A server

Abstract

The present application discloses a server. A plurality of graphics processing nodes are arranged in a first cabinet and are vertically stacked in sequence. A plurality of switching nodes are arranged in a second cabinet and are horizontally stacked in sequence. Among them, the graphics processing nodes are connected to the switching nodes. Therefore, the technical problem of poor reliability of the server can be solved, the difficulty of installation and maintenance can be reduced, and the integrity of signal transmission can be improved, thereby improving the reliability of the server.

Description

Technical Field

[0001] This application relates to the technical field of computing devices, and particularly to a server. Background Art

[0002] As the number of parameters of AI large models increases to tens of billions or even hundreds of billions, the video memory capacity and computing power of a single GPU can no longer meet the requirements. In related technologies, multiple GPUs can be integrated into a physical unit through high-bandwidth interconnection technology to form a server cluster with stronger collaborative computing capabilities, and this server cluster is called a supernode server.

[0003] In order to achieve cross-node communication inside the supernode server, a cable tray needs to be deployed at the rear of the cabinet. The cable tray can accommodate thousands of high-speed cables to connect GPUs and switching nodes. However, due to the high-density wiring of the cable tray, it is difficult to locate and repair a single cable failure, and the whole needs to be replaced, resulting in extremely poor maintainability. Moreover, long-distance wiring will cause serious signal integrity problems, leading to poor reliability of the server. Summary of the Invention

[0004] This application provides a server to at least solve the problem of poor reliability of the server in related technologies.

[0005] This application provides a server, including a first cabinet, a second cabinet, multiple graphics processing nodes, multiple switching nodes, a first power cabinet, and a second power cabinet. Among them,

[0006] The first power cabinet, the second power cabinet, and multiple graphics processing nodes are arranged in the first cabinet, and the first power cabinet, multiple graphics processing nodes, and the second power cabinet are vertically stacked in sequence;

[0007] Multiple switching nodes are arranged in the second cabinet, and multiple switching nodes are horizontally stacked in sequence;

[0008] Among them, the graphics processing nodes are connected to the switching nodes.

[0009] In a possible implementation manner, multiple connectors are included in the graphics processing nodes, and the multiple connectors in the graphics processing nodes are located on the first side of the first cabinet;

[0010] Multiple connectors are included in the switching nodes, and the multiple connectors in the switching nodes are located on the second side of the second cabinet;

[0011] The first side of the first cabinet is disposed opposite to the second side of the second cabinet, and the graphics processing nodes are connected to the switching nodes through the connectors.

[0012] In a possible implementation, the graphics processing unit node includes N connectors, where N is the number of switching nodes and N is an integer greater than 1;

[0013] The switching node includes M connectors, where M is the number of graphics processing unit nodes and M is an integer greater than 1;

[0014] Each of the N connectors in the graphics processing unit node is connected to one connector in each switching node.

[0015] In a possible implementation, the first cabinet further includes a second power supply copper busbar, and the second cabinet further includes a first power supply copper busbar;

[0016] The first power supply copper busbar is used to supply power to multiple graphics processing unit nodes, and the second power supply copper busbar is used to supply power to multiple switching nodes;

[0017] The first power supply copper busbar is located at the middle position among multiple switching nodes and is close to the first side of the first cabinet, and the second power supply copper busbar is located at the middle position among multiple graphics processing unit nodes and is close to the second side of the second cabinet;

[0018] Wherein, both ends of the first power supply copper busbar are respectively connected to the first power supply cabinet and the second power supply cabinet, and the first power supply copper busbar is connected to the second power supply copper busbar.

[0019] In a possible implementation, the graphics processing unit node includes a first power supply copper busbar, and the switching node includes a second power supply copper busbar;

[0020] The first power supply copper busbar in the graphics processing unit node is connected to the first power supply copper busbar of the second cabinet;

[0021] The second power supply copper busbar in the switching node is connected to the second power supply copper busbar of the first cabinet.

[0022] In a possible implementation, the first cabinet further includes multiple second radiators, and the second cabinet further includes multiple first radiators;

[0023] The first radiators are used to dissipate heat from multiple graphics processing unit nodes, and the second radiators are used to dissipate heat from multiple switching nodes;

[0024] The multiple second radiators are respectively located between the first power supply cabinet and multiple graphics processing unit nodes, and between multiple graphics processing unit nodes and the second power supply cabinet;

[0025] The multiple first radiators are respectively located on the left and right sides in the second cabinet.

