Cabinet, method for identifying position of electronic device, and electronic device

Abstract

Embodiments of the present application provide a cabinet, a method for identifying the position of an electronic device, and an electronic device. The cabinet includes: a cabinet body, the cabinet body includes a target component that penetrates all rack units, and one electronic device is installed in each rack unit; a plurality of first sensing mechanisms arranged on the target component; a plurality of second sensing mechanisms respectively arranged on the chassis of different electronic devices; a monitoring system that determines a target code based on N target sensing states between the second sensing mechanism on the target electronic device and the corresponding first sensing mechanism, and determines the position of the rack unit associated with the target code as the position of the rack unit of the target electronic device. Through the present application, the problem in the related art that due to differences in the architectures of different cabinets, adding additional positioning devices requires complex processing procedures, resulting in poor flexibility in positioning devices in the cabinet is solved, and thus the effect of improving the flexibility of positioning devices in the cabinet is achieved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computers, and more specifically, to a cabinet, a method for identifying the positions of electronic devices, and an electronic device. Background Art

[0002] With the wide application of new-generation information technologies such as cloud computing, big data, the Internet of Things, and artificial intelligence, the computing power demand of data centers continues to increase. As a key infrastructure to support the training and inference of AI models, the new-generation computing cluster architecture is facing severe performance challenges. The marginal effect of enhancing the performance of a single processor is weakening day by day, almost reaching the physical limit and unable to adapt to the continuous growth demand for computing power of AI (Artificial Intelligence) large models. Against this background, the AI whole-cabinet system redefines the organization and cooperation mode of computing units, and realizes the efficient aggregation of GPU computing nodes through an integrated topology design. Comprehensive optimization is carried out in aspects such as multi-source computing power integration, high-speed advanced interconnection, heat dissipation, high-power density power supply, and whole-cabinet management to provide higher computing power and meet the requirements of cutting-edge large model training and inference.

[0003] However, the whole cabinet lacks a convenient and reliable means of U (Rack Unit, a standard measurement unit) position positioning. U position positioning means that the server nodes in the cabinet can locate their own physical U positions. In related technologies, FRU positioning (Field Replaceable Unit positioning) is added to the cable tray for U position positioning. However, this method will increase the processing complexity and additional production cost of the whole cabinet, require additional space in the whole cabinet, and have poor reliability, and is not applicable to the whole-cabinet system without a cable tray, resulting in poor universality. In related technologies, the management switch is also used to burn and allocate U position information to each node through the front-window RJ45 network cable. However, if the physical position of the device is changed during the node maintenance process, it is difficult to detect and correct the wrong U position of the device. The efficiency of re-detecting the U position information by the management switch after each node replacement is low. Summary of the Invention

[0004] The embodiments of the present application provide a cabinet, a method for identifying the positions of electronic devices, and an electronic device, so as to at least solve the problem in related technologies that due to the differences in the architectures of different cabinets, adding additional positioning devices requires complex processing procedures, resulting in poor flexibility in positioning the devices in the cabinet.

[0005] According to an embodiment of the present application, a cabinet is provided, including: a cabinet body, in which a target component penetrating all rack units is included, and each rack unit is used to install an electronic device; a plurality of first sensing mechanisms arranged on the target component, wherein each first sensing mechanism corresponds to one rack unit, and N first sensing elements are arranged on each first sensing mechanism; a plurality of second sensing mechanisms respectively arranged on the chassis of different electronic devices, wherein N second sensing elements are arranged on each second sensing mechanism, and the contact states of the N second sensing elements on each second sensing mechanism with the N first sensing elements on a corresponding first sensing mechanism form M sensing states, and M to the power of N is greater than or equal to the number of all rack units, and M and N are positive integers greater than or equal to 2; a monitoring system, configured to determine a target code based on the N target sensing states between the second sensing mechanism on the target electronic device and the corresponding first sensing mechanism, and determine the position of the rack unit associated with the target code as the position of the rack unit of the target electronic device, wherein each rack unit is associated with a unique code and a position, and each unique code is formed by a unique combination of N sensing states.

[0006] In an exemplary embodiment, the contact states of the N first sensing elements of each first sensing mechanism with the N second sensing elements of a second sensing mechanism are preset, and the N contact states between the N first sensing elements of the first sensing mechanism corresponding to each rack unit and the N second sensing elements of the second sensing mechanism of the electronic device on the rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

[0007] In an exemplary embodiment, the first sensing element and the second sensing element are conductors, the first end of each first sensing element is grounded, the second end of each first sensing element is in contact or not in contact with the first end of a second sensing element, the second end of each second sensing element is connected to a power supply terminal, and the high and low level states of the second ends of the N second sensing elements of the second sensing mechanism of the electronic device of each rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

[0008] In an exemplary embodiment, the first sensing mechanism further includes: a base arranged on the target component, the first sensing element is a spring pin, each spring pin is connected to the base through a fixed end, and the spring pin with the telescopic end set to the extended state is in contact with the second sensing element on the second sensing mechanism of the electronic device on the rack unit, and the spring pin with the telescopic end set to the retracted state is not in contact with the second sensing element on the second sensing mechanism of the electronic device on the rack unit.

[0009] In an exemplary embodiment, the second sensing mechanism further includes: N signal lines, one end of each signal line is connected to a second sensing element, and the other end of each signal line is connected to the monitoring system through the interface of the I2C expander.

[0010] In an exemplary embodiment, the second sensing element is a conducting part, the signal line is a wire, an insulating pad is provided between the first end of the conducting part and the outer surface of the chassis of the electronic device, and an opening is provided at the second end of the conducting part for clamping the wire.

[0011] In an exemplary embodiment, the monitoring system includes a baseboard management controller for each electronic device. Each baseboard management controller acquires N sensing states between the second sensing elements on the second sensing structure on the electronic device to which it belongs and the first sensing elements on the corresponding first sensing mechanism, and determines the position of the rack unit associated with the code generated by the N sensing states as the position of the rack unit of the electronic device to which it belongs.

[0012] In an exemplary embodiment, the monitoring system further includes a management switch, which is connected to the baseboard management controller of each electronic device, and is used to acquire the position of the rack unit of the electronic device to which each baseboard management controller belongs, and send the position of the rack unit of each electronic device to the management software.

[0013] In an exemplary embodiment, the target component is a power supply bus bar, and the first sensing mechanism is provided on the housing of the power supply bus bar.

[0014] In an exemplary embodiment, the target component is a cable tray, and the cable tray is used to support the connection cables between the computing nodes and the switching nodes in the electronic device.

[0015] In an exemplary embodiment, among the M to the Nth power of different codes, the codes of the numbers of all rack units are valid codes, and the remaining codes are disabled codes.

[0016] According to another embodiment of the present application, an optional cabinet is provided, including: a cabinet body, the cabinet body includes a target component passing through all rack units, each rack unit is used to install an electronic device; a plurality of first sensing mechanisms, provided on the target component, each first sensing mechanism corresponds to a rack unit, and each rack unit is associated with a unique code and a position information, wherein each unique code is formed by a unique combination of N sensing states, the N sensing states are the sensing states between the N first sensing elements on a first sensing mechanism and different second sensing elements on a second sensing mechanism on the chassis of the electronic device on the rack unit, each sensing state belongs to one of M sensing states, M to the Nth power is greater than or equal to the number of all rack units, and M and N are positive integers greater than or equal to 2.

[0017] In an exemplary embodiment, the contact states between the N first sensing elements of each first sensing mechanism and the N second sensing elements of a second sensing mechanism are preset. The N contact states between the N first sensing elements of the corresponding first sensing mechanism of each rack unit and the N second sensing elements of the second sensing mechanism of the electronic device on the rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

[0018] In an exemplary embodiment, the first sensing element and the second sensing element are conductors. The first end of each first sensing element is grounded. The second end of each first sensing element is in contact or not in contact with the first end of a second sensing element. The second end of each second sensing element is connected to a power supply terminal. The high and low level states of the second ends of the N second sensing elements of the second sensing mechanism of the electronic device of each rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

[0019] In an exemplary embodiment, the first sensing mechanism further includes: a base disposed on the target component. The first sensing element is a spring pin. Each spring pin is connected to the base through a fixed end. The spring pin with the telescopic end set to the extended state is in contact with the second sensing element on the second sensing mechanism of the electronic device on the rack unit, and the spring pin with the telescopic end set to the retracted state is not in contact with the second sensing element on the second sensing mechanism of the electronic device on the rack unit.

