server

Through the nested architecture design, the second computing module is installed in the recessed part of the first computing module, which solves the problem of the high height of the server and realizes the compact layout and space optimization of the server in the computer room.

CN120371095BActive Publication Date: 2025-09-12INSPUR SUZHOU INTELLIGENT TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202510873136.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-06-26
Publication Date
2025-09-12
Estimated Expiration
2045-06-26

AI Technical Summary

Technical Problem

The server is relatively tall, which occupies a larger height space in the computer room and affects the layout of the server in the computer room.

Method used

A nested architecture design is adopted, and the second computing module is installed in the recessed part of the first computing module to realize the staggered arrangement of the computing modules and reduce the space occupied by the server in the height direction.

Benefits of technology

The overall height of the server is reduced, the internal structure of the chassis is optimized, the space utilization is improved, and the reasonable layout of the server in the computer room is facilitated.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120371095B_ABST
    Figure CN120371095B_ABST
Patent Text Reader

Abstract

The present application discloses a server, which relates to the field of server technology. A second computing module is installed in a recessed portion of a first computing module to implement a nested architecture design of the computing modules. This allows the first computing module and the second computing module to be staggered in a second direction, significantly reducing the total height of the combined structure of the first computing module and the second computing module in the second direction. This makes the internal structure of the server more compact, helps reduce the height of the chassis, and further reduces the height of the server. This solves the technical problem in the related art that the server is too high, resulting in the server occupying a large height space in the computer room and affecting the layout of the server in the computer room, and achieves the technical effect of reducing the height of the server.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of servers, and in particular to a server. Background Art

[0002] With the continuous development of the AI ​​market, high-performance AI servers are becoming increasingly popular. To ensure heat dissipation performance and a compact server size, servers are typically equipped with multiple PCIe expansion cards, memory cards, CPU (Central Processing Unit) components, and GPU (Graphics Processing Unit) components. Heat dissipation is provided by radiators to prevent overheating.

[0003] However, the height of the server in the above-mentioned related art is relatively high, resulting in the server occupying a relatively large height space in the computer room, thus affecting the layout of the server in the computer room. Summary of the Invention

[0004] The present application provides a server to at least solve the problem in the related art that the server is relatively high, resulting in the server occupying a large height space in the computer room and affecting the layout of the server in the computer room.

[0005] This application provides a server, which includes:

[0006] A chassis having an accommodating cavity therein, and a first window and a second window opposite to each other and communicating with the accommodating cavity along a first direction; the first direction is consistent with an extending direction of the front and rear windows of the chassis;

[0007] a first computing module disposed on a side of the chassis near the first window, wherein a recessed portion is formed on a side of the first computing module facing the top of the chassis along a second direction, wherein the second direction is a height direction of the chassis;

[0008] The second computing module is arranged in the chassis and installed in the recessed portion, and the second computing module is used to assist the first computing module in processing computing tasks.

[0009] Through the present application, since a recess is formed on the side of the first computing module facing the top of the chassis, and the second computing module is installed in the recess of the first computing module, a nested architecture design of the computing module is realized, so that the first computing module and the second computing module are staggered in the second direction, and the space reuse of the first computing module in the height direction of the chassis is realized, rather than the form of staggered arrangement in the second direction in the prior art. In this way, the total height of the combined structure of the first computing module and the second computing module in the second direction is significantly reduced compared with the prior art, making the internal structure of the server more compact, reducing the space occupied by the first computing module and the second computing module in the second direction of the chassis, helping to reduce the height of the chassis, and further reducing the height of the server, for example, reducing the height of the server to 5U. Therefore, the technical problem of the high height of the server in the related art, which causes the server to occupy a large height space in the computer room and affects the layout of the server in the computer room, can be solved, and the technical effect of reducing the height of the server, reducing the space occupied by the server, and facilitating the reasonable layout of the server in the computer room can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0011] Figure 1 A schematic diagram of the structure of a server provided in an embodiment of the present application;

[0012] Figure 2 An exploded diagram of a server provided in an embodiment of the present application;

[0013] Figure 3 A schematic diagram of the structure of a chassis provided in an embodiment of the present application;

[0014] Figure 4 for Figure 1 Cross-section at AA;

[0015] Figure 5 A schematic structural diagram of a first computing module from a first perspective provided in an embodiment of the present application;

[0016] Figure 6 A schematic structural diagram of a first computing module from a second perspective provided in an embodiment of the present application;

[0017] Figure 7 A schematic diagram of the structure of a first window side of a server provided in an embodiment of the present application;