[0026] In a possible implementation, the multiple first radiators include a first water inlet radiator and a first water outlet radiator. The first water inlet radiator is connected to the cabinet water inlet pipe, and the first water outlet radiator is connected to the cabinet water outlet pipe. The multiple graphics processing nodes are respectively connected to the first water inlet radiator and the first water outlet radiator.

[0027] In a possible implementation, the graphics processing node further includes a first water inlet interface and a first water outlet interface;

[0028] The first water inlet interface is connected to the first water inlet radiator, and the first water outlet interface is connected to the first water outlet radiator.

[0029] In a possible implementation, the multiple second radiators include a second water inlet radiator and a second water outlet radiator. The second water inlet radiator is connected to the first water inlet radiator through a water inlet adapter pipe, and the second water outlet radiator is connected to the first water outlet radiator through a water outlet adapter pipe. The multiple switching nodes are respectively connected to the second water inlet radiator and the second water outlet radiator.

[0030] In a possible implementation, the switching node further includes a second water inlet interface and a second water outlet interface;

[0031] The second water inlet interface is connected to the second water inlet radiator, and the second water outlet interface is connected to the second water outlet radiator.

[0032] In a possible implementation, the graphics processing node includes multiple graphics processors, multiple central processors, multiple memories, multiple hard disks, multiple network cards, integrated circuit chips, and multiple connectors;

[0033] The multiple graphics processors are connected to the multiple connectors through the integrated circuit chips. The graphics processors are connected to the central processors, the central processors are connected to the memories, and the central processors are respectively connected to the multiple hard disks and the multiple network cards.

[0034] In a possible implementation, the graphics processing node further includes a first power supply copper bar, a first water inlet interface, and a first water outlet interface;

[0035] The first power supply copper bar is located at the middle position among the multiple connectors, and the first water inlet interface and the first water outlet interface are respectively located on both sides of the multiple connectors.

[0036] In a possible implementation, the switching node includes multiple switching chips, multiple cable connectors, multiple optical modules, and multiple connectors;

[0037] The multiple switching chips are respectively connected to the multiple cable connectors and the multiple connectors. The multiple cable connectors are respectively connected to the multiple connectors, and the multiple switching chips are connected to the multiple optical modules.

[0038] In a possible implementation, the switching node further includes a second power supply copper bar, a second water inlet interface, and a second water outlet interface;

[0039] The second power supply copper bar is located at the middle position among the plurality of connectors, dividing the plurality of connectors into a plurality of first connectors and a plurality of second connectors. The second water inlet interface and the second water outlet interface are respectively located on both sides of the plurality of connectors;

[0040] The plurality of cable connectors are divided into a plurality of first cable connectors and a plurality of second cable connectors. The plurality of first cable connectors are connected to the plurality of second connectors through cables, and the plurality of second cable connectors are connected to the plurality of first connectors through cables;

[0041] The plurality of switching chips include a first switching chip and a second switching chip. The first switching chip is connected to the plurality of first cable connectors through a printed circuit board, and the first switching chip is also connected to the plurality of first connectors through a printed circuit board. The second switching chip is connected to the plurality of second cable connectors through a printed circuit board, and the second switching chip is also connected to the plurality of second connectors through a printed circuit board.

[0042] In a possible implementation, the second water inlet interface is connected to the heat dissipation pipeline of the second switching chip. The heat dissipation pipeline of the second switching chip is connected to the heat dissipation pipelines of the plurality of optical modules. The heat dissipation pipelines of the plurality of optical modules are connected to the heat dissipation pipeline of the first switching chip. The heat dissipation pipeline of the first switching chip is connected to the second water outlet interface.

