Server management system and method

Abstract

The invention discloses a server management system and method, and relates to the technical field of server management. The server node module comprises a first server node and a second server node which are the same, the first end of the first server node is connected with the substrate management controller, and the second end of the first server node is connected with the second end of the second server node; and the baseboard management controller is used for managing one of the first server node and the second server node based on the server management instruction. According to the invention, the server management cost is reduced.

Description

Technical Field

[0001] This application relates to the field of server management technology, and in particular to a server management system and method. Background Technology

[0002] As servers, such as CPUs (Central Processing Units), become increasingly prevalent in various application scenarios, users are placing higher demands on server management methods.

[0003] Traditional server management uses a single BMC (Baseboard Management Controller) to manage a single server node, which is controlled by a single switch. This method has a significant drawback: a single BMC can only manage one server node, resulting in high server management costs.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a server management system and method, which aims to solve the technical problem of high server management costs.

[0006] To achieve the above objectives, this application provides a server management system, which includes:

[0007] Baseboard management controller;

[0008] The server node module includes an identical first server node and a second server node. A first end of the first server node is connected to the baseboard management controller, and a second end of the first server node is connected to a second end of the second server node. The baseboard management controller is used to manage one of the first server node and the second server node based on server management instructions.

[0009] In one embodiment, the baseboard management controller includes a first communication terminal and a second communication terminal, and the first server node includes:

[0010] A first switching unit, wherein a first end of the first switching unit is connected to the first communication end;

[0011] A first programmable logic device, wherein a first end of the first programmable logic device is connected to a second communication end, and a second end of the first programmable logic device is connected to a second end of the switching unit;

[0012] A first connector, wherein a first end of the first connector is connected to the second communication terminal and the third end of the first switching unit, and a second end of the first connector is connected to the second end of the second connector in the second server node, wherein the first connector and the second connector are connected by a cascaded cable;

[0013] A first internal communication unit is connected to the fourth terminal and the fifth terminal of the first switching unit.

[0014] In one embodiment, the second end of the first programmable logic device includes a first control terminal, a second control terminal, and a third control terminal; the first end of the first connector includes a first connection terminal, a second connection terminal, and a third connection terminal; wherein the third connection terminal is connected to the second communication terminal; and the first switching unit includes:

[0015] A first switching switch, wherein the signal terminal of the first switching switch is connected to the first terminal of the first internal communication unit, the control terminal of the first switching switch is connected to the first control terminal, and the first selection terminal of the first switching switch is connected to the second terminal of the first internal communication unit;

[0016] A second switch, wherein the signal terminal of the second switch is connected to the second selection terminal of the first switch, the control terminal of the second switch is connected to the second control terminal, and the first selection terminal of the second switch is connected to the first connection terminal;

[0017] The third switch has a signal terminal connected to the first communication terminal, a control terminal connected to the third control terminal, a first selection terminal connected to the second connection terminal, and a second selection terminal connected to the second selection terminal of the second switch.

[0018] In one embodiment, the first internal communication unit includes:

[0019] The first serial peripheral interface memory is connected to the signal terminal of the first switching switch.

[0020] A first central processing unit is connected to the first selection terminal of the first switch.

[0021] In one embodiment, the server management system further includes:

[0022] The motherboard connector is used to connect the baseboard management controller to the first server node.

[0023] Furthermore, to achieve the above objectives, this application also provides a server management method, which is applied to the aforementioned server management system. The server management method includes:

[0024] Obtain server management instructions and determine the target service node and target service type in the server management instructions;

[0025] One of the first server node and the second server node is managed according to the target service node and the target service type.

[0026] In one embodiment, the target service type includes firmware access service, and the step of managing one of the first server node and the second server node according to the target service node and the target service type includes:

[0027] When the target service node is a first server node, the first switching unit in the first server node is controlled to be in a first conducting state based on the firmware access service. The first conducting state includes the first switching unit being connected to the first serial peripheral interface memory and the first central processing unit in the first server node.

[0028] When the target service node is the second server node, and the first switching unit is in a preset on state, the second switching unit in the second server node is controlled to be in a second on state based on the firmware access service. The second on state includes the second switching unit being connected to the second serial peripheral interface memory and the second central processing unit in the second server node.

[0029] In one embodiment, the target service type includes a firmware update service, and the step of managing one of the first server node and the second server node according to the target service node and the target service type includes:

[0030] When the target service node is the first server node, the first switching unit in the first server node is controlled to be in a third conduction state based on the firmware update service. The third conduction state includes the first switching unit being in a state of connecting the first serial peripheral interface memory and the baseboard management controller in the first server node.