[0020] According to another embodiment of the present application, a method for identifying the position of an electronic device is provided, including: obtaining N target sensing states collected by a second sensing mechanism on the chassis of the target electronic device, where the N target sensing states are the sensing states between the N second sensing elements on the second sensing mechanism and different first sensing elements on the corresponding first sensing mechanism. Each sensing state belongs to one of M sensing states, and M to the power of N is greater than or equal to the number of all rack units. M and N are positive integers greater than or equal to 2; determining a target code according to the N target sensing states; determining the position of the rack unit associated with the target code as the position of the rack unit of the target electronic device, where each rack unit is associated with a unique code and a position, and each unique code is formed by a unique combination of N sensing states.

[0021] In an exemplary embodiment, the method further includes: comparing the pre-stored position of the rack unit of the target electronic device with the determined position of the rack unit when the position of the rack unit of the target electronic device is pre-stored; determining that the target electronic device is incorrectly installed and sending a prompt message when the pre-stored position of the rack unit is different from the determined position of the rack unit.

[0022] According to another embodiment of the present application, there is also provided a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any one of the above method embodiments when running.

[0023] According to another embodiment of the present application, there is also provided an electronic device including a memory and a processor, where the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0024] According to another embodiment of the present application, there is also provided a computer program product including a computer program, and the computer program realizes the steps in any one of the above method embodiments when executed by a processor.

[0025] Through the present application, since the contact forms between the first and second sensing mechanisms form multiple sensing states, it is ensured that the electronic devices on each rack unit can be uniquely and accurately positioned. By establishing an association between the sensing states and the unique codes of the rack units, the installation positions of the electronic devices are automatically identified, greatly simplifying the initial configuration process of the server. During subsequent maintenance, equipment replacement, or expansion, the device positions can be quickly located, saving time and labor costs. The design of the sensing mechanism can be adjusted according to actual needs to adapt to cabinets of different scales and architectures. Therefore, it can solve the problem in the related art that due to differences in the architectures of different cabinets, additional positioning devices require complex processing processes, resulting in poor flexibility in positioning the devices in the cabinets, and achieve the effect of improving the flexibility in positioning the devices in the cabinets. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic diagram of the front of the cabinet of an AI supernode whole cabinet server according to an embodiment of the present application;

[0027] Figure 2 is a schematic diagram of the back of the cabinet of an AI supernode whole cabinet server according to an embodiment of the present application;

[0028] Figure 3 is a connection schematic diagram of the computing node and the switching node of an AI supernode whole cabinet server according to an embodiment of the present application Figure 1 ;

[0029] Figure 4 is a connection schematic diagram of the computing node and the switching node of an AI supernode whole cabinet server according to an embodiment of the present application Figure 2 ;

[0030] Figure 5Schematic diagram of the connection between the computing node and the switching node of an AI super-node whole cabinet server according to an embodiment of the present application Figure 3 ;

[0031] Figure 6 Schematic structural diagram of a cabinet according to an embodiment of the present application;

[0032] Figure 7 Schematic diagram of a first sensing mechanism according to an embodiment of the present application;

[0033] Figure 8 Schematic structural diagram of the first sensing mechanism installed on the power supply busbar according to an embodiment of the present application;

[0034] Figure 9 Schematic structural diagram of an optional cabinet according to an embodiment of the present application;

[0035] Figure 10 Flowchart of a method for identifying the position of an electronic device according to an embodiment of the present application;

[0036] Figure 11 Schematic diagram of a method for positioning a whole cabinet according to an embodiment of the present application. Detailed implementation manners

[0037] In the following, embodiments of the present application will be described in detail with reference to the accompanying drawings and in conjunction with the embodiments.

[0038] It should be noted that in the description of the present application, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. The terms "first", "second", etc. in the present application are used to distinguish similar objects and not to describe a specific order or sequence.

[0039] With the continuous development of artificial intelligence technology, the training and inference of AI (Artificial Intelligence) models have attracted increasing attention. As the key infrastructure supporting the training and inference of AI models, the new generation of computing cluster architectures are facing severe performance challenges. The marginal benefit of the performance of a single server processor is decreasing, approaching the physical limit, and it is difficult to meet the continuous growing demand for computing power of large AI models. At the same time, the strategy of horizontally expanding the computing cluster by simply increasing the number of servers has also encountered bottlenecks in efficiency and scalability, facing challenges in aspects such as cost, data synchronization, and energy consumption, which hinder the efficient execution of large-scale parallel computing. Against this background, the AI supernode whole cabinet system has emerged. It systematically integrates multiple groups of GPU (Graphics Processing Unit) computing units and switching units through an innovative topology, comprehensively optimizes in aspects such as multi-source computing power integration, high-speed advanced interconnection, heat dissipation, high-power density power supply, and whole cabinet management, and provides a scalable high-bandwidth domain (HBD, HighBandwidth Domain) supernode system with 32 cards or more to meet the needs of cutting-edge large model training and inference.

[0040] Figure 1 It is a schematic diagram of the front of the cabinet of an AI supernode whole cabinet server according to an embodiment of the present application. As Figure 1 shown, the number of Us represents the number of layers of the cabinet of the whole cabinet server, and the height of each U position can be adjusted. Generally, but not limited to, the square hole bars of each node of the cabinet can be set to three specifications: OU (48mm), RU (44.45mm), or SU (46.5mm). The components of the server can be deployed separately in each layer of the whole cabinet of the server. For example, a management switch can be deployed at the 44OU and 42OU positions, a cable management tray can be deployed at the 43OU position, Powershelf (power shelf, used to provide centralized power management for devices in the server rack) nodes can be deployed at the 37 - 38OU positions and the 5 - 6OU positions, Rack Stiffener (rack stiffener, used to increase the structural strength and stability of the server rack) nodes can be deployed at the 36OU position and the 7OU position, and CDU (Cooling Distribution Unit, a cooling distribution unit used to manage and distribute coolant or cold air) nodes can be deployed at the 1 - 4OU positions. Figure 1 In the shown distribution, the 39 - 41OU positions are not considered for what components to be deployed, so they are marked as the NA (Not Applicable) layer. In actual applications, specific functional components can be considered to be deployed in the NA layer according to the functional requirements of the server. The computing nodes are placed on both the upper and lower sides of the cabinet (for example Figure 1For the 28 OU - 35 OU and 8 OU - 15 OU deployments shown in [reference], computing nodes are deployed, and the switching nodes are placed in the middle of the cabinet (for example Figure 1 For the 16 OU - 27 OU deployment shown in [reference], switching nodes are deployed. Such a design can make the deployment of power supply cables and communication cables more balanced. However Figure 1 What is shown in [reference] is just a typical configuration, and the number of switching nodes and the number of computing nodes can be adjusted according to the actual situation.

[0041] Figure 2 is a schematic diagram of the back of a cabinet of an AI super - node whole - cabinet server according to an embodiment of the present application. As Figure 2 shown, the back of the whole - cabinet server mainly includes three components: Busbar (busway), Manifold (manifold), and Cable Tray (cable tray). Among them, the Busbar is located in the middle (or on one side), and the Powershelf nodes are connected to the Busbar to supply power to the whole cabinet. The Cable Tray is located on the left and right sides of the Busbar. The Cable Tray is used to connect the high - speed signals of the computing nodes and the switching nodes according to the designed topological interconnection relationship. The Manifold is located on the left and right sides at the back of the cabinet, providing water inlet and outlet, and providing a liquid path for liquid cooling of the whole cabinet. The joint action of these components can ensure the efficient and safe operation of the power supply and cooling systems of the whole - cabinet server.