[0043] Through the present application, a plurality of graphics processing nodes are arranged in the first cabinet, and the plurality of graphics processing nodes are vertically stacked in sequence. A plurality of switching nodes are arranged in the second cabinet, and the plurality of switching nodes are horizontally stacked in sequence. Among them, the graphics processing nodes and the switching nodes are connected. Therefore, the technical problem of poor reliability of the server can be solved, the difficulty of installation and maintenance can be reduced, and the integrity of signal transmission can be improved, thereby improving the reliability of the server. Description of the Drawings

[0044] In order to more clearly illustrate the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0045] Figure 1 Schematic diagram of the front side structure of a related server provided by an embodiment of the present application;

[0046] Figure 2 Schematic diagram of the rear side structure of a related server provided by an embodiment of the present application;

[0047] Figure 3 A schematic structural diagram of a server provided by an embodiment of the present application;

[0048] Figure 4 A schematic diagram of the connection relationship between a graphics processing unit node and a switching node provided by an embodiment of the present application;

[0049] Figure 5 A schematic structural diagram of a power supply copper bar provided by an embodiment of the present application;

[0050] Figure 6 A schematic structural diagram of a server provided by an embodiment of the present application;

[0051] Figure 7 Another schematic structural diagram of a server provided by an embodiment of the present application;

[0052] Figure 8 A schematic structural diagram of a radiator provided by an embodiment of the present application;

[0053] Figure 9 A schematic structural diagram of a graphics processing unit node provided by an embodiment of the present application;

[0054] Figure 10 A schematic structural diagram of a switching node provided by an embodiment of the present application;

[0055] Figure 11 A schematic diagram of the heat dissipation pipeline of a switching node provided by an embodiment of the present application. Detailed implementation manners

[0056] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present application.

[0057] It should be noted that in the description of the present application, the terms "including", "comprising" or any other variants 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 explicitly 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, rather than to describe a specific order or sequence.

[0058] As the number of parameters of AI large models increases to tens of billions or even hundreds of billions, the video memory capacity and computing power of a single GPU can no longer meet the requirements. In related technologies, multiple GPUs can be integrated into a physical unit through high-bandwidth interconnection technology to form a server cluster with stronger collaborative computing capabilities, and this server cluster is called a supernode server.

[0059] To achieve cross-node communication within the supernode server, a cable tray needs to be deployed at the rear of the cabinet. The cable tray can accommodate thousands of high-speed cables to connect the GPUs and the switching nodes.

[0060] Next, in combination with Figure 1 and Figure 2 , an example will be given to illustrate the structure of the supernode server in related technologies.

[0061] Please refer to Figure 1 and Figure 2 , Figure 1 which is a schematic diagram of the front-side structure of a related server provided by an embodiment of the present application. Figure 2 which is a schematic diagram of the rear-side structure of a related server provided by an embodiment of the present application. Figure 1 It may include the front-side structure of the cabinet. Figure 2 It may include the rear-side structure of the cabinet.

[0062] The front-side structure of the cabinet may include a power cabinet, multiple graphics processing nodes, and multiple switching nodes. The power cabinet is located at the upper and lower ends of the cabinet.

[0063] The power cabinet can be used to supply power to the graphics processing nodes and the switching nodes to ensure the normal operation of each node.

[0064] The graphics processing nodes can be used to execute computing tasks.

[0065] The switching nodes can be used for data exchange and communication between nodes to build a data transmission network.

[0066] The graphics processing nodes can cooperate with the switching nodes to achieve data processing and transmission.

[0067] Multiple switching nodes can be located between multiple graphics processing nodes.

[0068] For example, the server may include 16 graphics processing nodes and 8 switching nodes. The 8 switching nodes are located in the middle of the 16 graphics processing nodes, and power cabinets are also included on the upper and lower sides of the graphics processing nodes.

[0069] The rear structure of the cabinet may include multiple radiators, multiple cable trays, and a power supply copper busbar. The multiple radiators are located on the left and right sides, the multiple cable trays are located in the middle of the radiators, and the power supply copper busbar is located in the middle of the multiple cable trays.

[0070] The radiators can be used to help heat-generating components such as graphics processor nodes and switching nodes dissipate heat, maintaining the server within an appropriate operating temperature range.

[0071] The cable trays can be used for the connection between graphics processor nodes and switching nodes. Usually, thousands of cables are arranged in the cable trays, which are responsible for transmitting high-speed data signals required for inter-node communication.

[0072] The power supply copper busbar can be used for power transmission of the server.

[0073] However, in the case of high-density wiring of the cable trays, it is difficult to locate and repair a single cable failure, and the whole needs to be replaced, resulting in extremely poor maintainability. Moreover, long-distance wire routing will cause serious signal integrity problems, leading to poor reliability of the server.

