine the corresponding configuration file through the BMC file system and send the address of the configuration file to the virtual BMC 21 in the form of an instruction, and the virtual BMC 21 obtains the corresponding configuration file through the instruction and loads it into the virtual BMC 21.

[0107] S405, the virtual BMC 21 loads the configuration file.

[0108] The virtual BMC 21 loads the acquired configuration file into the system of the virtual BMC 21 so that the virtual BMC 21 has the function of correctly managing the processor 322 and ensures the normal operation of the operation and maintenance system.

[0109] S406 , the hardware management platform 20 configures the virtual BMC 21 to obtain an IP address.

[0110] The hardware management platform 20 is connected to the operation and maintenance system 10 through an IP address. Through the IP protocol connection, the operation and maintenance system 10 can obtain the basic information of the hardware management platform and the status information of BMC321. Among them, at least one IP address is stored in BMC321. After the hardware management platform 20 creates the virtual BMC21, in order to enable the virtual BMC21 and the operation and maintenance system 10 to achieve independent connection and communication, the virtual BMC21 is configured to obtain the IP address stored in BMC321.

[0111] It is worth noting that the IP address can be a fourth version of the Internet Protocol IPv4 address or a sixth version of the Internet Protocol IPv6 address, which is not limited in the embodiments of the present application.

[0112] S407, the virtual BMC 21 connects to the operation and maintenance system 10 via the IP address.

[0113] After the virtual BMC 21 obtains the corresponding IP address, it binds the IP address, uses the IP address as the source address, and uses the obtained IP address of the operation and maintenance system 10 as the destination address to initiate a connection request to the operation and maintenance system to complete network initialization.

[0114] Thereby, the virtual BMC 21 is connected to the operation and maintenance system 10 , and the connection is independent.

[0115] S408 , the virtual BMC 21 manages the processor 322 through the hardware management channel.

[0116] The virtual BMC 21 obtains the status information of the processor 322 according to the hardware management channel configured by the hardware management platform 20 , and manages the processor 322 through the hardware management channel.

[0117] Through this configuration, each virtual BMC can independently manage the corresponding processor. When the processor changes, only the virtual BMC is affected, avoiding the impact on other virtual BMCs, thereby improving the stability of the operation and maintenance system.

[0118] S409 , the virtual BMC 21 is connected to the operation and maintenance system 10 , and the processor 322 is registered in the operation and maintenance system 10 .

[0119] After the virtual BMC 21 establishes an independent connection with the operation and maintenance system 10 and is connected to the processor 322 through the corresponding hardware management channel, the processor 322 is registered in the operation and maintenance system 10, so that the operation and maintenance system 10 manages the processor 322, which is conducive to further management and monitoring of the processor 322 in the future.

[0120] S410 , the virtual BMC 21 sends status information of the processor 322 .

[0121] The virtual BMC 21 obtains the status information of the processor 322 through the hardware management channel and sends the status information to the operation and maintenance system 10, so that the operation and maintenance system 10 can obtain the status information of the processor 322 in real time and manage and control it in time according to the status information of the processor.

[0122] An important feature of the physical hardware of the resource pool is that computing devices support hot-swappable processors. When traditional BMCs handle this scenario, they need to dynamically adjust the external interface after sensing a change in the number of processors and notify the operation and maintenance system of the change in the number of processors. The operation and maintenance system can only perform management after completing a processor number query through the interface.

[0123] The embodiment of the present application makes a one-to-one correspondence between the virtual BMC and the processor. When the number of processors changes, management can be completed by simply creating or deleting the virtual BMC system synchronously. Taking the scenario of adding a processor as an example, the specific steps in an example provided in the embodiment of the present application can be found in Figure 4 The steps in the hardware management platform, that is, obtaining the basic information of the newly added processor, creating a corresponding virtual BMC according to the basic information of the newly added processor, and configuring the virtual BMC to establish an independent connection with the operation and maintenance system, will not be repeated in this article.

[0124] In another example provided by the embodiment of the present application, a virtual BMC can be created in advance. The virtual BMC is a stateless BMC. Since there is no corresponding processor in the computing device that can be associated with it, the virtual BMC can be regarded as an idle virtual BMC. When the hardware management platform obtains the basic information of the newly added processor, the idle virtual BMC is configured to load the corresponding configuration file according to the basic information, so as to realize the management function of the newly added processor and improve the speed of the operation and maintenance system to manage the newly added processor.