[0042] Figure 3 is a schematic diagram of the connection between the computing nodes and the switching nodes of an AI super - node whole - cabinet server according to an embodiment of the present application Figure 1 As Figure 3As shown in the figure, #1 to #N are N computing nodes. One CPU (Central Processing Unit) can be deployed on each computing node, but is not limited to this. The CPU can be connected to 4 GPUs through a PCIe (Peripheral Component Interconnect Express) switch chip. Taking the example of deploying 16 computing nodes in an AI full rack server, there are a total of 16×4 = 64 GPUs in this AI full rack server. If N computing nodes are deployed in the AI full rack server, there are a total of 4×N GPUs in this AI full rack server. A DPU (Data Processing Unit) network card, a storage device, and a BMC (Baseboard Management Controller) management module can also be deployed on the computing node, but are not limited to this. The storage device can be an SSD (Solid State Drive), but is not limited to this. Figure 3 The switching nodes 1 to switching node M shown in the figure are M switching nodes. Each switching node can have one MAC (Media Access Control) chip, but is not limited to this. The high-speed signals of the switching nodes and the computing nodes are interconnected through a Cable Tray. In the AI full rack server, all the computing nodes and switching nodes are connected to the Rack CableTray (a cable bridge inside or near the rack) through high-density connectors to achieve Serdes (Serializer / Deserializer) high-speed interconnection between all the GPUs in the AI full rack server. Figure 4 It is a schematic diagram of the connection between the computing nodes and the switching nodes of an AI supernode full rack server according to an embodiment of the present application. Figure 2 For the above-mentioned Serdes high-speed interconnection method between all the GPUs, as Figure 4 shown in the figure. It should be noted that in the solution where the computing nodes in the AI full rack server are fully interconnected through the switching nodes, it is necessary to ensure that the high-speed signal resources of each GPU in each computing node are evenly distributed to each switching node. This is a Scale up (vertical expansion) topology. Figure 5 It is a schematic diagram of the connection between the computing nodes and the switching nodes of an AI supernode full rack server according to an embodiment of the present application. Figure 3 As Figure 5As shown, the computing nodes 1-8 are interconnected through the switching nodes 1-8. The connector of each computing node inside the AI whole cabinet server is connected to 4 GPUs. The signals at the same positions of the 8 computing nodes fill up one switch (i.e., the switching node). For example, Figure 5 the computing node signals and the switching node signals with the same numbers shown in it have a connection relationship.

[0043] In this embodiment, a cabinet is provided. Figure 6 It is a schematic structural diagram of the cabinet according to the embodiment of the present application. As Figure 6 shown, the cabinet includes:

[0044] A cabinet body, which includes a target component running through all rack units, and each rack unit is used to install an electronic device;

[0045] In some alternative embodiments, the cabinet body can be the cabinet body of an AI whole cabinet, the rack unit is also the U position, and the target component can be a power supply bus bar. As Figure 1 shown, various electronic devices are equipped on the AI whole cabinet, such as a management switch, a cable management tray, a power supply rack, a rack reinforcement, a computing node (a server for performing data-intensive computing tasks), a switching node (a server for processing and managing network communications), and a CDU (Cooling Distribution Unit), etc. The computing nodes are placed on the upper and lower sides of the cabinet respectively, and the switching nodes are placed in the middle, which can make the power supply and communication lines more balanced. The number of switching nodes and the number of computing nodes can be adjusted according to the actual situation. The height of each U position can also be adjusted. The square hole bar for fixing each node of the cabinet can be set to OU (48mm), or set to RU(44.45mm), or set to SU(46.5mm). In addition to the Figure 2 whole cabinet shown, there is also a whole cabinet where there is no high-speed interconnection between the nodes where each GPU is located. The Figure 2 cable tray in it can be removed.

[0046] A plurality of first sensing mechanisms are provided on the target component. Wherein, each first sensing mechanism corresponds to a rack unit, and N first sensing elements are provided on each first sensing mechanism.

[0047] In some alternative embodiments, the first sensing mechanism may be a spring pin and its base disposed on the power supply bus bar. On the target component, each rack unit, i.e., each U-position, is equipped with a first sensing mechanism. On each first sensing mechanism, N first sensing elements are provided. The number and layout design of the first sensing elements can be determined according to the coding requirements for differentiating all U-positions. The value of N needs to ensure that M^N is greater than or equal to the number of all rack units, where M is the number of distinguishable sensing states of each first sensing element. When each first sensing element (such as a spring pin) physically contacts a sensor (second sensing element) on the server chassis, it can transmit a unique sensing state. These states combine to form a code for identifying a specific U-position.

[0048] A plurality of second sensing mechanisms are respectively disposed on the chassis of different electronic devices. Among them, N second sensing elements are provided on each second sensing mechanism. The contact forms between the N second sensing elements on each second sensing mechanism and the N first sensing elements on a corresponding first sensing mechanism form M sensing states. M to the power of N is greater than or equal to the number of all rack units. M and N are positive integers greater than or equal to 2.

[0049] In some alternative embodiments, on each second sensing mechanism, N second sensing elements are also provided. The second sensing elements can be in the form of sensors and are capable of detecting the contact state of the corresponding first sensing element. The second sensing element can also be a screw, and the signal of the sensing state is transmitted through the contact state and non-contact state between the screw and the spring pin. The first sensing mechanism on each U-position corresponds to the second sensing mechanism disposed on the chassis of the electronic device at that U-position. This corresponding position enables each first sensing element of the first sensing mechanism to be in contact with the second sensing element of the corresponding second sensing mechanism.

[0050] When a server is installed in a specific U-position of the cabinet, the second sensing mechanism on the chassis will come into contact with the first sensing mechanism at that U-position. The contact forms of the N first sensing elements and the N second sensing elements combine to form M sensing states, where the value of M depends on the number of different states that each sensing element can identify.

[0051] M is the number of different states recognized by the sensor. M can be two states (such as contact or non-contact) or more states (such as varying degrees of pressure detected by a pressure sensor). For example, if M equals 2 (contact and non-contact only) and N equals 6 (six sensors per chassis), then M raised to the power of N (2^6) equals 64, sufficient to cover the range of 1 to 52 U-slots in the entire cabinet. The additional states can be used for error detection or system redundancy. Because each server's U-slot within the cabinet must be uniquely identifiable, the values of M and N must satisfy the condition that M raised to the power of N is greater than or equal to the number of rack units.

[0052] A monitoring system is configured to determine a target code based on N target sensing states between a second sensing mechanism and a corresponding first sensing mechanism on a target electronic device, and to determine the position of a rack unit associated with the target code as the position of the rack unit of the target electronic device, wherein each rack unit is associated with a unique code and a position, and each unique code is formed by a unique combination of the N sensing states.

[0053] In some optional embodiments, the monitoring system includes a BMC (Baseboard Management Controller). The BMC reads the contact status of the N second sensing elements on the second sensing mechanism and the corresponding first sensing elements in the first sensing mechanism, that is, the target sensing status. Based on the N target sensing states read, a target code is generated through a preset mapping rule. This code can be a mixed sequence of numbers or alphanumeric characters, which can uniquely identify the specific U position where the server is located. The mapping rule determines the correspondence between each sensing state combination and a specific U position. The monitoring system compares the target code with the preset U position code database. Each rack unit is associated with a unique code, so the location of the U position where the target electronic device is located can be determined by the target code.

[0054] Furthermore, the monitoring system can update server location information in real time and verify that the server is correctly inserted into the intended U position. If the sensing status indicates a misalignment with the expected U position, the monitoring system triggers an alarm, notifying operations personnel to make corrections to ensure stable data center operations. The BMC communicates with the second sensing mechanism via I2C (Inter-Integrated Circuit) or another standard bus to obtain sensing status information and upload it to the monitoring system for further processing.

[0055] Among them, the preset mapping rule can use hexadecimal encoding to represent the U bits. For example, the effective U-bit range is set: the hexadecimal encoding corresponding to 0 - 52 is 0x00 - 0x34, and the disabled encoding: the hexadecimal encoding corresponding to 53 - 63 is 0x35 - 0x3F. Table 1 is the mapping relationship table between the probe activation state and the U bits.

[0056] Table 1

[0057]

[0058] The binary encoding representing the U bits is stored by defining the ProbeBits (bit field) structure. Each member (A0, A1, etc.) in the structure can be an unsigned char type with a bit width (a data type in the programming language) and can store 0 or 1. The ProbeBits structure is used to represent the U-bit encoding obtained from the contact states of the probes (A0, A1, A2, B0, B1, B2). The U-bit encoding function receives a ProbeBits structure as a parameter and calculates the corresponding decimal U-bit value. By multiplying the value of each member in the structure by the corresponding weight and summing (the weight corresponds to the binary bit value), a decimal value representing the U bit can be calculated. The function then checks whether the calculated value is within the valid range (1 to 52). If it exceeds the range, an error message is output and the program exits. If it is within the range, the decimal value and hexadecimal value of the U bit are output.