[0074] To solve the above technical problems, the embodiments of the present application provide a server. By arranging multiple graphics processor nodes in a first cabinet and stacking the multiple graphics processor nodes vertically in sequence, and arranging multiple switching nodes in a second cabinet and stacking the multiple switching nodes horizontally in sequence, where the graphics processor nodes are connected to the switching nodes. In this way, cable trays are not required, the difficulty of installation and maintenance is reduced, and the integrity of signal transmission is improved, enhancing the reliability of the server.

[0075] To enable those skilled in the art of the present technology to better understand the solution of the present application, the following further elaborates on the present application in conjunction with the drawings and specific implementation manners.

[0076] In combination with the specific hardware architecture on which the execution of the server depends, the specific hardware architecture is described herein.

[0077] Next, in combination with Figure 3 , the server of the present application is explained.

[0078] The server may include a first cabinet, a second cabinet, multiple graphics processor nodes, multiple switching nodes, a first power cabinet, and a second power cabinet.

[0079] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of a server provided by an embodiment of the present application. Figure 3 It may include a first cabinet and a second cabinet.

[0080] The first power cabinet, the second power cabinet, and multiple graphics processing nodes are arranged in the first cabinet, and the first power cabinet, the multiple graphics processing nodes, and the second power cabinet are vertically stacked in sequence.

[0081] Multiple switching nodes are arranged in the second cabinet, and the multiple switching nodes are horizontally stacked in sequence;

[0082] Among them, the graphics processing nodes are connected to the switching nodes.

[0083] Among them, in the first cabinet from top to bottom are: the first power cabinet, multiple graphics processing nodes, and the second power cabinet.

[0084] The power cabinet can be used to supply power to the server.

[0085] The graphics processing nodes can be used to perform tasks such as graphics processing and computing. Multiple graphics processing nodes can work in parallel to improve computing efficiency and meet complex computing requirements.

[0086] The multiple switching nodes in the second cabinet are arranged in sequence in a horizontally stacked manner.

[0087] In this way, the switching nodes are horizontally arranged to form an independent heat dissipation area, avoiding heat interference with high-power graphics processing nodes.

[0088] The switching nodes can be used to transmit the data generated during the operation or the data that needs to be obtained, so as to realize the data circulation and processing of the entire system.

[0089] The graphics processing nodes can be connected to the switching nodes through connectors.

[0090] The connectors can include orthogonal connectors, etc., which are not limited here.

[0091] In this way, there is no need for a cable tray, reducing the difficulty of installation and maintenance, and improving the integrity of signal transmission, thereby improving the reliability of the server.

[0092] In a possible implementation manner, multiple connectors are included in the graphics processing nodes, and the multiple connectors in the graphics processing nodes are located on the first side of the first cabinet.

[0093] Multiple connectors are included in the switching nodes, and the multiple connectors in the switching nodes are located on the second side of the second cabinet.

[0094] The first side of the first cabinet is disposed opposite to the second side of the second cabinet, and the graphics processing nodes are connected to the switching nodes through connectors.

[0095] Among them, the connectors can be interface components for realizing data transmission.

[0096] The first side of the first cabinet is arranged opposite to the second side of the second cabinet. This layout of opposite arrangement is to facilitate the connection between the graphics processing unit (GPU) nodes and the switch nodes through connectors. Through this face-to-face layout and the docking of connectors, the GPU nodes and the switch nodes can establish a stable connection channel, realize data transmission and interaction between the two, enabling the entire system to work collaboratively to complete tasks such as AI training and data processing.

[0097] In a possible implementation, the GPU node includes N connectors, where N is the number of switch nodes and N is an integer greater than 1;

[0098] The switch node includes M connectors, where M is the number of GPU nodes and M is an integer greater than 1;

[0099] The N connectors in the GPU node are respectively connected to one connector in each switch node.

[0100] Among them, N can be twice M.

[0101] Next, in combination with Figure 4 , the connection relationship between the GPU node and the switch node will be explained.

[0102] Figure 4 FIG. [FIGURE NUMBER] is a schematic diagram of the connection relationship between a GPU node and a switch node provided by an embodiment of the present application. Please refer to Figure 4 , Figure 4 , which includes multiple GPU nodes and multiple switch nodes.