[0018] Figure 8 A schematic diagram of the internal structure of a first computing module provided in an embodiment of the present application;

[0019] Figure 9 A schematic structural diagram of a first storage unit provided in an embodiment of the present application;

[0020] Figure 10 A schematic structural diagram of a second storage unit provided in an embodiment of the present application;

[0021] Figure 11 A schematic diagram of the structure of a second computing module provided in an embodiment of the present application;

[0022] Figure 12 A schematic diagram of the internal structure of a second computing module provided in an embodiment of the present application;

[0023] Figure 13 A schematic structural diagram of a rear plug-in box module from a first perspective provided in an embodiment of the present application;

[0024] Figure 14 A schematic structural diagram of a rear plug-in box module from a second perspective provided in an embodiment of the present application;

[0025] Figure 15 An exploded view of a rear plug-in box module provided in an embodiment of the present application;

[0026] Figure 16 A schematic diagram of the structure of a power distribution unit provided in an embodiment of the present application;

[0027] Figure 17 A schematic structural diagram of the second window side of a chassis provided in an embodiment of the present application;

[0028] Figure 18 for Figure 3 Cross-section at BB.

[0029] The above drawings include the following reference numerals:

[0030] 100, chassis;

[0031] 110, accommodating chamber; 120, first window; 130, second window; 140, first side panel;

[0032] 150, second side panel; 160, first handle; 121, opening;

[0033] 200. First computing module;

[0034] 210, recessed portion; 220, expansion unit; 230, first storage unit;

[0035] 240, first computing unit; 250, second storage unit; 260, first housing;

[0036] 270, third connector; 280, first liquid cooling unit;

[0037] 211. Open your mouth;

[0038] 221, expansion card;

[0039] 231, second housing; 232, second cavity; 233, first storage element;

[0040] 234, first connecting member; 235, second connecting member;

[0041] 251, third housing; 252, third cavity; 253, second storage element;

[0042] 254. Third connecting member; 255. Input / output assembly;

[0043] 261. Second in command;

[0044] 281. First cold plate; 282. First joint assembly; 2821. First bracket;

[0045] 2822, first joint;

[0046] 300. Second computing module;

[0047] 310, fourth housing; 320, fourth cavity; 330, second computing unit;

[0048] 340. Second liquid cooling unit; 350. Liquid leakage detection connector;

[0049] 311, third handle; 341, second joint assembly; 342, third joint assembly;

[0050] 3411, second bracket; 3412, second connector; 3421, third bracket;

[0051] 3422, third connector;

[0052] 400. A first power supply unit;

[0053] 500, rear plug-in box module;

[0054] 510, fifth housing; 520, fifth cavity; 530, middle back plate unit;

[0055] 540, power distribution unit; 550, air cooling unit; 560, first baffle;

[0056] 570, second baffle; 580, third baffle;

[0057] 511, first port; 512, second port; 513, first compartment; 514, fourth handle;

[0058] 531, second power supply unit; 532, second plug interface; 533, back panel;

[0059] 541, first power connector; 542, second power connector; 543, second connector;

[0060] 544, air guide duct; 551, fan; 561, first limiting hole; 571, second limiting hole;

[0061] 581, the third limiting hole;

[0062] 600, first sliding mechanism;

[0063] 610, second pulley set; 620, first slide rail;

[0064] 700, second sliding mechanism;

[0065] 710, fourth pulley set; 720, third slide rail. DETAILED DESCRIPTION

[0066] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0067] It should be noted that the terms "center," "longitudinal," "transverse," "length," "width," "thickness," "upper," "lower," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," "circumferential," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended solely for ease of description and simplification of the present application. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present application. The terms "mounted," "connected," and "connected" should be interpreted broadly, and may include, for example, fixed, removable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediary; and internal communication between two components. The terms "parallel," "perpendicular," and "equal" encompass the described conditions and conditions similar to the described conditions, provided that the range of the similar conditions is within an acceptable range of deviation, as determined by a person of ordinary skill in the art taking into account the measurement in question and the errors associated with the measurement of the particular quantity (i.e., the limitations of the measurement system). For example, "parallel" includes both absolute parallelism and approximate parallelism, where the acceptable deviation range for approximate parallelism may be, for example, within 5°; "perpendicular" includes both absolute perpendicularity and approximate perpendicularity, where the acceptable deviation range for approximate perpendicularity may also be, for example, within 5°. "Equal" includes both absolute equality and approximate equality, where the acceptable deviation range for approximate equality may be, for example, that the difference between the two is less than or equal to 5% of either. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0068] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0069] As described in the background, servers in related art are relatively tall, which causes them to occupy a significant amount of space in the computer room, impacting their layout. This problem arises because the CPU and GPU components in related art are stacked in the height direction of the chassis. The height of the chassis is affected by the sum of the heights of the CPU and GPU components. The greater the sum of the two heights, the taller the chassis needs to be, resulting in a taller server, which in turn occupies a significant amount of space in the computer room.