[0059] To ensure the uniqueness of the encoding corresponding to the U bits, the contact states between each pair of the first sensing mechanism and the second sensing mechanism are preset. In an exemplary embodiment, the contact states of the N first sensing elements of each first sensing mechanism with the N second sensing elements of a second sensing mechanism are preset. The N contact states between the N first sensing elements of the corresponding first sensing mechanism of each rack unit and the N second sensing elements of the second sensing mechanism of the electronic device on the rack unit form a unique encoding, and 2 to the power of N is greater than or equal to the number of all rack units.

[0060] Specifically, the contact state between the N first sensing elements (such as spring pins) on each first sensing mechanism and the N second sensing elements (such as contact sensors) on the second sensing mechanism installed on the server chassis is preset. For each U-position of the cabinet, it is determined which first sensing elements will contact the second sensing elements, forming a fixed coding table. For example, the coding rules for each U-position are set as follows: U-position 1: The A0 contact is in contact, and A1, A2, B0, B1, B2 are not in contact, coded as binary 000001, that is, decimal 1 or hexadecimal 0x01. U-position 2: The A1 contact is in contact, and A0, A2, B0, B1, B2 are not in contact, coded as binary 000010, that is, decimal 2 or hexadecimal 0x02. And so on until U-position 52. By detecting the contact states of the A0 - A2 and B0 - B2 contacts, a specific binary code is generated, and then the position of the U-position where the server is located is parsed out.

[0061] Based on the coding rules and positioning method of the contact state, this embodiment can effectively distinguish the positions of all server nodes, provide the functions of automatic verification and error alarm, enhance the accuracy and reliability of server deployment and management in the data center, optimize the operation and maintenance process, and improve the overall operation efficiency.

[0062] The code output by the second sensing element can be determined by the high and low level states formed by whether it is grounded. In an exemplary embodiment, the first sensing element and the second sensing element are conductors. The first end of each first sensing element is grounded, and the second end of each first sensing element contacts or does not contact the first end of a second sensing element. The second end of each second sensing element is connected to the power supply terminal. The high and low level states of the second ends of the N second sensing elements of the second sensing mechanism of the electronic device in each rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

[0063] Specifically, the first sensing element and the second sensing element can be made of conductor materials to ensure effective electrical contact between the first sensing element and the second sensing element. The first end of each first sensing element is grounded to form a stable potential reference point. When the server is inserted into a specific U-position of the cabinet, the second end of the first sensing element contacts the first end of the second sensing element. Since the second end of the second sensing element is connected to the power supply terminal, the change in the contact state will be directly reflected in the level state of the second end of the second sensing element, that is, a low level or a high level is formed when in contact, and the initial level state is maintained when not in contact.

[0064] The second sensing mechanism on the electronic device per rack unit forms N combinations of high and low levels according to the contact state, and this combination forms a unique code. The monitoring system regularly detects the level states of the second ends of the N second sensing elements on the second sensing mechanism. According to the different combinations of the level states, the monitoring system can parse a specific digital code, and then determine the position of the U-bit where the server is located through the mapping relationship between the code and the U-bit.

[0065] For example, as Figure 6 shown, the first sensing element is a spring pin, the first end of the spring pin is connected to a metal base to be grounded, the second sensing element is a screw. When the second sensing element contacts the first sensing element, since the first sensing element is grounded, a low level is formed at the second end of the second sensing element. When not in contact, the second end of the second sensing element remains at a high level (high impedance state), or is connected to a pull-up resistor to form a high level. It is set that each second sensing element is at a low level (0) when contacting the first sensing element and at a high level (1) when not in contact. For U-bit 1, the possible contact state is A0 contact (low level), and the rest A1 - A2, B0 - B2 are not in contact (high level), and its code is binary 111110, that is, decimal 62 or hexadecimal 0x3E. For U-bit 2, the contact state may be A1 contact and the rest are not in contact, and its code is binary 111101, that is, decimal 61 or hexadecimal 0x3D. And so on until the codes of all U-bits are set.

[0066] In this embodiment, a unique code is generated through the high and low level states formed by contact, which simplifies the hardware design of the positioning system and improves the accuracy and reliability of positioning. Due to the stability of the circuit and the conductivity of the sensing elements, the detection of high and low level states is relatively less susceptible to interference from external environmental factors, enabling the system to operate stably in a complex environment such as a data center. In addition, through the analysis of digital level states, the monitoring system can quickly process and update the server location information, improving the efficiency of data center operation and maintenance management.

[0067] The first sensing element and the second sensing element are made to contact or not contact through the extension and retraction states of the spring pin. In an exemplary embodiment, the first sensing mechanism further includes: a base disposed on the target component, the first sensing element is a spring pin, each spring pin is connected to the base through a fixed end, and the spring pin with the telescopic end set to the extended state contacts the second sensing element on the second sensing mechanism of the electronic device per rack unit, and the spring pin with the telescopic end set to the retracted state does not contact the second sensing element on the second sensing mechanism of the electronic device per rack unit.

[0068] Specifically, as Figure 6As shown, each first sensing mechanism designs a base for fixing spring pins. The metal base is fixed by screws, and one side of the spring pin is fixed to the base using threads. Each base on each side supports a maximum of three spring pins. A total of six spring pins are supported on both sides. On the chassis side, no design modification is made to the Power clip (an electrical contact component used to connect the server node to the power supply busbar or other power supply systems within the cabinet). Instead, sensors that contact the spring pins are added to the nearby chassis. The sensors can use screws, and this is not limited to such a connection method. It is only necessary to provide a sensor that can detect conduction to the spring pins and then lead out wires to connect to the BMC within the node.

[0069] For example, Figure 7 is a schematic diagram of the first sensing mechanism according to an embodiment of the present application. As Figure 7 shown, each spring pin base of the first sensing mechanism includes at least three spring pin positions, and six spring pins can be set. Each spring pin has two states: contact and non-contact. By calculating the contact state combinations of these spring pins, a unique target code can be generated, and this code corresponds to a specific U position according to a preset mapping rule. Figure 8 is a schematic structural diagram of the first sensing mechanism according to an embodiment of the present application installed on the power supply busbar. As Figure 8 shown, the first sensing mechanism corresponding to U1 includes three probes (i.e., spring pins) A0 / A1 / A2 on the left base and three probes B0 / B1 / B2 on the right base. The first sensing mechanism corresponding to U2 includes three probes (i.e., spring pins) C0 / C1 / C2 on the left base and three probes D0 / D1 / D2 on the right base.

[0070] In this embodiment, through the design of the first sensing mechanism that combines spring pins and the base, and through a preset contact mode, combined with electrical detection, a simple and effective server U position positioning technology is formed. It can accurately identify the position of the server in the cabinet, provide an anti-misoperation mechanism for server insertion and extraction. Once the server is inserted into the wrong U position, the mismatch of the contact mode will be immediately recognized by the monitoring system, thereby triggering an error alarm or a position correction process, ensuring the security and efficiency of data center operation and maintenance. In addition, the telescopic design of the spring pins makes the system have a certain adaptability, which can cope with slight changes in the server size or slight vibrations in the cabinet internal environment, increasing the stability and reliability of the positioning system.

[0071] The monitoring system obtains the sensing state through the signal line connected to the second sensing component. In an exemplary embodiment, the second sensing mechanism further includes: N signal lines, one end of each signal line is connected to a second sensing component, and the other end of each signal line is connected to the monitoring system through the interface of the I2C expander.

[0072] Specifically, each signal line in the second sensing mechanism is responsible for transmitting sensing information. One end of the signal line is connected to the second sensing element on the chassis to capture changes in the contact state, and the other end is connected to the monitoring system through the interface of the I2C expander, which is responsible for converting the contact state of the second sensing element into a signal recognizable by the monitoring system to achieve digital transmission of information. The I2C expander is responsible for converting the contact states of multiple signal lines captured by the second sensing mechanism into a unified digital signal and then transmitting these signals to the monitoring system through the I2C bus.

[0073] In this embodiment, through the setting of N signal lines, the sensing state of the second sensing element can be accurately captured and transmitted to the monitoring system, and then through the interface of the I2C expander for digital conversion and efficient communication, finally realizing the U-position positioning of the server in the cabinet. It reduces the wiring complexity, improves the scalability and reliability of the system, enhances the accuracy and real-time performance of server positioning, and also reduces the complexity and cost of system implementation, providing important technical support for the efficient management and operation and maintenance of the data center.

[0074] To ensure the stability of signal transmission, an insulating pad and an opening are provided on the conducting part. In an exemplary embodiment, the second sensing element is the conducting part, the signal line is a wire, an insulating pad is provided between the first end of the conducting part and the outer surface of the chassis of the electronic device, and an opening is provided at the second end of the conducting part for clamping the wire.