[0125] Please see further Figure 5 ,like Figure 5 As shown, Figure 5 This is a flow chart of a hardware management method based on virtualization technology provided by an embodiment of the present application, specifically:

[0126] S501 , before acquiring the newly added processor 324 , create a virtual BMC 23 .

[0127] Before the processor 324 is hot-inserted into the computing device, the hardware management platform 20 first creates a virtual BMC 23. The virtual BMC 23 can be a stateless virtual BMC, and can also remain in a dormant state before the processor 324 is inserted. When the processor 324 is inserted, the virtual BMC can be quickly awakened.

[0128] It is worth noting that the hardware management platform 20 can determine the number of idle virtual BMCs to be created based on its own operating conditions, such as power, computing power, etc., and the embodiments of the present application do not limit this.

[0129] S502 , the hardware management platform 20 obtains basic information of the processor 324 .

[0130] When the processor 324 is inserted into the computing device, the computing device detects the insertion of the processor 324 and notifies the hardware management platform 20 running on the BMC 321 through a hardware signal such as CPLD or GPIO. The hardware management platform 20 obtains the basic information of the processor 324 through the IPMI interface or the system bus. The basic information includes the quantity information, type information, model information, corresponding physical location, etc. of the processor 324. When the processor 322 is a GPU, the basic information obtained by the hardware management platform includes the type information of the processor 322 being a GPU, and the basic information also includes the hardware management channel corresponding to the processor 322.

[0131] S503 , the hardware management platform 20 determines a configuration file according to basic information of the processor 324 .

[0132] The hardware management platform 20 determines the configuration file of the processor 324 according to the type information in the basic information of the processor 324 obtained. The BMC 321 may pre-store configuration files corresponding to different types of processors. When the hardware management platform 20 obtains the basic information of the processor 322, it determines the configuration file corresponding to the processor 324 from the BMC 321.

[0133] Steps S504 to S510 and Figure 4 Steps S404 to S410 shown in FIG. 4 are the same and will not be described in detail herein.

[0134] Another embodiment of the present application takes the scenario of removing the processor as an example, please refer to Figure 6 ,like Figure 6 As shown, Figure 6 This is a flow chart of a hardware management method based on virtualization technology provided by an embodiment of the present application, specifically:

[0135] S601 , the hardware management platform 20 obtains information about removing the processor 322 from the computing device 22 .

[0136] When the computing device 22 is running and the processor 322 is hot-plugged from the computing device 22, the computing device 22 obtains information about the unplugged processor 322 through hardware detection, notifies BMC 321 through the hardware management channel, and the hardware management platform 20 obtains uninstall information of the processor 322, which includes the physical location information, hardware management channel, model information, type information, etc. of the processor 322.

[0137] S602: The hardware management platform 20 uninstalls the virtual BMC 22 according to the information.

[0138] The hardware management platform 20 determines the virtual BMC 22 for managing the processor 322 according to the uninstallation, uninstalls the virtual BMC 22, and reclaims the IP address configured for the virtual BMC 22, thereby reducing the consumption of computing resources and saving the computing resources of the BMC.

[0139] S603, notify the operation and maintenance system.

[0140] After uninstalling BMC22, the hardware management platform 20 notifies the operation and maintenance system 10, so that the operation and maintenance system 10 updates the computing device information in a timely manner to improve the accuracy of the operation and maintenance system information.

[0141] The present application also provides a hardware management platform, which runs on a baseboard management controller BMC, which is connected to a processor provided on a computing device, wherein the hardware management platform is connected to an operation and maintenance system, which is used to obtain status information of the computing device, see below Figure 7 , Figure 7 This is a schematic diagram of the structure of a hardware management platform provided in an embodiment of the present application. The details are as follows:

[0142] The hardware management platform 20 includes an information acquisition module 901, a creation module 902, a configuration module 903 and an uninstallation module 904, wherein:

[0143] The information acquisition module 901 is used to acquire basic information of a processor, and is also used to acquire uninstallation information of removing a processor from a computing device.

[0144] The creation module 902 is used to create a virtual BMC on the BMC according to the basic information.

[0145] The configuration module 903 is used to configure the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system. It is also used to configure the virtual BMC according to the configuration file so that the virtual BMC loads the configuration file. It is also used to configure the virtual BMC so that the virtual BMC registers the processor to the operation and maintenance system.