[0070] refer to Figure 1 、 Figure 2 、 Figure 3 and Figure 4An embodiment of the present application provides a server, which may include a chassis 100, a first computing module 200 and a second computing module 300.

[0071] The chassis 100 may have an accommodating cavity 110 inside, and a first direction (eg Figure 2 The first window 120 and the second window 130 are opposite to and communicate with the accommodating cavity 110 in the X direction. The first direction is consistent with the extending direction of the front and rear windows of the chassis 100.

[0072] It is understood that the interior of the chassis 100 having the accommodating cavity 110, as well as the first window 120 and the second window 130, which are opposite and connected to the accommodating cavity 110 along a first direction, means that the chassis 100 has a front-to-back installation channel. Specifically, this can be achieved using a metal frame with a removable panel. The first window 120 is used to install the computing module, and the second window 130 is used to connect the heat sink and power module, facilitating modular maintenance and airflow circulation. Specifically, the first window 120 is a front window facing the front of the chassis 100, and the second window 130 is a rear window facing the rear of the chassis 100.

[0073] The first computing module 200 is disposed on a side of the chassis 100 close to the first window 120 and along the second direction (eg Figure 2 The first computing module 200 has a concave portion 210 formed on one side of the chassis 100 facing the top, and the second direction is the height direction of the chassis 100. The first computing module 200 can be disposed in the receiving cavity 110 of the chassis 100 through the first window 120.

[0074] It is understood that the first computing module 200 can adopt a stepped layout design to form the lower recess 210. For example, by staggering the components in the first computing module 200 in the second direction to form a stepped structure and simultaneously forming a recess, the height difference is utilized to save vertical space and provide a mounting location for the second computing module 300.

[0075] The second computing module 300 is disposed within the chassis 100 and mounted within the lower recess 210 , thereby achieving a nested design between the second computing module 300 and the first computing module 200 and reusing the space within the accommodating cavity 110 occupied by the first computing module 200 in the height direction. The second computing module 300 is used to assist the first computing module 200 in processing computing tasks. The second computing module 300 can be disposed within the accommodating cavity 110 of the chassis 100 through the first window 120 .

[0076] In a specific implementation, the second computing module 300 can be first installed in the lower recess 210 of the first computing module 200 , and then the combined structure of the first computing module 200 and the second computing module 300 is installed in the accommodating cavity 110 through the first window 120 of the chassis 100 .

[0077] In some embodiments, the second computing module 300 can be directly placed in the concave portion 210 , so that the first computing module 200 can carry and support the second computing module 300 .

[0078] Alternatively, the second computing module 300 and the first computing module 200 can also be assembled in a detachable connection manner, which can improve the connection stability between the first computing module 200 and the second computing module 300, and can prevent the combination of the first computing module 200 and the second computing module 300 from shaking during installation in the accommodating cavity 110, thereby improving the assembly efficiency of the first computing module 200, the second computing module 300 and the chassis 100, or can prevent the second computing module 300 from shaking relative to the first computing module 200 in the chassis 100, thereby improving the structural stability of the server.

[0079] In some embodiments, in the second direction, the second computing module 300 can be completely located in the recess 210 of the first computing module 200. For example, the surface of the second computing module 300 facing the top of the chassis 100 does not protrude from the opening 211 of the recess 210 toward the top of the chassis 100. In this case, the second computing module 300 will not affect the height of the chassis 100.

[0080] Alternatively, in the second direction, the recess 210 can accommodate part of the second computing module 300, and a part of the second computing module 300 is located in the recess 210. In this way, only the part of the structure of the second computing module 300 protruding from the recess 210 will affect the height of the chassis 100, and will also reduce the total height of the second computing module 300 and the second computing module 300 to a certain extent.

[0081] An embodiment of the present application provides a server, which realizes a nested architectural design of the computing modules by forming a recess 210 on one side of the first computing module 200 facing the top of the chassis 100, and installing the second computing module 300 in the recess 210 of the first computing module 200, so that the first computing module 200 and the second computing module 300 are staggered in the second direction, and spatial reuse of the first computing module 200 in the height direction of the chassis 100 is realized, rather than the staggered arrangement in the second direction in the prior art. In this way, the total height of the combined structure of the first computing module 200 and the second computing module 300 in the second direction is significantly reduced compared with the prior art, making the internal structure of the server more compact, reducing the space occupied by the first computing module 200 and the second computing module 300 in the second direction inside the chassis 100, helping to reduce the height of the chassis 100, and further reducing the height of the server, for example, reducing the height of the server to 5U, where U can be inches and 5U can be understood as 5 inches.