[0075] Specifically, as Figure 6 shown, the conducting part can be a screw. One end of the conducting part is near the outer surface of the chassis for detecting the contact state with the spring pin. To prevent accidental conduction between the conducting part and the metal outer surface of the chassis, an insulating pad is provided between this end and the outer surface of the chassis to ensure electrical isolation and avoid electrical faults such as short circuits in the system. The other end of the conducting part is designed with an opening, and the function of this opening is to clamp the wire, that is, the signal line of the sensor. When the spring pin contacts the conducting part, a conductive path will be formed, and the signal line is connected to the conducting part through the opening to transmit the signal of the contact state to the monitoring system. The design of the opening makes the connection of the wire more stable, reducing the risk of signal interruption caused by loose wires or poor contact, and ensuring the continuity and accuracy of the signal.

[0076] In this embodiment, through the design of the conducting part of the second sensing element and its connection method with the signal line (wire), a key component of an efficient and reliable U-position positioning system is formed. It realizes the accurate detection of the contact state, ensures the stability of signal transmission and the accuracy of data processing, providing a solid technical guarantee for the precise positioning of the server in the cabinet. At the same time, the setting of the insulating pad and the opening further improves the electrical safety and mechanical stability of the system.

[0077] In an exemplary embodiment, the monitoring system includes a baseboard management controller for each electronic device. Each baseboard management controller acquires N sensing states between a second sensing element on a second sensing structure on the electronic device to which it belongs and a first sensing element on a corresponding first sensing mechanism, and determines the position of the rack unit associated with the code generated from the N sensing states as the position of the rack unit of the electronic device to which it belongs.

[0078] Specifically, the baseboard management controller on each server node continuously monitors the N sensing states between the second sensing element on the second sensing structure and the first sensing element. The N sensing states are contact signals or on / off states fed back to the baseboard management controller by N independent sensing points on the second sensing structure. Through an I2C expander or other interfaces, the baseboard management controller can read the changes in these sensing states and obtain real-time contact information.

[0079] The baseboard management controller generates a specific code based on the acquired N sensing states. The parsing process of the code needs to follow a preset mapping rule to ensure a one-to-one correspondence between the code and the position of the U-bit, so as to realize the accurate identification of the server's position. The baseboard management controller compares the parsed code with the pre-stored position information of the U-bit to determine the position of the rack unit where the server node is located.

[0080] The baseboard management controller of this embodiment realizes the precise positioning of the server node in the cabinet by acquiring the N sensing states on the second sensing structure, generating a code and parsing the corresponding position of the U-bit.

[0081] After the baseboard management controller parses the U-bit of the electronic device, the management switch sends the U-bit to the management software. In an exemplary embodiment, the monitoring system further includes a management switch. The management switch is connected to the baseboard management controller of each electronic device and is used to acquire the position of the rack unit of the electronic device to which each baseboard management controller belongs and send the position of the rack unit of each electronic device to the management software.

[0082] Specifically, the baseboard management controller of each server node establishes a communication link with the management switch through a network interface (such as an Ethernet interface or a dedicated management network interface), enabling the baseboard management controller to upload the encoded sensing states it detects and the parsed position information of the rack unit (U-bit) to the management switch in real time.

[0083] The management switch receives rack unit location information from the baseboard management controllers of all server nodes, aggregates it, and processes it. By analyzing this massive amount of location data, the management switch constructs a detailed map of the entire server cabinet layout, clearly displaying the actual U-position of each node. After completing the data aggregation, the management switch transmits this location information to management software, such as the data center's automated operations and maintenance platform or server cluster management software.

[0084] The management software receives rack unit location data from the management switch and uses this information to comprehensively monitor and manage the data center's server layout. The software interface intuitively displays the current U-positions of all servers, allowing operations personnel to quickly locate the exact location of any server. The interaction between the management switch and the management software also supports error identification and maintenance support. If the baseboard management controller detects incorrect U-position information, the management switch immediately notifies the management software of these anomalies. The software then generates alerts, notifying operations personnel of servers that may be misplaced and require inspection or relocation.

[0085] This embodiment enhances the real-time performance of the monitoring system by uploading the U-position information to the management switch, which then sends it to the management software. This provides the operation and maintenance team with a timely problem discovery and resolution mechanism, and reduces the operation and maintenance costs and potential risks caused by server location errors.

[0086] In order to improve the flexibility of positioning equipment in the cabinet and save costs, the first sensing mechanism can be set on the power supply busbar. In an exemplary embodiment, the target component is the power supply busbar, and the first sensing mechanism is set on the shell of the power supply busbar.

[0087] Specifically, if Figure 7 As shown, the first sensing mechanism (the pogo pin and its base) can be arranged on the power busbar housing to ensure that each U-position has a corresponding sensing point. Based on the width and thickness of the power busbar and the spacing between the U-positions, the first sensing mechanism is reasonably distributed in space to ensure accurate contact height without interfering with the normal power transmission of the power busbar.

[0088] Because the power busbar carries high current, for electrical safety reasons, an insulating pad is placed between the sensing mechanism and the busbar housing to prevent accidental short circuits or electrical failures. Furthermore, the first sensing mechanism should not interfere with the power supply to the busbar when in contact, while maintaining good electrical isolation when not in contact. The first sensing mechanism, mounted on the housing, must possess excellent mechanical stability to withstand the impact forces generated when server nodes are inserted and removed, while ensuring reliable and durable contact.

[0089] In this embodiment, by arranging the first sensing mechanism on the power supply busbar, the production cost is saved, and the effectiveness and long-term stability of the U-position positioning scheme in the server cabinet are ensured. Through the reasonable layout and design on the housing of the power supply busbar, the first sensing mechanism can cooperate with the second sensing element on the server node to achieve fast and accurate U-position identification, providing important support for the efficient operation and maintenance of the data center.

[0090] The first sensing mechanism can also be arranged on the cable tray. In an exemplary embodiment, the target component is the cable tray, and the cable tray is used to support the connection cables between the computing nodes and the switching nodes in the electronic device.

[0091] Specifically, if the cabinet includes a cable tray component, the first sensing mechanism can be designed and installed at an appropriate position on the cable tray. The layout of the first sensing mechanism needs to consider the space limitation of the cable tray and the routing path of the cables to ensure that it does not interfere with the cable layout and can meet the requirements of contact detection.

[0092] The mechanical structure of the cable tray needs to be optimized to facilitate the installation and maintenance of the first sensing mechanism. For example, installation holes are reserved on the side plate or bottom plate of the cable tray for fixing the first sensing mechanism, and sufficient space is provided to ensure the reliability of electrical contact. To ensure electrical safety, the design of the first sensing mechanism needs to avoid contact with the high-voltage cables carried on the cable tray, and the first sensing mechanism is isolated from the high-voltage cables through insulating materials or isolation structures. In addition, the electrical connection between the first sensing mechanism and the contact points (such as the conduction part) on the server node needs to be designed to be stable enough to prevent positioning errors or system failures caused by poor contact or incorrect connection.

[0093] In this embodiment, by integrating the first sensing mechanism on the cable tray, accurate identification of the U-position of the server node is achieved, further optimizing cable management and improving the operation and maintenance efficiency and safety of the data center.

[0094] The encoding that does not need to correspond to the U-position is set as the disabled encoding. In an exemplary embodiment, among the Nth power of M different encodings, the encodings of the number of all rack units are valid encodings, and the remaining encodings are disabled encodings.

[0095] Specifically, the number of encodings determined based on the number N of sensing elements and the number M of sensing state types can adapt to server nodes with different numbers of U bits. If N is 6 and M is 2, a total of 64 encodings can be generated, which can cover the effective positioning range from 1 to 52 U bits (0x01 - 0x34). For the remaining extra encodings, U bits from 53 to 63 can be specified as disabled states, thereby enhancing the adaptability and scalability of the U-bit positioning scheme. The disabled encodings can be used for error detection or representing the system initialization state, and they do not correspond to the positions of any actual U bits.

[0096] Through the above encoding strategy, this embodiment can accurately identify the positions of the U bits of each server node, and also has error detection and anti-fooling functions, improving the stability and security of system operation.