[0146] Uninstall module 904, which is used to uninstall the virtual BMC from the BMC according to the uninstall information

[0147] It is worth noting that the users in the above embodiments can be replaced by multiple other users, and the above modules can all achieve corresponding technical functions, which is not limited in the embodiments of the present application.

[0148] Among them, the information acquisition module 901, the creation module 902, the configuration module 903, and the uninstallation module 904 can all be implemented by software, or can be implemented by hardware. Exemplarily, the implementation of the information acquisition module 901 is introduced below by taking the information acquisition module 901 as an example. Similarly, the implementation of the creation module 902, the configuration module 903, and the uninstallation module 904 can refer to the implementation of the information acquisition module 901.

[0149] As an example of a software functional unit, the information acquisition module 901 may include code running on a computing instance, and the virtual BMC may be regarded as a computing instance. Among them, the computing instance may include at least one of a physical host, a virtual machine, and a container. Further, the above-mentioned computing instance may be one or more. For example, the information acquisition module 901 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same area or in different areas. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone or in different availability zones, and each availability zone includes a data center or multiple data centers with similar geographical locations. Among them, generally a region may include multiple availability zones.

[0150] As an example of a hardware functional unit, the information acquisition module 901 may include at least one computing device, such as a server, etc. Alternatively, the information acquisition module 901 may also be a device implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), etc. The PLD may be a CPLD, a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0151] It should be noted that, in other embodiments, the information acquisition module 901 can be used to execute any step in the hardware management method based on virtualization technology, the creation module 902 can be used to execute any step in the hardware management method based on virtualization technology, the configuration module 903 can be used to execute any step in the hardware management method based on virtualization technology, and the uninstall module 904 can be used to execute any step in the hardware management method based on virtualization technology. The steps that the information acquisition module 901, the creation module 902, the configuration module 903, and the uninstall module 904 are responsible for implementing can be specified as needed. The information acquisition module 901, the creation module 902, the configuration module 903, and the uninstall module 904 respectively implement different steps in the hardware management method based on virtualization technology to realize the full functions of the hardware management platform.

[0152] The above details the hardware management platform, device and system of the embodiment of the present application. In order to facilitate better implementation of the above scheme of the embodiment of the present application, correspondingly, related equipment for cooperating in implementing the above scheme is also provided below.

[0153] This application provides a BMC, see below Figure 8 , Figure 8 1 is a schematic diagram of the structure of a BMC of a hardware management method based on virtualization technology provided in an embodiment of the present application. BMC321 includes: BMC bus 307, BMC processor 308, BMC memory 306 and BMC communication interface 309. BMC processor 308, BMC memory 306 and BMC communication interface 309 communicate with each other through BMC bus 307. It should be understood that the present application does not limit the number of BMC processors and BMC memories in BMC321.

[0154] The BMC bus 307 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 The BMC bus 307 may include a path for transmitting information between various components of the BMC 321 (eg, the BMC memory 306, the BMC processor 308, and the BMC communication interface 309).

[0155] The BMC processor 308 may include any one or more processors such as a CPU, a microprocessor (MP), or a digital signal processor (DSP).

[0156] The BMC memory 306 may include a volatile memory, such as a random access memory (RAM). The BMC processor 308 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0157] The BMC memory 306 stores executable program codes, and the BMC processor 308 executes the executable program codes to respectively implement the functions of the information acquisition module 901, the creation module 902, the configuration module 903, and the uninstallation module 904, thereby implementing the hardware management method based on virtualization technology. That is, the BMC memory 306 stores instructions for the hardware management platform to execute the hardware management method based on virtualization technology.

[0158] The BMC communication interface 309 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the BMC 321 and other devices or a communication network.

[0159] The embodiment of the present application also provides a BMC cluster. The BMC cluster includes at least one BMC. The BMC can be deployed in a server, such as a central server, an edge server, or a local server in a local data center.

[0160] See below Fig. 9 , Fig. 9 Schematic diagram of a BMC cluster structure of a hardware management method based on virtualization technology in an embodiment of the present application. Fig. 9 As shown, the BMC cluster includes at least one BMC 321 , and the BMC memory 306 in one or more BMCs 321 in the BMC cluster may store the same hardware management platform for executing instructions of the hardware management method based on virtualization technology.