[0031] When the target service node is the second server node, and the first switching unit is in a preset on state, the second switching unit in the second server node is controlled to be in a fourth on state based on the firmware update service. The fourth on state includes the second switching unit being in a state where it is connected to the first serial peripheral interface memory and the baseboard management controller in the second server node.

[0032] In one embodiment, after the target service node is a second server node, the server management method further includes:

[0033] Based on the target service type, the first switching unit in the first server node is controlled to be in a preset on state, wherein the preset on state includes the first switching unit being in a state where the first connector in the first server node is connected to the baseboard management controller.

[0034] In one embodiment, the server management method further includes:

[0035] Obtain the connection status of the baseboard management controller;

[0036] When the connection status is a dual-server node connection status, the step of obtaining server management instructions is executed;

[0037] When the connection status is a single server node connection status, server management instructions are obtained, and the server node is managed based on the server management instructions.

[0038] This application proposes a server management system, including a baseboard management controller (BMC) and a server node module. The server node module includes an identical first server node and a second server node. A first end of the first server node is connected to the BMC, and a second end of the first server node is connected to a second end of the second server node. The BMC is used to manage one of the first and second server nodes based on server management commands. This server management system connects the first end of the first server node to the BMC via the BMC, and simultaneously connects the second end of the first server node to the second end of the second server node. This allows the BMC to manage one of the first and second server nodes based on server management commands, avoiding the problem in the prior art where one BMC can only manage one server node. This server management system can establish a connection between the second and first server nodes, and thus manage one of the first and second server nodes based on server management commands, thereby reducing server management costs. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of one module of the server management system of this application;

[0040] Figure 2 This is a schematic diagram of the framework connection of the first embodiment of the server management system of this application;

[0041] Figure 3 This is a circuit connection diagram of the first embodiment of the server management system of this application;

[0042] Figure 4 This is a schematic diagram of the framework connection of the second embodiment of the server management system of this application;

[0043] Figure 5 This is a circuit connection diagram of the second embodiment of the server management system of this application;

[0044] Figure 6 This is a flowchart illustrating the server management method of this application.

[0045] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0046] Explanation of icon numbers:

[0047] 100. Baseboard Management Controller; 200. Server Node Module; 10. First Server Node; 20. Second Server Node; 11. First Switching Unit; 12. First Programmable Logic Controller; 13. First Connector; 14. First Internal Communication Unit; S. First Communication Terminal; I. Second Communication Terminal; 110. Management Motherboard; 120. First Motherboard; 130. First Motherboard Connector; 141. First Serial Peripheral Interface Memory; 142. First Central Processing Unit; X1. First Switching Switch; X2. Second Switching Switch; X3. Third Switching Switch; SEL1. First Control Terminal; SEL2. Second Control Terminal; SEL3. Third control terminal; D, signal terminal of the switch; C, control terminal of the switch; S1, first selection terminal of the switch; S2, second selection terminal of the switch; 21, second switch unit; 22, second programmable logic device; 23, second connector; 24, second internal communication unit; 140, second motherboard; 150, second motherboard connector; 241, second serial peripheral interface memory; 242, second central processing unit; X4, fourth switch; X5, fifth switch; X6, sixth switch; SEL4, fourth control terminal; SEL5, fifth control terminal; SEL6, sixth control terminal; 160, cascade cable. Detailed Implementation

[0048] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0049] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0050] With the improvement of CPU performance and the continued growth in demand for rapid market launch of server products, users have also placed higher demands on server management methods. Because each server node's CPU has its own BIOS (the first firmware program when the computer starts up), SPI (Serial Peripheral Interface), and Flash (non-volatile memory), each server node needs to be managed by a BMC. That is, a single BMC cannot manage the BIOS SPI Flash on two nodes simultaneously, thus causing a waste of BMC resources.

[0051] Therefore, based on the shortcomings of the above server management solutions, the server management system of this application is proposed. The main solution of the embodiments of this application is: the first end of the first server node is connected through the baseboard management controller, and the second end of the first server node is connected to the second end of the second server node. Then, the baseboard management controller can be used to manage one of the first server node and the second server node based on server management instructions, so as to avoid the problem that one BMC can only manage one server node in the prior art. This server management system can establish a connection between the second server node and the first server node, and then manage one of the first server node and the second server node based on server management instructions, thereby reducing the cost of server management.

[0052] Based on this, embodiments of this application provide a server management system, referring to... Figure 1 , Figure 1 This is a schematic diagram of one module of the server management system of this application.

[0053] Reference Figure 1 This application provides a server management system, which includes:

[0054] Baseboard management controller 100;

[0055] Server node module 200 includes an identical first server node 10 and second server node 20. A first end of the first server node 10 is connected to a baseboard management controller 100, and a second end of the first server node 10 is connected to a second end of the second server node 20. The baseboard management controller 100 is used to manage one of the first server node 10 and the second server node 20 based on server management instructions.