[0103] A mesh topology is formed between the GPU nodes and the switch nodes. Multiple GPU nodes and multiple switch nodes are connected pairwise.

[0104] In a possible implementation, the first cabinet further includes a second power supply copper bar, and the second cabinet further includes a first power supply copper bar.

[0105] Next, in combination with Figure 5 and Figure 6 , the power supply copper bar will be explained.

[0106] Figure 5 FIG. [FIGURE NUMBER] is a schematic diagram of the structure of a power supply copper bar provided by an embodiment of the present application. Figure 6 FIG. [FIGURE NUMBER] is a schematic diagram of the server structure provided by an embodiment of the present application. Please refer to Figure 5 , Figure 5 , which includes a first power supply copper bar, a second power supply copper bar, a first power supply cabinet, and a second power supply cabinet. Please refer to Figure 6 , Figure 6It includes a first cabinet and a second cabinet. The first cabinet includes a first power cabinet, multiple graphics processing unit (GPU) nodes, a second power supply copper busbar, and a second power cabinet. The second cabinet includes multiple switching nodes and a first power supply copper busbar.

[0107] The first power supply copper busbar is used to supply power to the multiple GPU nodes, and the second power supply copper busbar is used to supply power to the multiple switching nodes.

[0108] The first power supply copper busbar is located at the middle position among the multiple switching nodes and is close to the first side of the first cabinet. The second power supply copper busbar is located at the middle position among the multiple GPU nodes and is close to the second side of the second cabinet.

[0109] Wherein, both ends of the first power supply copper busbar are respectively connected to the first power cabinet and the second power cabinet, and the first power supply copper busbar is connected to the second power supply copper busbar.

[0110] Wherein, the first power supply copper busbar can be connected to the second power supply copper busbar through a connector.

[0111] In this way, the first power supply copper busbar is dedicated to supplying power to the multiple GPU nodes, and the second power supply copper busbar is responsible for supplying power to the multiple switching nodes. This clearly defined power supply method can ensure that nodes with different functions can obtain stable power, meet the working requirements such as high-energy-consuming operations of the GPU nodes and data transmission of the switching nodes, and guarantee the stable operation of the system.

[0112] In a possible implementation manner, the first power supply copper busbar is included in the GPU nodes, and the second power supply copper busbar is included in the switching nodes.

[0113] The first power supply copper busbar in the GPU nodes is connected to the first power supply copper busbar of the second cabinet.

[0114] The second power supply copper busbar in the switching nodes is connected to the second power supply copper busbar of the first cabinet.

[0115] In a possible implementation manner, the first cabinet further includes multiple second radiators, and the second cabinet further includes multiple first radiators.

[0116] Next, in combination with Figure 7 , the positions of the radiators in the server will be explained.

[0117] Figure 7 It is another schematic diagram of the server structure provided by the embodiment of the present application. Please refer to Figure 7 , Figure 7 It includes a first cabinet and a second cabinet. The first cabinet includes a first power cabinet, second radiators, multiple GPU nodes, a second power supply copper busbar, and a second power cabinet. The second cabinet includes first radiators, multiple switching nodes, and a first power supply copper busbar.

[0118] The first radiator is used to dissipate heat for multiple graphics processing unit nodes, and the second radiator is used to dissipate heat for multiple switching nodes;

[0119] Multiple second radiators are respectively located between the first power cabinet and multiple graphics processing unit nodes, and between multiple graphics processing unit nodes and the second power cabinet;

[0120] Multiple first radiators are respectively located on the left and right sides in the second cabinet.

[0121] Next, in conjunction with Figure 8 , the structure of the radiator will be explained.

[0122] Figure 8 The following is a schematic structural diagram of a radiator provided by an embodiment of the present application. Please refer to Figure 8 , Figure 8 It includes multiple first radiators and multiple second radiators.

[0123] The multiple first radiators include a first water inlet radiator and a first water outlet radiator. The first water inlet radiator is connected to the cabinet water inlet pipe, the first water outlet radiator is connected to the cabinet water outlet pipe, and multiple graphics processing unit nodes are respectively connected to the first water inlet radiator and the first water outlet radiator.