[0082] Therefore, the technical problem in the related technology that the server is too high, causing the server to occupy a larger height space in the computer room and affecting the layout of the server in the computer room can be solved, and the technical effect of lowering the server height, reducing the space occupied by the server, and facilitating the reasonable layout of the server in the computer room can be achieved.

[0083] refer to Figure 3 In some embodiments, the first computing module 200 may be a CPU (Central Processing Unit) computing module, and the second computing module 300 may be a GPU (Graphics Processing Unit) computing module.

[0084] In high-performance AI servers, the CPU computing module is responsible for general computing tasks, while the GPU computing module focuses on deep learning and large-scale parallel computing tasks.

[0085] refer to Figure 3 In some embodiments, along the second direction, the lower recess 210 has an opening 211 facing the top of the chassis 100 , and the second computing module 300 does not protrude from the opening 211 of the lower recess 210 .

[0086] The opening 211 may refer to the open area formed in the height direction of the top of the lower recess 210. "Non-protrusion" may mean that the second computing module 300 is completely accommodated in the height direction of the lower recess 210. This can be achieved by controlling the installation height of the second computing module 300 to be flush with or lower than the plane of the opening 211, for example, by using a mounting structure that conforms to the inner wall of the lower recess 210.

[0087] Through the above technical solution, by making the second computing module 300 not protrude from the opening 211 of the lower recess 210, it is possible to avoid the second computing module 300 increasing the space in the height direction of the chassis 100, and thus the height of the chassis 100 can be designed only according to the height of the first computing module 200, thereby reducing the height of the chassis 100 and the height of the server.

[0088] refer to Figure 3 In some embodiments, the server may further include a first handle 160 , which is disposed on an outer side wall of the chassis 100 .

[0089] If the chassis 100 includes a first side wall and a second side wall opposite to each other along the third direction, a plurality of first handles 160 can be respectively provided on the first side wall and the second side wall.

[0090] The first handle 160 provided on the outer sidewall of the chassis 100 provides a convenient grip for the operator. This makes it easier for the operator to carry or move the server. This improves the ease of transport and reduces physical exertion. Furthermore, the design of the first handle 160 makes the server more stable during transport, reducing the risk of damage due to improper handling.

[0091] refer to Figure 4 、 Figure 5 、 Figure 6 、 Figure 7 and Figure 8 In some embodiments, the first computing module 200 may include an expansion unit 220 , a first storage unit 230 , and a first computing unit 240 .

[0092] The extension unit 220 can be arranged closer to the bottom of the chassis 100 relative to the top of the chassis 100 , and the extension unit 220 can play a certain supporting role.

[0093] The expansion unit 220 may refer to a component for providing computing expansion capability, for example, a plurality of expansion units along a third direction (eg, Figure 5 The expansion card 221 is arranged in the Y direction (in the middle Y direction), and the third direction is perpendicular to the first direction and the second direction. The expansion unit 220 can provide additional functional support for the server, such as data transmission or signal processing, by increasing the number of interfaces or processing capabilities.

[0094] Along the second direction, the first storage unit 230 is disposed on a side of the expansion unit 220 that faces the top of the chassis 100. The expansion unit 220 can be used to support the first storage unit 230. The first storage unit 230 can be a module for data storage and can be implemented as a structure comprising multiple storage components, such as hard disks or solid-state drives. The first storage unit 230 is arranged vertically above the expansion unit 220, fully utilizing the vertical space and avoiding interference with the expansion unit 220.

[0095] Along the first direction, the first computing unit 240 is disposed on a side of the first storage unit 230 facing the second window 130 , and the second computing module 300 is disposed on a side of the first computing unit 240 facing the top of the chassis 100 .

[0096] The first computing unit 240 may refer to a component that performs core computing tasks, and may be implemented using a motherboard structure equipped with a processor. The processor may be a CPU, and the processor may be thermally managed through a liquid cooling unit.