[0056] In this embodiment, the baseboard management controller 100 connects to the first end of the first server node 10 and simultaneously connects the second end of the first server node 10 to the second end of the second server node 20, thereby enabling the management of both server nodes. It is worth noting that the first server node 10 and the second server node 20 are identical; that is, their internal composition and connection relationships are the same. The only difference between the two server nodes lies in their external connection relationships. When managing the two server nodes, one of the first server node 10 and the second server node 20 can be managed directly based on server management instructions. For example, if the server management instructions require management of the first server node 10, then management of the first server node 10 can be performed directly. Similarly, if the server management instructions require management of the second server node 20, then management of the second server node 20 can be achieved through the connection relationship of the first server node 10. Server management instructions refer to instructions for managing server nodes and may include the managed object and instructions for controlling the connection relationships between the lines in the first server node 10 and the second server node 20. Therefore, based on the connection between the first server node 10 and the second server node 20, and the server management commands, one of the first server node 10 and the second server node 20 can be managed, thus achieving the management of both server nodes and reducing the cost of server management. It is worth noting that the first server node 10 and the second server node 20 both refer to a server module, which contains its own components and the connections between them, and do not refer to a single point.

[0057] In this embodiment, a server management system is provided, including a baseboard management controller 100 and a server node module 200. The server node module 200 includes an identical first server node 10 and a second server node 20. A first end of the first server node 10 is connected to the baseboard management controller 100, and a second end of the first server node 10 is connected to a second end of the second server node 20. The baseboard management controller 100 is used to manage one of the first server node 10 and the second server node 20 based on server management instructions. This server management system connects the first end of the first server node 10 to the baseboard management controller 100, and simultaneously connects the second end of the first server node 10 to the second end of the second server node 20. Therefore, the baseboard management controller 100 can manage one of the first server node 10 and the second server node 20 based on server management instructions, avoiding the problem in the prior art where one baseboard management controller can only manage one server node. This server management system can establish a connection between the second server node 20 and the first server node 10, and thus manage one of the first server node 10 and the second server node 20 based on server management instructions, thereby reducing the cost of server management.

[0058] Furthermore, based on the first embodiment of this application described above, a second embodiment of the server management system of this application is proposed. In this embodiment, reference is made to... Figure 2 , Figure 2 This is a schematic diagram of the framework connection of the first embodiment of the server management system of this application. The baseboard management controller 100 includes a first communication terminal S and a second communication terminal I, and the first server node 10 includes:

[0059] The first switching unit 11, the first end of the first switching unit 11 is connected to the first communication terminal S;

[0060] A first programmable logic device 12, the first end of which is connected to a second communication terminal I, and the second end of which is connected to a switching unit 11.

[0061] The first connector 13 has a first end connected to the second communication terminal I and the third end of the first switching unit 11, and a second end connected to the second end of the second connector 23 in the second server node 20. The first connector 13 and the second connector 23 are connected by a cascade cable 160.

[0062] The first internal communication unit 14 is connected to the fourth terminal and the fifth terminal of the first switching unit 11.

[0063] In one embodiment, the server management system further includes:

[0064] The motherboard connector and the baseboard management controller 100 are connected to the first server node 10 via the motherboard connector.

[0065] In this embodiment, the substrate management controller 100 includes a first communication terminal S and a second communication terminal I. Taking SPI communication and I2C (Inter-Integrated Circuit) communication as an example, the first communication terminal S represents the port using SPI communication, and the second communication terminal I represents the port using I2C communication. The first server node 10 includes a first switching unit 11, a first programmable logic device 12, such as a CPLD (Complex Programmable Logic Device), a first connector 13, and a first internal communication unit 14. Since the first connector 13 and the second connector 23 are connected via a cascade cable 160, the connection between the first server node 10 and the second server node 20 is realized, enabling the management of both server nodes and reducing the cost of server management. On one hand, the first internal communication unit 14 in the first server node 10 can realize internal BIOS communication by controlling the first switching unit 11. The first programmable logic device 12 is used to control the state of the first switching unit 11, thereby enabling the internal circuit of the first internal communication unit 14 to be connected. It can also control the first switching unit 11 to enable the first internal communication unit 14 to communicate with the baseboard management controller 100, so that the baseboard management controller 100 can manage the first server node 10. On the other hand, the first switching unit 11 can also be controlled to connect the baseboard management controller 100 to the first connector 13, so that the baseboard management controller 100 can manage the second server node 20 connected to the first connector 13, thereby realizing the management of the two server nodes.