[0124] The multiple second radiators include a second water inlet radiator and a second water outlet radiator. The second water inlet radiator is connected to the first water inlet radiator through a water inlet adapter pipe, the second water outlet radiator is connected to the first water outlet radiator through a water outlet adapter pipe, and multiple switching nodes are respectively connected to the second water inlet radiator and the second water outlet radiator.

[0125] In this way, for the graphics processing unit nodes, they are respectively connected to the first water inlet radiator and the first water outlet radiator, and the switching nodes are connected to the second water inlet radiator and the second water outlet radiator, which can directly take away a large amount of heat generated during the operation of the nodes, ensure that the nodes work at an appropriate temperature, and maintain the system performance and stability. Moreover, the multiple first radiators and the multiple second radiators are connected to each other through adapter pipes, integrating the heat dissipation systems of the graphics processing unit nodes and the switching nodes, simplifying the structure and layout of the overall heat dissipation system, facilitating installation, maintenance and management, and reducing the system complexity and maintenance cost.

[0126] In a possible implementation manner, the graphics processing unit node further includes a first water inlet interface and a first water outlet interface;

[0127] The first water inlet interface is connected to the first water inlet radiator, and the first water outlet interface is connected to the first water outlet radiator.

[0128] In a possible implementation manner, the switching node further includes a second water inlet interface and a second water outlet interface;

[0129] The second water inlet interface is connected to the second water inlet radiator, and the second water outlet interface is connected to the second water outlet radiator.

[0130] In a possible implementation, the graphics processing unit node includes multiple graphics processing units, multiple central processing units, multiple memories, multiple hard disks, multiple network cards, integrated circuit chips, and multiple connectors.

[0131] The multiple graphics processing units are connected to the multiple connectors through the integrated circuit chips. The graphics processing unit is connected to the central processing unit, the central processing unit is connected to the memory, and the central processing unit is respectively connected to the multiple hard disks and the multiple network cards.

[0132] In a possible implementation, the graphics processing unit node further includes a first power supply copper bus, a first water inlet interface, and a first water outlet interface;

[0133] The first power supply copper bus is located at the middle position among the multiple connectors, and the first water inlet interface and the first water outlet interface are respectively located on both sides of the multiple connectors.

[0134] Next, in combination with Figure 9 , the graphics processing unit node will be explained.

[0135] Figure 9 This is a schematic structural diagram of a graphics processing unit node provided by an embodiment of the present application. Please refer to Figure 9 , Figure 9 including a graphics processing unit node.

[0136] The graphics processing unit node may include multiple graphics processing units, multiple central processing units, multiple memories, multiple hard disks, multiple network cards, integrated circuit chips, multiple connectors, a first power supply copper bus, a first water inlet interface, and a first water outlet interface.

[0137] The first water inlet interface and the first water outlet interface can be used to connect the coolant pipelines of the radiator, and the heat generated during the operation of the node can be carried away through the circulating flow of the coolant, ensuring that each component works within an appropriate temperature range.

[0138] The integrated circuit chip can be used to process and control the signals of each component inside the node.

[0139] In a possible implementation, the switching node includes multiple switching chips, multiple cable connectors, multiple optical modules, and multiple connectors;

[0140] The multiple switching chips are respectively connected to the multiple cable connectors and the multiple connectors. The multiple cable connectors are respectively connected to the multiple connectors, and the multiple switching chips are connected to the multiple optical modules.

[0141] The switching node further includes a second power supply copper bus, a second water inlet interface, and a second water outlet interface;

[0142] The second power supply copper bar is located at the middle position among the multiple connectors, which divides the multiple connectors into multiple first connectors and multiple second connectors. The second water inlet interface and the second water outlet interface are respectively located on both sides of the multiple connectors.

[0143] The multiple cable connectors are divided into multiple first cable connectors and multiple second cable connectors. The multiple first cable connectors are connected to the multiple second connectors through cables, and the multiple second cable connectors are connected to the multiple first connectors through cables.

[0144] The multiple switching chips include a first switching chip and a second switching chip. The first switching chip is connected to the multiple first cable connectors through a printed circuit board, and the first switching chip is also connected to the multiple first connectors through a printed circuit board. The second switching chip is connected to the multiple second cable connectors through a printed circuit board, and the second switching chip is also connected to the multiple second connectors through a printed circuit board.