[0161] In some possible implementations, one or more BMCs 321 in the BMC cluster may also be used to execute some instructions of the hardware management platform for executing the hardware management method based on virtualization technology. In other words, a combination of one or more BMCs 321 may jointly execute instructions of the hardware management platform for executing the hardware management method based on virtualization technology.

[0162] It should be noted that the BMC memory 306 in different BMCs 321 in the BMC cluster can store different instructions for executing some functions of the hardware management platform. That is, the instructions stored in the BMC memory 306 in different BMCs 321 can implement the functions of one or more modules in the information acquisition module 901, the creation module 902, the configuration module 903, and the uninstallation module 904.

[0163] In some possible implementations, the BMC memory 306 of one or more BMCs 321 in the BMC cluster may also store partial instructions for executing the hardware management method based on virtualization technology. In other words, the combination of one or more BMCs 321 can jointly execute instructions for executing the hardware management method based on virtualization technology.

[0164] This application provides a computing device, see below Fig.10 , Fig.10 806 and 809. The embodiment of the present application provides a structural diagram of a computing device of a hardware management method based on virtualization technology. The computing device 32 includes: a bus 807, a processor 322, a memory 806 and a communication interface 809. The processor 322, the memory 806 and the communication interface 809 communicate with each other through the bus 807. It should be understood that the present application does not limit the number of processors, memories and BMCs in the computing device 32.

[0165] The bus 807 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.10 The bus 807 may include a path for transmitting information between various components of the computing device 32 (eg, the memory 806, the processor 322, and the communication interface 809).

[0166] The processor 322 may include any one or more processors such as a CPU, a microprocessor (MP), or a digital signal processor (DSP).

[0167] The memory 806 may include a volatile memory, such as a random access memory (RAM). The processor 322 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0168] The communication interface 809 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 32 and other devices or a communication network.

[0169] The operation and maintenance system 10 can obtain the status information of the processor 322 through the virtual BMC 21 running on the BMC 321, thereby realizing the operation and maintenance management of the processor.

[0170] The embodiment of the present application also provides a computing device cluster, which includes at least one computing device, each computing device includes at least one BMC and at least one processor. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center.

[0171] See below Fig.11 , Fig.11 Schematic diagram of a computing device cluster of a hardware management method based on virtualization technology in an embodiment of the present application. Fig.11 As shown, the computing device cluster includes at least one computing device 32, and the BMC 321 in one or more computing devices 32 in the computing device cluster can store and run the same hardware management platform and execute instructions of the hardware management method based on virtualization technology.

[0172] In some possible implementations, the BMC 321 in one or more computing devices 32 in the computing device cluster may also be used to execute some instructions of the hardware management method based on virtualization technology. In other words, any combination of the BMC 321 in one or more computing devices 32 may be used together to execute instructions of the hardware management method based on virtualization technology.

[0173] The operation and maintenance system 10 can obtain the status information of the processor 322 through the virtual BMC 21 running on the BMC 321 in the computing device 32, thereby realizing the operation and maintenance management of the processor.

[0174] The embodiment of the present application also provides a computer program product including instructions. The computer program product may be a software or program product including instructions that can be run on a computing device or stored in any available medium. When the computer program product is run on at least one computer device, the at least one computer device executes the above-mentioned hardware management method applied to the hardware management platform for executing virtualization technology.

[0175] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by the computing device or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute the above-mentioned hardware management method applied to the hardware management platform for executing virtualization technology.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

[0177] Those skilled in the art can clearly understand that the specific working process of the system, device or unit described above can refer to the corresponding process in the aforementioned hardware management platform embodiment, which will not be repeated here.

[0178] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A hardware management method based on virtualization technology, characterized in that: The method is applied to a hardware management platform, the hardware management platform runs on a baseboard management controller BMC, the BMC is connected to a processor provided on a computing device, wherein the hardware management platform is connected to an operation and maintenance system, the operation and maintenance system is used to obtain status information of the computing device, and the method includes: Obtaining basic information of the processor; Creating a virtual BMC on the BMC according to the basic information, wherein the virtual BMC is used to manage the processor; The virtual BMC is configured so that a connection established between the virtual BMC and the operation and maintenance system is independent of connections established between other virtual BMCs and the operation and maintenance system, wherein the operation and maintenance system is used to obtain the status information of the first processor through the virtual BMC.