[0066] It is worth noting that the first server node 10 can be set on the first motherboard 120, with a first motherboard connector 130 to connect to the baseboard management controller 100. The second server node 20 can be set on the second motherboard 140, with a second motherboard connector 150 to connect to the baseboard management controller 100. However, to reduce the internal logic control and software design costs of the baseboard management controller 100, only one motherboard connector can be used to connect to the baseboard management controller 100, leaving the other motherboard connector unconnected. Communication between the motherboards is achieved through a cascading cable 160 between the first connector 13 and the second connector 23. The cascading cable 160 can be any cable used for normal data transmission and is not limited here. The baseboard management controller 100 can also be set on the management motherboard 110, and then connect to the first motherboard 120 through the first motherboard connector 130. It is worth noting that the management motherboard 110 can be a DC-SCM model, and the motherboard connector can be a DC-SCM connector.

[0067] Furthermore, based on the first and / or second embodiments of this application described above, a third embodiment of the server management system of this application is proposed, with reference to... Figure 3 , Figure 3 This is a circuit connection diagram of the first embodiment of the server management system of this application. The second terminal of the first programmable logic device 12 includes a first control terminal SEL1, a second control terminal SEL2, and a third control terminal SEL3. The first terminal of the first connector 13 includes a first connection terminal, a second connection terminal, and a third connection terminal, wherein the third connection terminal is connected to the second communication terminal I. The first switching unit 11 includes:

[0068] The first switch X1 has its signal terminal connected to the first terminal of the first internal communication unit 14, its control terminal connected to the first control terminal SEL1, and its first selection terminal connected to the second terminal of the first internal communication unit 14.

[0069] The second switch X2 has its signal terminal connected to the second selection terminal of the first switch X1, its control terminal connected to the second control terminal SEL2, and its first selection terminal connected to the first connection terminal.

[0070] The third switch X3 has its signal terminal connected to the first communication terminal S, its control terminal connected to the third control terminal SEL3, its first selection terminal connected to the second connection terminal, and its second selection terminal connected to the second selection terminal of the second switch X2.

[0071] Furthermore, the first internal communication unit 14 includes:

[0072] The first serial peripheral interface memory 141 is connected to the signal terminal of the first switching switch X1.

[0073] The first central processing unit 142 is connected to the first selection terminal of the first switching switch X1.

[0074] In this embodiment, the second end of the first programmable logic device 12 includes a first control terminal SEL1, a second control terminal SEL2, and a third control terminal SEL3. The first end of the first connector 13 includes a first connection terminal, a second connection terminal, and a third connection terminal, wherein the third connection terminal is connected to the second communication terminal I. The first switching unit 11 includes three switching switches, and... Figure 3 The connection relationships are established. In the switch, the signal terminal D, the second selection terminal S2, and the first selection terminal S1 of the switch all represent signal input or output ports. The control terminal C of the switch represents the port controlling the line connecting the signal terminal D, the second selection terminal S2, and the first selection terminal S1 of the switch. If the control terminal C of the switch inputs 1, then the signal terminal D of the switch is connected to the second selection terminal S2 and disconnected from the first selection terminal S1; if the control terminal C of the switch inputs 0, then the signal terminal D of the switch is disconnected from the second selection terminal S2 and connected to the first selection terminal S1. The functions and connection relationships of each component in the first server node 10 are explained below:

[0075] BMC: The management core, which can update the BIOS via the SPI bus, configure the first CPLD via the I2C bus, and connect to the internal circuitry of the motherboard via the DC-SCM connector. The BMC's I2C bus is connected to the first CPLD and the first connector 13 on the motherboard, and the BMC's SPI is connected to the D port of the third toggle switch X3;

[0076] First CPLD: integrates SEL1 / SEL2 / SEL3 gating control logic, switches the control objects of the SPI bus (local CPU and SPI Flash / cascaded first connector 13) to the control ports C of the first switch X1, the second switch X2 and the third switch X3 respectively, and receives control commands from the BMC through the I2C bus;

[0077] DC-SCM connector: physical interface, supports SPI, I2C and other signal transmission, compatible with single motherboard and multi-motherboard cascading;

[0078] Cascading Cable 160: Adopts a cross-wiring design to enable SPI signal pass-through between multiple motherboards, expand the management scope, and realize the physical connection between two motherboards.