[0145] In this way, through the connection method of the printed circuit board, the number of cables required inside the switching node is reduced, and the signal integrity is improved.

[0146] Next, in combination with Figure 10 , the switching node will be explained.

[0147] Figure 10 FIG. Figure 10 , Figure 10 is a schematic structural diagram of a switching node provided by an embodiment of the present application. Please refer to

[0148] The switching node may include multiple switching chips, multiple cable connectors, multiple optical modules, multiple connectors, a second power supply copper bar, a second water inlet interface, and a second water outlet interface.

[0149] In a possible implementation manner, the multiple switching chips include a first switching chip and a second switching chip. The first switching chip includes a first heat dissipation pipe, and the second switching chip includes a second heat dissipation pipe.

[0150] Next, in combination with Figure 11 , the heat dissipation pipe will be explained.

[0151] Figure 11 FIG. Figure 11 , Figure 11 is a schematic diagram of the heat dissipation pipe of a switching node provided by an embodiment of the present application. Please refer to

[0152] The second water inlet interface is connected to the heat dissipation pipeline of the second switching chip. The heat dissipation pipeline of the second switching chip is connected to the heat dissipation pipelines of multiple optical modules. The heat dissipation pipelines of multiple optical modules are connected to the heat dissipation pipeline of the first switching chip. The heat dissipation pipeline of the first switching chip is connected to the second water outlet interface.

[0153] In this application, "at least one" means one or more, and "multiple" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B may be singular or plural. The character " / " generally represents an "or" relationship between the preceding and following associated objects; in a formula, the character " / " represents a "division" relationship between the preceding and following associated objects. "At least one (item)" or its similar expression refers to any combination of these items, including any combination of single-item (item) or plural items (items). For example, at least one (item) of a, b, or c may represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c may be single or multiple.

[0154] It can be understood that the various numerical numbers involved in the embodiments of this application are only for the convenience of description and are not used to limit the scope of the embodiments of this application. In the embodiments of this application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The execution order of each process should be determined by its function and internal logic and should not constitute any limitation to the implementation process of the embodiments of this application.

[0155] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A server, characterized in that, Including a first cabinet, a second cabinet, a plurality of graphics processing nodes, a plurality of switching nodes, a first power supply cabinet and a second power supply cabinet, wherein, The first power supply cabinet, the second power supply cabinet and the plurality of graphics processing nodes are arranged in the first cabinet, and the first power supply cabinet, the plurality of graphics processing nodes and the second power supply cabinet are vertically stacked in sequence; The plurality of switching nodes are arranged in the second cabinet, and the plurality of switching nodes are horizontally stacked in sequence; Wherein, the graphics processing nodes are connected to the switching nodes; The first cabinet further includes a second power supply copper bar, and the second cabinet further includes a first power supply copper bar; The first power supply copper bar is used to supply power to the plurality of graphics processing nodes, and the second power supply copper bar is used to supply power to the plurality of switching nodes; The first power supply copper bar is located at the middle position among the plurality of switching nodes and close to the first side of the first cabinet, and the second power supply copper bar is located at the middle position among the plurality of graphics processing nodes and close to the second side of the second cabinet; Wherein, both ends of the first power supply copper bar are respectively connected to the first power supply cabinet and the second power supply cabinet, and the first power supply copper bar is connected to the second power supply copper bar.

2. The server according to claim 1, wherein The graphics processing nodes include a plurality of connectors, and the plurality of connectors in the graphics processing nodes are located on the first side of the first cabinet; The switching nodes include a plurality of connectors, and the plurality of connectors in the switching nodes are located on the second side of the second cabinet; The first side of the first cabinet is arranged opposite to the second side of the second cabinet, and the graphics processing nodes and the switching nodes are connected through connectors.

3. The server according to claim 2, wherein The graphics processing nodes include N connectors, where N is the number of the switching nodes, and N is an integer greater than 1; The switching nodes include M connectors, where M is the number of the graphics processing nodes, and M is an integer greater than 1; The N connectors in the graphics processing nodes are respectively connected to one connector in each switching node.

4. The server according to any one of claims 1-3, characterized in that, The graphics processing nodes include a first power supply copper bar, and the switching nodes include a second power supply copper bar; The first power supply copper bar in the graphics processing nodes is connected to the first power supply copper bar of the second cabinet; The second power supply copper bar in the switching nodes is connected to the second power supply copper bar of the first cabinet.