[0097] According to another embodiment of the present application, an optional cabinet is provided. Figure 9 It is a schematic structural diagram of an optional cabinet according to an embodiment of the present application, as Figure 9 shown, including:

[0098] A cabinet body, which includes target components running through all rack units, and each rack unit is used to install an electronic device;

[0099] In some optional embodiments, the cabinet body can be the cabinet body of an AI full cabinet, the rack unit is also the U bit, and the target component can be a power supply bus bar. As Figure 1 shown, the AI full cabinet is equipped with a variety of electronic devices, such as a management switch, a cable management tray, a power supply rack, a rack reinforcement, a computing node (a server for performing data-intensive computing tasks), a switching node (a server for processing and managing network communications), and a cooling distribution unit, etc. The computing nodes are placed on the upper and lower sides of the cabinet respectively, and the switching nodes are placed in the middle, which can make the power supply and communication lines more balanced. The number of switching nodes and the number of computing nodes can be adjusted according to the actual situation. The height of each U bit can also be adjusted. The square hole bar for fixing each node of the cabinet can be set to OU (48mm), or set to RU (44.45mm), or set to SU (46.5mm).

[0100] As Figure 2As shown, the back of the AI whole cabinet includes a power supply busbar, a fluid distributor, and a cable tray. Among them, the power supply busbar is a conductor used for centralized power transmission and serves as the main trunk line for power distribution in the cabinet or electrical equipment. The fluid distributor in the server liquid cooling system is a pipeline structure for distributing and collecting coolant to ensure that each server node that needs to be cooled can be effectively cooled. The cable tray is a structure in the cabinet for supporting and managing cables, which helps to keep the cables inside the cabinet tidy and facilitates maintenance and management. Among them, the power supply busbar is located in the middle or on one side, and the Powershelf (power supply rack) node is connected to the power supply busbar to supply power to the whole cabinet. The cable trays are located on the left and right sides, and connect the high-speed signals of the computing nodes and the switching nodes according to the designed topological interconnection relationship. The fluid distributors are located on the left and right sides at the back of the cabinet, providing water inlet and outlet, and providing a liquid path for liquid cooling of the whole cabinet. There is also a whole cabinet where there is no high-speed interconnection between the nodes where each GPU is located. The Figure 2 cable tray in it can be removed.

[0101] A plurality of first sensing mechanisms are arranged on the target component. Each first sensing mechanism corresponds to a rack unit. Each rack unit is associated with a unique code and a position information. Among them, each unique code is formed by a unique combination of N sensing states. The N sensing states are the sensing states between N first sensing elements on a first sensing mechanism and different second sensing elements on a second sensing mechanism on the chassis of the electronic device on the rack unit. Each sensing state belongs to one of M sensing states, and M to the power of N is greater than or equal to the number of all rack units. M and N are positive integers greater than or equal to 2.

[0102] In some alternative embodiments, the first sensing mechanism can be a spring pin and its base arranged on the power supply busbar, or a contact sensor or other forms of sensing devices. On the target component, each rack unit, that is, each U-position, is equipped with a first sensing mechanism. Each U-position is associated with a unique code and corresponding position information. The unique code is formed by a specific combination of N sensing states, where N represents the number of first sensing elements associated with the first sensing mechanism. For example, N first sensing elements can be arranged on both sides of the power supply busbar, and each first sensing element corresponds to a specific sensing state. These states together form a binary code or a more complex coding system to represent the position of the U-position.

[0103] The formation of the sensing state can be based on the contact or non-contact state between the first sensing element on the first sensing mechanism and the second sensing element on the second sensing mechanism (such as a conduction part) on the chassis of the electronic device. Each sensing state belongs to one of M possible sensing states, where M is the total number of states that each first sensing element may have, such as contact, non-contact, or conduction states with different resistance values. By detecting these sensing states, a specific coding combination can be formed, and then the position of the U-bit of the server node can be identified.

[0104] For the cabinet in this embodiment, a corresponding sensing state can be preset for each U-bit. When the electronic device is installed in a certain U-bit, the connection between the chassis of the electronic device and the mechanism automatically triggers the sensing state preset for this U-bit. For example, the first sensing element can be a spring pin, and the second sensing element can be a screw on the chassis of the electronic device. For each U-bit on the cabinet, according to the unique code preset for this U-bit, the extending and retracting states of N spring pins on the first sensing mechanism are adjusted correspondingly. After an electronic device is placed on a certain U-bit, the screw on the chassis of the electronic device automatically connects to the spring pins extending on this U-bit, generating N sensing states corresponding to this U-bit.

[0105] Based on the contact state between the first sensing element and the second sensing element, a unique code is determined. In an exemplary embodiment, the contact or non-contact states of the N first sensing elements of each first sensing mechanism and the N second sensing elements of a second sensing mechanism are preset. The N contact or non-contact states between the N first sensing elements of the corresponding first sensing mechanism of each rack unit and the N second sensing elements of the second sensing mechanism of the electronic device on the rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

[0106] Specifically, the contact states between the N first sensing elements (such as spring pins) on each first sensing mechanism and the N second sensing elements (such as contact sensors) on the second sensing mechanism installed on the server chassis are preset. For each U-bit of the cabinet, it is determined which first sensing elements will contact the second sensing elements, forming a fixed contact mode table. For example, the coding rules for each U-bit are set as follows: U-bit 1: Contact of A0 contact point, A1, A2, B0, B1, B2 not in contact, coded as binary 000001, that is, decimal 1 or hexadecimal 0x01. U-bit 2: Contact of A1 contact point, A0, A2, B0, B1, B2 not in contact, coded as binary 000010, that is, decimal 2 or hexadecimal 0x02. And so on, until U-bit 52. By detecting the contact states of A0 - A2 and B0 - B2 contact points, a specific binary code is generated, and then the position of the U-bit where the server is located is parsed.

[0107] Based on the encoding rules and positioning method for the contact state, this embodiment can effectively distinguish the positions of all server nodes, provides functions of automatic verification and error alarm, enhances the accuracy and reliability of server deployment and management in the data center, optimizes the operation and maintenance process, and improves the overall operation efficiency.

[0108] The unique encoding output by the second sensing element can be determined by the high and low level states formed by grounding or not. In an exemplary embodiment, the first sensing element and the second sensing element are conductors. The first end of each first sensing element is grounded, and the second end of each first sensing element is in contact or not in contact with the first end of a second sensing element. The second end of each second sensing element is connected to a power supply terminal. The high and low level states of the second ends of the N second sensing elements of the second sensing mechanism of the electronic device in each rack unit form a unique encoding, and 2 to the power of N is greater than or equal to the number of all rack units.

[0109] Specifically, the first sensing element and the second sensing element can be made of conductor materials to ensure effective electrical contact between the first sensing element and the second sensing element. The first end of each first sensing element is grounded to form a stable potential reference point. When the server is inserted into a specific U-position of the cabinet, the second end of the first sensing element is in contact with the first end of the second sensing element. Since the second end of the second sensing element is connected to the power supply terminal, the change in the contact state will be directly reflected in the level state of the second end of the second sensing element, that is, a low level or a high level is formed when in contact, and the initial level state is maintained when not in contact.

[0110] The second sensing mechanism on the electronic device in each rack unit will form a combination of N high and low levels according to the contact state, and this combination forms a unique encoding. The monitoring system will regularly detect the level states of the second ends of the N second sensing elements on the second sensing mechanism. According to the different combinations of the level states, the monitoring system can parse out a specific digital encoding, and then determine the position of the U-position where the server is located through the mapping relationship between the encoding and the U-position.

[0111] For example, as Figure 6As shown, the first sensing element is a spring pin. The first end of the spring pin is connected to the metal base to be grounded. The second sensing element is a screw. When the second sensing element contacts the first sensing element, since the first sensing element is grounded, a low level will be formed at the second end of the second sensing element. When not in contact, the second end of the second sensing element remains at a high level (high impedance state), or is connected to a pull-up resistor to form a high level. It is set that each second sensing element is at a low level (0) when in contact with the first sensing element and at a high level (1) when not in contact. For U-bit 1, the possible contact states are A0 contact (low level), and the rest A1 - A2, B0 - B2 are not in contact (high level), and its encoding is binary 111110, that is, decimal 62 or hexadecimal 0x3E. For U-bit 2, the contact state may be A1 contact, and the rest are not in contact, and its encoding is binary 111101, that is, decimal 61 or hexadecimal 0x3D. And so on until the encodings of all U-bits are set.