[0079] Therefore, based on the above connection relationships, when management is required, the connection relationships can be realized by controlling the three switches in the first switching unit 11. For example, if the CPU needs to access the SPI Flash, the first switching switch X1 can be controlled to connect the first serial peripheral interface memory 141 (SPI Flash) and the first central processing unit 142 (CPU). If the board management controller 100 needs to update the firmware in the first serial peripheral interface memory 141, the three switches in the first switching unit 11 can be controlled to connect the first serial peripheral interface memory 141 with the board management controller 100. Further, refer to... Figure 4 , Figure 4 This is a schematic diagram of the framework connection of the second embodiment of the server management system of this application. It shows the internal composition and connection relationship of the second server node 20. As can be seen from the diagram, the internal composition and connection relationship of the second server node 20 are the same as those of the first server node 10, except that it is not directly connected to the baseboard management controller 100. Instead, it is connected to the first server node 10 via the second connector 23 and then to the baseboard management controller 100. Therefore, the second server node 20 can be managed based on the internal circuit control of the first server node 10. (Refer to...) Figure 5 , Figure 5 This is a circuit connection diagram of the second embodiment of the server management system of this application. The management of the second server node 20 is based on three switching switches inside the first server node 10 to enable the baseboard management controller 100 to connect to the second server node 20 through the first connector 13, and then access the CPU and SPI Flash. The fourth switching switch X4 can be further controlled to connect the second serial peripheral interface memory 241 (SPI Flash) and the second central processing unit 242 (CPU). If it is necessary to enable the baseboard management controller 100 to update the firmware in the second serial peripheral interface memory 141, the three switches in the second switching switch unit 21 can be further controlled to enable the second serial peripheral interface memory 141 to connect to the baseboard management controller 100. Based on internal communication and control, the baseboard management controller 100 can manage one of the first server node 10 and the second server node 20, thereby reducing the cost of server management.

[0080] Based on the above embodiments of the server management system of this application, a first embodiment of the server management method of this application is proposed. In this embodiment, the server management method is applied to the above-described server management system, referring to... Figure 6 , Figure 6This is a flowchart illustrating the server management method of this application. The server management method includes:

[0081] Step S10: Obtain server management instructions and determine the target service node and target service type in the server management instructions;

[0082] Step S20: Manage one of the first server node 10 and the second server node 20 according to the target service node and the target service type.

[0083] In this embodiment, the server management system based on the above embodiments can manage two server nodes. By acquiring server management instructions (which can be user input or triggered based on internal triggering conditions), the target service node and target service type in the server management instructions are determined. The target service node refers to the node that needs to be managed, which can be one of the first server node 10 and the second server node 20. The target service type refers to the type of service that needs to be managed, such as updating the BIOS or reading the BIOS. By determining the target service node and target service type, one of the first server node 10 and the second server node 20 can be precisely managed. Based on the server management system of the above embodiments, and the target service node and target service type, one of the first server node 10 and the second server node 20 can be managed, thus completing the management of the two server nodes by the baseboard management controller 100 and reducing the cost of server management.

[0084] Based on the first embodiment of the server management method, a second embodiment of the server management method of this application is proposed. In one embodiment, the target service type includes firmware access service. The step of managing one of the first server node 10 and the second server node 20 according to the target service node and the target service type includes:

[0085] Step S201: When the target service node is the first server node 10, the first switching unit 11 in the first server node 10 is controlled to be in a first conducting state based on the firmware access service. The first conducting state includes the first switching unit 11 being connected to the first serial peripheral interface memory 141 and the first central processing unit 142 in the first server node 10.

[0086] Step S202: When the target service node is the second server node 20, and the first switching unit 11 is in a preset conducting state, the second switching unit 21 in the second server node 20 is in a second conducting state based on the firmware access service control. The second conducting state includes the second switching unit 21 being connected to the second serial peripheral interface memory 241 and the second central processing unit 242 in the second server node 20.