5. The server according to any one of claims 1 to 3, characterized in that The first cabinet further includes a plurality of second radiators, and the second cabinet further includes a plurality of first radiators; The first radiators are used to dissipate heat from the plurality of graphics processing nodes, and the second radiators are used to dissipate heat from the plurality of switching nodes; The plurality of second radiators are respectively located between the first power supply cabinet and the plurality of graphics processing nodes, and between the plurality of graphics processing nodes and the second power supply cabinet; The plurality of first radiators are respectively located on the left and right sides in the second cabinet.

6. The server according to claim 5, wherein The multiple first radiators include a first water inlet radiator and a first water outlet radiator. The first water inlet radiator is connected to the cabinet water inlet pipe, and the first water outlet radiator is connected to the cabinet water outlet pipe. The multiple graphics processing nodes are respectively connected to the first water inlet radiator and the first water outlet radiator.

7. The server according to claim 6, wherein The graphics processing node further includes a first water inlet interface and a first water outlet interface; The first water inlet interface is connected to the first water inlet radiator, and the first water outlet interface is connected to the first water outlet radiator.

8. The server according to claim 5, wherein The multiple second radiators include a second water inlet radiator and a second water outlet radiator. The second water inlet radiator is connected to the first water inlet radiator through a water inlet adapter pipe, and the second water outlet radiator is connected to the first water outlet radiator through a water outlet adapter pipe. The multiple switching nodes are respectively connected to the second water inlet radiator and the second water outlet radiator.

9. The server according to claim 8, characterized in that The switching node further includes a second water inlet interface and a second water outlet interface; The second water inlet interface is connected to the second water inlet radiator, and the second water outlet interface is connected to the second water outlet radiator.

10. The server according to any one of claims 1-3, characterized in that, The graphics processing node includes multiple graphics processors, multiple central processors, multiple memories, multiple hard disks, multiple network cards, integrated circuit chips, and multiple connectors; The multiple graphics processors are connected to the multiple connectors through the integrated circuit chips. The graphics processor is connected to the central processor, the central processor is connected to the memory, and the central processor is respectively connected to the multiple hard disks and the multiple network cards.

11. The server according to claim 10, wherein The graphics processing node further includes a first power supply copper bar, a first water inlet interface, and a first water outlet interface; The first power supply copper bar is located at the middle position among the multiple connectors, and the first water inlet interface and the first water outlet interface are respectively located on both sides of the multiple connectors.

12. The server according to any one of claims 1-3, characterized in that, The switching node includes multiple switching chips, multiple cable connectors, multiple optical modules, and multiple connectors; The multiple switching chips are respectively connected to the multiple cable connectors and the multiple connectors. The multiple cable connectors are respectively connected to the multiple connectors, and the multiple switching chips are connected to the multiple optical modules.

13. The server according to claim 12, wherein The switching node further includes a second power supply copper bar, a second water inlet interface, and a second water outlet interface; The second power supply copper bar is located at the middle position among the multiple connectors, dividing the multiple connectors into multiple first connectors and multiple second connectors. The second water inlet interface and the second water outlet interface are respectively located on both sides of the multiple connectors; The multiple cable connectors are divided into multiple first cable connectors and multiple second cable connectors. The multiple first cable connectors are connected to the multiple second connectors through cables, and the multiple second cable connectors are connected to the multiple first connectors through cables; The multiple switching chips include a first switching chip and a second switching chip. The first switching chip is connected to the multiple first cable connectors through a printed circuit board. The first switching chip is also connected to the multiple first connectors through a printed circuit board. The second switching chip is connected to the multiple second cable connectors through a printed circuit board. The second switching chip is also connected to the multiple second connectors through a printed circuit board.

14. The server according to claim 13, wherein The second water inlet interface is connected to the heat dissipation pipeline of the second switching chip. The heat dissipation pipeline of the second switching chip is connected to the heat dissipation pipelines of the multiple optical modules. The heat dissipation pipelines of the multiple optical modules are connected to the heat dissipation pipeline of the first switching chip. The heat dissipation pipeline of the first switching chip is connected to the second water outlet interface.

Images