[0112] In this embodiment, unique encodings are generated through the high and low level states formed by contact, which simplifies the hardware design of the positioning system and improves the accuracy and reliability of positioning. Due to the stability of the circuit and the conductivity of the sensing elements, the detection of high and low level states is relatively less susceptible to interference from external environmental factors, enabling the system to operate stably in a complex environment such as a data center. In addition, through the analysis of digital level states, the monitoring system can quickly process and update the server location information, improving the efficiency of data center operation and maintenance management.

[0113] The contact or non-contact between the first sensing element and the second sensing element is achieved through the extended and retracted states of the spring pin. In an exemplary embodiment, the first sensing mechanism further includes: a base disposed on the target component. The first sensing element is a spring pin, and each spring pin is connected to the base through a fixed end. The spring pin with the telescopic end set to the extended state contacts the second sensing element on the second sensing mechanism of the electronic device on the rack unit, and the spring pin with the telescopic end set to the retracted state does not contact the second sensing element on the second sensing mechanism of the electronic device on the rack unit.

[0114] Specifically, as Figure 6 shown, for each first sensing mechanism, a base for fixing the spring pins is designed, and the metal base is fixed by screws. One side of the spring pin is fixed to the base using threads. Each side of the base supports at most three spring pins. A total of 6 spring pins are supported on both sides. On the chassis side, no design modification is made to the Power clip (an electrical contact component used to connect the server node to the power bus bar or other power supply systems in the cabinet), but only sensors that contact the spring pins are added to the nearby chassis. The sensors can use screws, and are not limited to such connection methods. It is sufficient to provide a sensor that can detect conduction to the spring pin and then lead out a wire to connect to the baseboard management controller in the node.

[0115] In this embodiment, through the design of the first induction mechanism combining spring pins and the base, and by presetting the contact mode and combining electrical detection, a simple and effective server U-position positioning technology is formed. It can accurately identify the position of the server in the cabinet, providing an anti-error mechanism for server insertion and extraction. Once the server is inserted into the wrong U-position, the mismatch of the contact mode will be immediately recognized by the monitoring system, thus triggering an error alarm or a position correction process to ensure the security and efficiency of data center operation and maintenance. In addition, the telescopic design of the spring pins enables the system to have a certain adaptability, which can cope with minor changes in the server size or slight vibrations in the cabinet interior environment, increasing the stability and reliability of the positioning system.

[0116] According to another embodiment of the present application, a method for identifying the position of an electronic device is also provided. Figure 10 It is a flowchart of the method for identifying the position of an electronic device according to an embodiment of the present application, as Figure 10 shown. This process includes the following steps:

[0117] Step S1002, obtaining N target induction states collected by a second induction mechanism on the chassis of the target electronic device, where the N target induction states are the induction states between N second induction elements on the second induction mechanism and different first induction elements on the corresponding first induction mechanism, and each induction state belongs to one of M induction states, and the Nth power of M is greater than or equal to the number of all rack units, and M and N are positive integers greater than or equal to 2.

[0118] In some optional embodiments, the target electronic device may be a server, and the second induction mechanism is a sensor located inside each server chassis or near the rear window structure. These sensors are used to detect the contact state with the spring pins on the Busbar. The first induction mechanism may include spring pins and a base for fixing the spring pins. The first induction element may be a spring pin, and the second induction element may be a sensor. The sensors on each server chassis will detect the contact state (i.e., the induction state) with the spring pins at specific positions on the Busbar.

[0119] M may be the number of different states recognized by the sensor. M may be two states (such as contact or non-contact), or more states (the pressure intensity levels collected by the pressure sensor). For example, if M is equal to 2 (only two states of contact and non-contact) and N is equal to 6 (6 sensors are configured on each chassis), then the Nth power of M (2^6) is equal to 64, which is sufficient to cover the range of 1 to 52 U-positions of the entire cabinet. The extra states can be used for error detection or system redundancy.

[0120] Step S1004, determining a target code according to the N target induction states.

[0121] In some alternative embodiments, when identifying the position of a target electronic device in an entire cabinet, positioning can be performed based on N target sensing states collected from a second sensing mechanism on the chassis of the target electronic device. First, a target code is determined based on the N target sensing states. For example, M equals 2, N equals 6, the non-contact state is represented by "0", the contact state is represented by "1", and the N target sensing states are "000100". According to the mapping of binary, the target code corresponding to the N target sensing states is 0*2^5 + 0*2^4 + 0*2^3 + 1*2^2 + 0*2^1 + 0*2^0 = 4, and 4 can represent the 4th U position.

[0122] Step S1006: Determine the position of the rack unit associated with the target code as the position of the rack unit of the target electronic device, where each rack unit is associated with a unique code and a position, and each unique code is formed by a unique combination of N sensing states.

[0123] In some alternative embodiments, a target code is calculated through the combination of the sensing states of the sensor. This code is formed by a unique combination of N sensing states. Using a preset mapping rule, a one-to-one correspondence is established between the target code and the rack units (U positions) in the cabinet. That is, each valid code is assigned a U position, representing the exact position of the server in the cabinet. Ensure that each U position has a unique code, and each code is only associated with a specific U position. Convert the mapping relationship between the calculated target code and the U position into specific position information and output or record it. For example, if the target code is 4, it is determined that the server is located at the 4th U position.

[0124] Through the above steps, since the contact forms between the first and second sensing mechanisms form multiple sensing states, it is ensured that the electronic devices on each rack unit can be uniquely and accurately positioned. By establishing an association between the sensing states and the unique codes of the rack units, the installation position of the electronic device is automatically identified, greatly simplifying the initial configuration process of the server. During subsequent maintenance, equipment replacement, or expansion, the device position can be quickly located, saving time and labor costs. The design of the sensing mechanism can be adjusted according to actual needs to adapt to cabinets of different scales and architectures. Therefore, it is possible to solve the problem in the related art that due to differences in the architectures of different cabinets, additional positioning devices require complex processing procedures, resulting in poor flexibility in positioning the devices in the cabinet, and achieve the effect of improving the flexibility of positioning the devices in the cabinet.

[0125] Among them, the execution subject of the above steps can be a server, a terminal, etc., but is not limited thereto.

[0126] In an exemplary embodiment, the method further includes: when the position of the rack unit of the target electronic device is pre-stored, comparing the pre-stored position of the rack unit with the determined position of the rack unit; when the pre-stored position of the rack unit is different from the determined position of the rack unit, determining that the target electronic device is incorrectly installed and sending a prompt message.

[0127] Specifically, during the installation or initialization of servers in a cabinet, the preset U-position of each electronic device is recorded. For example, if the target electronic device is server A, the preset position of server A is U10. The contact state with the spring pin base on the power supply busbar is detected by a sensor, and the actual U-position of the current target electronic device is calculated. For example, the state detected by the sensor corresponds to the code 0x14. According to the hexadecimal mapping rule, this represents that the target electronic device is currently actually located at the U20 position. The real-time determined position (U20) is compared with the pre-stored position information (U10). If the two are consistent, it indicates that the device is correctly installed; if they are different, it means that the device may be installed in the wrong position. When it is detected that the pre-stored position is inconsistent with the determined position, the installation error is automatically identified, and a prompt message is sent in a predetermined manner (such as through a management software interface, email, text message, or audible and visual alarm) to notify the operation and maintenance personnel in the data center that device A may be incorrectly installed at the U20 position instead of its preset U10 position.

[0128] The automatic verification and error detection mechanism of this embodiment can significantly improve the maintenance efficiency of the data center. When a server node is moved or replaced, the system can immediately detect the position change and verify whether it conforms to the preset layout, thus avoiding the cumbersome process of manual inspection one by one, and reducing the fault response time and cost caused by incorrect device insertion. After detecting an installation error, the system can further provide detailed error information and guidance to help the operation and maintenance personnel quickly locate the problem and take correct recovery measures. At the same time, by recording and analyzing these errors, future similar problems can be prevented, and the stability and availability of the entire data center can be improved.

[0129] According to another embodiment of the present application, a positioning method for an entire cabinet is further provided. Figure 11 is a schematic diagram of the positioning method for an entire cabinet according to an embodiment of the present application, as Figure 11 shown, the method includes: collecting the induction state of the contact sensor between the power supply busbar and the rear window of the chassis through an I2C bus expander chip, and the I2C bus expander chip transmits the collected induction state to the baseboard management controller through an input / output port. The baseboard management controller generates a specific code according to the obtained induction state, and parses the code to determine the position of the U-position corresponding to the code.