[0087] In this embodiment, the target service type includes firmware access service, which refers to the service where the first central processing unit 142 accesses the BIOS and other firmware in the first serial peripheral interface memory 141 (or the second central processing unit 242 accesses the BIOS and other firmware in the second serial peripheral interface memory 241). In this case, processing is divided into two scenarios: the target service node is the first server node 10 or the target service node is the second server node 20. When the target service node is the first server node 10, the first switching unit 11 in the first server node 10 is controlled to be in a first conducting state based on the firmware access service. The first conducting state includes the first switching unit 11 being connected to the first serial peripheral interface memory 141 and the first central processing unit 142 in the first server node 10. That is, the first serial peripheral interface memory 141 and the first central processing unit 142 are connected to enable firmware access. See also... Figure 5 The BMC connects to the first CPLD via a DC-SCM connector. The first CPLD controls the first switching switch X1 via SEL1, connecting the D port of MUX1 to the S1 port, thus enabling the first CPU to access the first SPI Flash (i.e., the first serial peripheral interface memory 141) and read / write the BIOS flash normally. At this time, the BMC controls the output SEL1 of the first CPLD via the I2C bus, switching the first switching switch X1 to the first CPU path, allowing the first CPU to access the SPI Flash storing the BIOS firmware. When the target service node is the second server node 20, the first switching switch unit 11 is first controlled to be in a preset on state. Then, based on the firmware access service, the second switching switch unit 21 in the second server node 20 is controlled to be in a second on state. This second on state includes the second switching switch unit 21 being connected to the second serial peripheral interface memory 241 and the second central processing unit 242 in the second server node 20. That is, the second serial peripheral interface memory 241 and the second central processing unit 242 are connected to enable firmware access. (See reference...) Figure 5The BMC first sends a command to the first CPLD on the first motherboard 120 via the I2C bus, causing the first CPLD to control the third switch X3 via SEL3. This connects the D port of the third switch X3 to the S1 port, and the S1 port of the third switch X3 to the first connector 13 on the first motherboard 120. At this time, the first switch unit 11 is in a preset on state. Then, the second on state control of the second switch unit 21 is performed. Because the first motherboard 120 is connected to the second connector 23 on the second motherboard 140 via a cable, a command is sent to the second CPLD via I2C to connect the D port of the fourth switch X4 to the S1 port. This approach returns SPI access control to the second CPU of the second motherboard 140, enabling it to access the SPI Flash normally and complete firmware access service management for the two server nodes, thereby significantly reducing server management costs.

[0088] In one embodiment, the target service type includes a firmware update service. The step of managing one of the first server node 10 and the second server node 20 according to the target service node and the target service type includes:

[0089] Step S211: When the target service node is the first server node 10, the first switching unit 11 in the first server node 10 is controlled to be in a third conduction state based on the firmware update service. The third conduction state includes the first switching unit 11 being in a state that connects the first serial peripheral interface memory 141 in the first server node 10 to the baseboard management controller 100.

[0090] In step S212, when the target service node is the second server node 20, the second switching unit 21 in the second server node 20 is controlled to be in a fourth conducting state based on the firmware update service when the first switching unit 11 is in a preset conducting state. The fourth conducting state includes the second switching unit 21 being in a state that connects the first serial peripheral interface memory 241 in the second server node 20 to the baseboard management controller 100.

[0091] In this embodiment, the target service type includes firmware update service, that is, the service of updating the firmware of the first serial peripheral interface memory 141 using the baseboard management controller 100 (or the service of updating the firmware of the second serial peripheral interface memory 241 using the baseboard management controller 100). This will be handled in two cases: the target service node is the first server node 10 or the target service node is the second server node 20. When the target service node is the first server node 10, the first switching unit 11 in the first server node 10 is controlled to be in a third conducting state based on the firmware update service. The third conducting state includes the first switching unit 11 being in a state connecting the first serial peripheral interface memory 141 in the first server node 10 to the baseboard management controller 100. That is, at this time, the first serial peripheral interface memory 141 and the baseboard management controller 100 are connected to realize firmware update. See also... Figure 5 When the BMC needs to remotely update the BIOS firmware, the BMC sends a command to the first CPLD via I2C. The first CPLD controls the first switch X1, the second switch X2, and the third switch X3 by controlling SEL1, SEL2, and SEL3, enabling the BMC to access the SPI flash. Specifically, the first CPLD controls SEL1, connecting the D port of the first switch X1 to the S2 port; the first CPLD controls SEL2, connecting the D port of the second switch X2 to the S2 port; and the first CPLD controls SEL3, connecting the D port of the third switch X3 to the S2 port. This transfers SPI flash access to the BMC, allowing the BMC to perform BIOS firmware updates. Of course, the switches can also be controlled in other situations, such as when the firmware in the first serial peripheral interface memory 141 is accessed by the second central processing unit 242; in such cases, only the switches need to be controlled. When the target service node is the second server node 20, and the first switching unit 11 is in a preset conducting state (i.e., the conducting relationship mentioned in the above embodiment), the second switching unit 21 in the second server node 20 is controlled to be in a fourth conducting state based on the firmware update service. The fourth conducting state includes the second switching unit 21 being connected to the first serial peripheral interface memory 241 and the baseboard management controller 100 in the second server node 20. That is, at this time, the second serial peripheral interface memory 241 and the baseboard management controller 100 are connected to achieve firmware updates. See also... Figure 5When the BMC needs to update the BIOS SPI Flash on the second motherboard 140, the BMC first sends a command to the first CPLD on the first motherboard 120 via the I2C bus. This command causes the first CPLD to control the third switch X3 via SEL3, connecting the D port of the third switch X3 to the S1 port. The S1 port of the third switch X3 is then connected to the first connector 13 on the first motherboard 120, and via a cable, to the second connector 23 on the second motherboard 140. Next, the BMC commands the second CPLD on the second motherboard 140 via the I2C bus, causing its SEL4 to control the fourth switch X4, connecting the S2 port of the fourth switch X4 to the D port. Similarly, SEL5 controls the fifth switch X5, connecting the S1 port of the fifth switch X5 to the D port. This physically connects the BIOS SPI Flash on the second motherboard 140 to the BMC's SPI interface, enabling the BMC to remotely update the BIOS SPI Flash on the second motherboard 140, thus reducing management costs during firmware update services.