2. The method according to claim 1, characterized in that The basic information also includes management channel information of the processor, and the step of creating the virtual BMC according to the basic information includes: A virtual BMC for managing the processor through the management channel is created according to the basic information.

3. The method according to claim 1 or 2, characterized in that: The hardware management platform is connected to the operation and maintenance system via an Internet Protocol IP address, the BMC stores the IP address, and the configuring the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system includes: The virtual BMC is configured so that the virtual BMC is bound to the IP address, and the virtual BMC is also used to connect to the operation and maintenance system through the IP address.

4. The method according to any one of claims 1 to 3, characterized in that: The basic information also includes the type of the processor, and the BMC also stores a configuration file. After creating the virtual BMC, the method further includes: The virtual BMC is configured according to the configuration file so that the virtual BMC loads the configuration file, wherein the configuration file is used to instruct the virtual BMC to manage a function of a corresponding type of processor.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: obtaining uninstall information to remove the processor from the computing device; The virtual BMC is uninstalled from the BMC according to the information.

6. The method according to any one of claims 1 to 5, characterized in that: After configuring the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of connections established between other virtual BMCs and the operation and maintenance system, the method further includes: The virtual BMC is configured so that the virtual BMC registers the processor with the operation and maintenance system.

7. A hardware management platform, characterized in that: The hardware management platform runs on a baseboard management controller BMC, and the BMC is connected to a processor provided on a computing device, wherein the hardware management platform is connected to an operation and maintenance system, and the operation and maintenance system is used to obtain status information of the computing device, and the hardware management platform includes: An information acquisition module, the information acquisition module is used to acquire basic information of the processor; A creation module, wherein the creation module is used to create a virtual BMC on the BMC according to the basic information, and the virtual BMC is used to manage the processor; A configuration module, wherein the configuration module is used to configure the virtual BMC so that the connection established between the virtual BMC and the operation and maintenance system is independent of the connection established between other virtual BMCs and the operation and maintenance system, wherein the operation and maintenance system is used to obtain the status information of the first processor through the virtual BMC.

8. The hardware management platform according to claim 7, characterized in that: The basic information also includes the management channel information of the processor, then: The creation module is specifically configured to create, according to the basic information, a virtual BMC for managing the processor through the management channel.

9. The hardware management platform according to claim 7 or 8, characterized in that: The hardware management platform is connected to the operation and maintenance system via an Internet Protocol IP address, and the BMC stores the IP address. Then: The configuration module is specifically used to configure the virtual BMC so that the virtual BMC is bound to the IP address, and the virtual BMC is also used to connect to the operation and maintenance system through the IP address.

10. The hardware management platform according to any one of claims 7 to 9, characterized in that: The basic information also includes the type of the processor, the BMC also stores a configuration file, and the configuration module is further used to: The virtual BMC is configured according to the configuration file so that the virtual BMC loads the configuration file, wherein the configuration file is used to instruct the virtual BMC to manage a function of a corresponding type of processor.

11. The hardware management platform according to any one of claims 7 to 10, characterized in that: The information acquisition module is further used to obtain uninstall information for removing the processor from the computing device; The hardware management platform also includes: An uninstallation module is used to uninstall the virtual BMC from the BMC according to the information.

12. The hardware management platform according to any one of claims 7 to 11, characterized in that: The configuration module is also used to: The virtual BMC is configured so that the virtual BMC registers the processor with the operation and maintenance system.

13. A BMC cluster, characterized in that: Includes at least one BMC, each BMC includes a BMC processor and a BMC memory; The BMC processor of the at least one BMC is used to execute instructions stored in the BMC memory of the at least one BMC, so that the BMC cluster executes the hardware management platform according to any one of claims 1 to 6.

14. A computing device cluster, characterized in that: comprising at least one computing device, each computing device comprising at least one processor and at least one BMC; The at least one BMC is used to execute the hardware management method according to any one of claims 1 to 6, so that the operation and maintenance system obtains the status information of at least one processor in each computing device through the virtual BMC running in the at least one BMC.

15. A computer-readable storage medium, characterized in that: The method comprises computer program instructions. When the computer program instructions are executed by a computing device cluster, the computing device cluster executes the hardware management platform according to any one of claims 1 to 6.

16. A computer program product comprising instructions, characterized in that When the instructions are executed by a computing device cluster, the computing device cluster executes the hardware management platform according to any one of claims 1 to 6.

Images