[0130] In this embodiment, through the positioning method of the whole cabinet, by using the I2C bus expander chip, the status information of the contact sensor can be efficiently collected and transmitted, and then encoded and parsed by the baseboard management controller to determine the position of the U-bit of the server node. The automatic and accurate positioning of the server node is realized, and the efficiency and accuracy of the operation and maintenance of the data center are improved.

[0131] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence, or the part that makes a contribution to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM (Read-Only Memory), RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to enable a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0132] In this embodiment, a position recognition device for rack units is also provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that can implement a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0133] It should be noted that the above-mentioned various modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to this: the above modules are all located in the same processor; or, the above-mentioned various modules are separately located in different processors in any combination form.

[0134] The embodiment of the present application also provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium, and the computer program is set to execute the steps in any one of the above method embodiments when running.

[0135] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: USB flash drive, read-only memory (abbreviated as ROM), random access memory (abbreviated as RAM), mobile hard disk, magnetic disk, or optical disk and other various media that can store computer programs.

[0136] Embodiments of the present application further provide an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0137] In an exemplary embodiment, the above electronic device may further include a transmission device and an input / output device. The transmission device is connected to the above processor, and the input / output device is connected to the above processor.

[0138] Embodiments of the present application further provide a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0139] Embodiments of the present application further provide another computer program product, including a non-volatile computer-readable storage medium. The non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0140] Embodiments of the present application further provide a computer program. The computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in any one of the above method embodiments.

[0141] Specific examples in this embodiment may refer to the examples described in the above embodiments and exemplary embodiments, and will not be repeated here.

[0142] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the present application is not limited to any specific combination of hardware and software. <N

[0143] The above are only the preferred embodiments of the present application and are not used to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the principle of the present application shall be included in the protection scope of the present application.

Claims

1. A cabinet, characterized in that, Including: A cabinet body, which includes a target component penetrating all rack units, and each rack unit is used to install an electronic device; A plurality of first sensing mechanisms, which are arranged on the target component. Among them, each first sensing mechanism corresponds to a rack unit, and N first sensing elements are arranged on each first sensing mechanism; A plurality of second sensing mechanisms, which are respectively arranged on the chassis of different electronic devices. Among them, N second sensing elements are arranged on each second sensing mechanism, and the contact states of the N second sensing elements on each second sensing mechanism and the N first sensing elements on a corresponding first sensing mechanism form M sensing states, and M to the power of N is greater than or equal to the number of all rack units, and M and N are positive integers greater than or equal to 2; A monitoring system, which is used to determine a target code based on the N target sensing states between the second sensing mechanism on the target electronic device and the corresponding first sensing mechanism, and determine the position of the rack unit associated with the target code as the position of the rack unit of the target electronic device. Among them, each rack unit is associated with a unique code and a position, and each unique code is formed by a unique combination of N sensing states; Among them, the contact or non-contact states of the N first sensing elements of each first sensing mechanism and the N second sensing elements of the second sensing mechanism on the electronic device on the rack unit are preset, and the N contact or non-contact states between the N first sensing elements of the corresponding first sensing mechanism of each rack unit and the N second sensing elements of the second sensing mechanism of the electronic device on the rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

2. The cabinet according to claim 1, wherein The first sensing element and the second sensing element are conductors. The first end of each first sensing element is grounded, the second end of each first sensing element is in contact or not in contact with the first end of a second sensing element, the second end of each second sensing element is connected to a power supply terminal, and the high and low level states of the second ends of the N second sensing elements of the second sensing mechanism of the electronic device of each rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

3. The cabinet according to claim 1, characterized in that, The first sensing mechanism further includes: a base, which is arranged on the target component. The first sensing element is a spring pin, and each spring pin is connected to the base through a fixed end. The spring pin with the telescopic end set in the extended state is in contact with the second sensing element on the second sensing mechanism of the electronic device on the rack unit, and the spring pin with the telescopic end set in the retracted state is not in contact with the second sensing element on the second sensing mechanism of the electronic device on the rack unit.

4. The cabinet according to claim 2, wherein The second sensing mechanism further includes: N signal lines, one end of each signal line is connected to a second sensing element, and the other end of each signal line is connected to the monitoring system through the interface of an I2C expander.

5. The cabinet according to claim 4, wherein, The second sensing element is a conduction part, the signal line is a wire, an insulating pad is arranged between the first end of the conduction part and the outer surface of the chassis of the electronic device, and an opening is arranged at the second end of the conduction part for clamping the wire.

6. The cabinet according to claim 1, characterized in that, The monitoring system includes a baseboard management controller for each electronic device. Each baseboard management controller acquires N induction states between a second inducer on a second induction structure on the electronic device to which it belongs and a first inducer on a corresponding first induction mechanism, and determines the position of the rack unit associated with the code generated from the N induction states as the position of the rack unit of the electronic device to which it belongs.

7. The cabinet according to claim 6, wherein The monitoring system further includes a management switch. The management switch is connected to the baseboard management controller of each electronic device, and is configured to acquire the position of the rack unit of the electronic device to which each baseboard management controller belongs, and send the position of the rack unit of each electronic device to the management software.

8. The cabinet according to claim 1, wherein, The target component is a power supply busbar, and the first induction mechanism is disposed on the housing of the power supply busbar.

9. The cabinet according to claim 1, characterized in that, The target component is a cable tray, and the cable tray is used to support the connection cables between the computing nodes and the switching nodes in the electronic device.

10. The cabinet according to claim 1, characterized in that, Among the M to the power of N different codes, the codes of the number of all rack units are valid codes, and the remaining codes are disabled codes.

11. A cabinet, characterized in that, Including: A cabinet body, which includes a target component passing through all rack units, and each rack unit is used to install an electronic device; A plurality of first induction mechanisms, disposed on the target component, each first induction mechanism corresponding to a rack unit, each rack unit associated with a unique code and a position information, wherein each unique code is formed by a unique combination of N induction states, and the N induction states are the induction states between N first inductors on a first induction mechanism and different second inductors on a second induction structure on the chassis of the electronic device in the rack unit. Each induction state belongs to one of M induction states, M to the power of N is greater than or equal to the number of all rack units, and M and N are positive integers greater than or equal to 2; The contact or non-contact states of the N first inductors of each first induction mechanism and the N second inductors of a second induction mechanism are preset. The N contact or non-contact states between the N first inductors of the corresponding first induction mechanism of each rack unit and the N second inductors of the second induction mechanism of the electronic device in the rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

12. The cabinet according to claim 11, wherein, The first inductor and the second inductor are conductors. The first end of each first inductor is grounded. The second end of each first inductor is in contact or not in contact with the first end of a second inductor. The second end of each second inductor is connected to a power supply terminal. The high and low level states of the second ends of the N second inductors of the second induction mechanism of the electronic device in each rack unit form a unique code, and 2 to the power of N is greater than or equal to the number of all rack units.

13. The cabinet according to claim 11, characterized in that, The first sensing mechanism further includes: a base, which is arranged on the target component. The first sensing element is a spring pin, and each spring pin is connected to the base through a fixed end. The spring pin with the telescopic end set in the extended state contacts the second sensing element on the second sensing mechanism of the electronic device on the rack unit, and the spring pin with the telescopic end set in the retracted state does not contact the second sensing element on the second sensing mechanism of the electronic device on the rack unit.

14. A method for identifying the location of an electronic device, characterized in that, Applied to the cabinet according to any one of claims 1 to 13, including: Obtaining N target sensing states collected by the second sensing mechanism on the chassis of the target electronic device, where the N target sensing states are the sensing states between N second sensing elements on the second sensing mechanism and different first sensing elements on the corresponding first sensing mechanism, and each sensing state belongs to one of M sensing states, and M to the power of N is greater than or equal to the number of all rack units, and M and N are positive integers greater than or equal to 2; Determining a target code according to the N target sensing states; Determining the position of the rack unit of the target electronic device as the position of the rack unit associated with the target code, where each rack unit is associated with a unique code and a position, and each unique code is formed by a unique combination of N sensing states.

15. The method according to claim 14, wherein The method further includes: When the position of the rack unit of the target electronic device is pre-stored, comparing the pre-stored position of the rack unit with the determined position of the rack unit; When the pre-stored position of the rack unit is different from the determined position of the rack unit, determining that the target electronic device is installed incorrectly and sending a prompt message.

16. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 14 to 15 are implemented.

17. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, where when the computer program is executed by the processor, the steps of the method according to any one of claims 14 to 15 are implemented.

18. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 14 to 15 are implemented.

Images