[0092] In one embodiment, after the target service node is the second server node 20, the server management method further includes:

[0093] Step S30: Based on the target service type (which may be firmware access service or firmware update service), control the first switching unit 11 in the first server node 10 to be in a preset conducting state. The preset conducting state includes the first switching unit 11 being in a state where the first connector 13 in the first server node 10 is connected to the baseboard management controller 100.

[0094] In this embodiment, since the second server node 20 is indirectly controlled by the first server node 10, when managing the second server node 20, it is necessary to control the switching switch inside the first server node 10. That is, the firmware access service controls the first switching switch unit 11 in the first server node 10 to be in a preset conducting state. The preset conducting state includes the first switching switch unit 11 being connected to the first connector 13 in the first server node 10 and the baseboard management controller 100. Referring to Figure 3, the preset conducting state is that the BMC first sends a command to the first CPLD on the first motherboard 120 through the I2C bus, so that the first CPLD controls the third switching switch X3 through SEL3, so that the D port of the third switching switch X3 is connected to the S1 port, and the S1 port of the third switching switch X3 is connected to the first connector 13 on the first motherboard 120. At this time, the first switching switch unit 11 is in the preset conducting state to reduce subsequent management costs and ensure the accuracy of subsequent control of the second server node 20.

[0095] Based on the first and / or second embodiments of the server management method, a third embodiment of the server management method of this application is proposed. In one embodiment, the server management method further includes:

[0096] Step a, obtain the connection status of the baseboard management controller 100;

[0097] Step b: When the connection status is a dual-server node connection status, execute the step of obtaining server management instructions;

[0098] Step c: When the connection status is a single server node connection status, obtain server management instructions and manage the server node based on the server management instructions.

[0099] In this embodiment, the connection status of the baseboard management controller 100 refers to the connection status of the first connector 13 on the first server node 10, which can be connected or not connected. That is, the BMC can determine whether the second motherboard 140 exists through I2C. If the BMC determines that the second motherboard 140 does not exist through the I2C bus, the BMC enters the single-node management mode. If the BMC determines that the second motherboard 140 exists through the I2C bus, the BMC enters the two-node management mode. The connection status refers to the connection status of the first connector 13. Therefore, when the connection status is a dual-server node connection status, the step of obtaining server management instructions can be executed, that is, the steps of the above embodiment can be executed. When the connection status is a single-server node connection status, server management instructions will be obtained, and the server node will be managed based on the server management instructions. At this time, the server management instructions are only for target service types. Therefore, the first switching unit 11 in the first server node 10 is in the first conducting state for firmware access service control. The first conducting state includes the state in which the first switching unit 11 is connected to the first serial peripheral interface memory 141 and the first central processing unit 142 in the first server node 10. At this time, the first serial peripheral interface memory 141 and the first central processing unit 142 are connected to enable firmware access. For firmware update services, the second switching unit 21 in the second server node 20 is in a fourth conducting state, where the second switching unit 21 connects the first serial peripheral interface memory 241 in the second server node 20 to the baseboard management controller 100. That is, at this time, the second serial peripheral interface memory 241 and the baseboard management controller 100 are connected to enable firmware updates and management of a single server node. It is worth noting that the BMC can also read the status of the two CPLDs via the I2C bus to determine which device currently controls the SPI flash, thus gaining real-time understanding of the management status.

[0100] Furthermore, based on the above embodiments, this application also provides a server, which includes the above-described server management system and a server manager in the server management system (which can be directly controlled by the baseboard management controller) for the above-described server management method.

[0101] In this system, the server management system is connected to the first end of the first server node 10 via a baseboard management controller 100. Simultaneously, the second end of the first server node 10 is connected to the second end of the second server node 20. This allows the baseboard management controller 100 to manage one of the first server node 10 and the second server node 20 based on server management commands. This avoids the problem in existing technologies where one baseboard management controller (BMC) can only manage one server node. This server management system can establish a connection between the second server node 20 and the first server node 10, and thus manage one of the first server node 10 and the second server node 20 based on server management commands, thereby reducing server management costs. Of course, the server may also include other components, which will not be described in detail here.

[0102] The above are merely preferred embodiments of this application and do not limit the scope of the patent application. Any equivalent structural transformations made based on the inventive concept of this application and the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included within the scope of patent protection of this application.

Claims

1. A server management system, characterized in that, The server management system includes: Baseboard management controller; The server node module includes an identical first server node and a second server node. A first end of the first server node is connected to the baseboard management controller, and a second end of the first server node is connected to a second end of the second server node. The baseboard management controller is used to manage one of the first server node and the second server node based on server management instructions.

2. The server management system as described in claim 1, characterized in that, The baseboard management controller includes a first communication terminal and a second communication terminal, and the first server node includes: A first switching unit, wherein a first end of the first switching unit is connected to the first communication end; A first programmable logic device, wherein a first end of the first programmable logic device is connected to a second communication end, and a second end of the first programmable logic device is connected to a second end of the switching unit; A first connector, wherein a first end of the first connector is connected to the second communication terminal and the third end of the first switching unit, and a second end of the first connector is connected to the second end of the second connector in the second server node, wherein the first connector and the second connector are connected by a cascaded cable; A first internal communication unit is connected to the fourth terminal and the fifth terminal of the first switching unit.

3. The server management system as described in claim 2, characterized in that, The second end of the first programmable logic device includes a first control terminal, a second control terminal, and a third control terminal; the first end of the first connector includes a first connection terminal, a second connection terminal, and a third connection terminal, wherein the third connection terminal is connected to the second communication terminal; and the first switching unit includes: A first switching switch, wherein the signal terminal of the first switching switch is connected to the first terminal of the first internal communication unit, the control terminal of the first switching switch is connected to the first control terminal, and the first selection terminal of the first switching switch is connected to the second terminal of the first internal communication unit; A second switch, wherein the signal terminal of the second switch is connected to the second selection terminal of the first switch, the control terminal of the second switch is connected to the second control terminal, and the first selection terminal of the second switch is connected to the first connection terminal; The third switch has a signal terminal connected to the first communication terminal, a control terminal connected to the third control terminal, a first selection terminal connected to the second connection terminal, and a second selection terminal connected to the second selection terminal of the second switch.

4. The server management system as described in claim 3, characterized in that, The first internal communication unit includes: The first serial peripheral interface memory is connected to the signal terminal of the first switching switch. A first central processing unit is connected to the first selection terminal of the first switch.

5. The server management system as described in any one of claims 1 to 4, characterized in that, The server management system also includes: The motherboard connector is used to connect the baseboard management controller to the first server node.

6. A server management method, characterized in that, The server management method is applied to the server management system as described in any one of claims 1 to 5, wherein the server management method includes: Obtain server management instructions and determine the target service node and target service type in the server management instructions; One of the first server node and the second server node is managed according to the target service node and the target service type.

7. The server management method as described in claim 6, characterized in that, The target service type includes firmware access service. The step of managing one of the first server node and the second server node according to the target service node and the target service type includes: When the target service node is a first server node, the first switching unit in the first server node is controlled to be in a first conducting state based on the firmware access service. The first conducting state includes the first switching unit being connected to the first serial peripheral interface memory and the first central processing unit in the first server node. When the target service node is the second server node, and the first switching unit is in a preset on state, the second switching unit in the second server node is controlled to be in a second on state based on the firmware access service. The second on state includes the second switching unit being connected to the second serial peripheral interface memory and the second central processing unit in the second server node.

8. The server management method as described in claim 6, characterized in that, The target service type includes firmware update service. The step of managing one of the first server node and the second server node according to the target service node and the target service type includes: When the target service node is the first server node, the first switching unit in the first server node is controlled to be in a third conduction state based on the firmware update service. The third conduction state includes the first switching unit being in a state of connecting the first serial peripheral interface memory and the baseboard management controller in the first server node. When the target service node is the second server node, and the first switching unit is in a preset on state, the second switching unit in the second server node is controlled to be in a fourth on state based on the firmware update service. The fourth on state includes the second switching unit being in a state where it is connected to the first serial peripheral interface memory and the baseboard management controller in the second server node.

9. The server management method as described in any one of claims 7 or 8, characterized in that, When the target service node is a second server node, the server management method further includes: Based on the target service type, the first switching unit in the first server node is controlled to be in a preset on state, wherein the preset on state includes the first switching unit being in a state where the first connector in the first server node is connected to the baseboard management controller.

10. The server management method as described in claim 6, characterized in that, The server management method also includes: Obtain the connection status of the baseboard management controller; When the connection status is a dual-server node connection status, the step of obtaining server management instructions is executed; When the connection status is a single server node connection status, server management instructions are obtained, and the server node is managed based on the server management instructions.

Images