[0097] Along the second direction, the surface of the first computing unit 240 facing the top of the chassis 100 is lower than the surface of the first storage unit 230 facing the top of the chassis 100, so that the expansion unit 220, the first storage unit 230, and the first computing unit 240 together form a recessed portion 210. The first computing unit 240 is arranged behind the first storage unit 230 along the front-to-back direction, forming a stepped structure, so that the top of the first computing unit 240 is lower than the top of the first storage unit 230, thereby forming a recessed portion 210 together with the expansion unit 220. The first computing unit 240 can be used to support the second computing module 300. The expansion unit 220 and the first storage unit 230 can protect and limit the side of the second computing module 300 facing the first window 120, preventing the second computing module 300 from displacing relative to the first computing module 200 in the second direction, thereby improving the connection stability between the various components inside the chassis 100.

[0098] Through the above technical solution, the structure of the first computing module 200 is further refined. The height difference formed by the layout of the expansion unit 220, the first storage unit 230 and the first computing unit 240 naturally forms the lower concave portion 210 while ensuring that the height of the chassis 100 is relatively low. No additional support structure is required, which simplifies the assembly process and ensures structural strength.

[0099] This layered structure clearly defines the functional divisions within the first computing module 200. The expansion unit 220, first storage unit 230, and first computing unit 240 each perform distinct functions while working in conjunction with each other. The presence of the recessed portion 210 provides installation space for the second computing module 300, further optimizing space utilization and making the internal structure of the chassis 100 more compact. The coordinated operation of these various components enhances overall performance.

[0100] refer to Figure 5 and Figure 7 In some embodiments, the expansion unit 220 may include multiple expansion cards 221 arranged along a third direction, which is perpendicular to the first and second directions. The third direction may be the width of the chassis 100. The width of the chassis 100 is perpendicular to the front-to-back direction and the height of the chassis 100, and the width is perpendicular to the height of the chassis 100. Specifically, the third direction may be defined by a coordinate axis perpendicular to the front-to-back direction and the height of the chassis 100.

[0101] In some embodiments, expansion card 221 provides additional server functionality, such as data transmission or signal processing, by increasing the number of interfaces or processing power. Expansion can be achieved using full-height PCIE (Peripheral Component Interconnect Express) cards, up to 12 of which are arranged across the width of chassis 100 to enhance the server's computing and expansion capabilities.

[0102] The expansion unit 220 has a plurality of expansion cards 221 arranged transversely along the third direction, forming a stable support structure for the first computing unit 240 inside the chassis 100 .

[0103] The first computing module 200 may further include a second storage unit 250 . The second storage unit 250 is disposed below the first storage unit 230 and between two adjacent expansion cards 221 .

[0104] The second storage unit 250 is an additional storage module located below the first storage unit 230. Specifically, it can be implemented using a structure comprising multiple storage components, such as hard disks or solid-state drives. While ensuring the server's scalability, the second storage unit 250 can be further installed between the two expansion cards 221 to increase storage capacity and optimize space utilization.

[0105] The above technical solution provides a second storage unit 250 between expansion cards 221, fully utilizing the space and further increasing storage capacity. This design improves the server's storage capacity without increasing additional space, meeting the increased storage requirements of high-performance servers while maintaining the overall compactness of the structure.

[0106] refer to Figure 5 and Figure 7 In some embodiments, the first computing module 200 may further include a first housing 260 . The first housing 260 has a first cavity therein for accommodating the expansion unit 220 , the first storage unit 230 , the first computing unit 240 , and the second storage unit 250 .

[0107] The first housing 260 can be designed along the outer contours of the expansion unit 220, the first storage unit 230, and the first computing unit 240 to ensure that the concave portion 210 is formed at the corresponding portion of the first computing unit 240. For example, the first housing 260 can be stepped to ensure that the concave portion 210 is formed.

[0108] The first housing 260 may refer to the support structure that encloses the internal components of the first computing module 200. Specifically, it may be implemented as a combination of a metal frame and panels. Its function is to provide physical protection for the internal components and form a modular unit. The modular structure facilitates overall pull-out maintenance. The first cavity refers to the enclosed space formed within the first housing 260. Specifically, it may be implemented by partitioning the area with partitions. It is used to centrally accommodate the expansion unit 220, the first storage unit 230, and the first computing unit 240, reducing the installation complexity caused by the dispersed layout of the components.

[0109] The above technical solution integrates the expansion unit 220, first storage unit 230, first computing unit 240, and second storage unit 250 into a single cavity through the provision of a first housing 260, achieving a modular design. This design facilitates assembly and maintenance, improves production efficiency, and enhances structural stability. The modular design also makes connections between units more reliable, reduces the risk of failures due to external interference, and improves server reliability.

[0110] In addition, the first shell 260 can also support the second computing module 300 to prevent the first computing unit 240 from being damaged due to direct pressure, thereby improving the protection performance of the first computing unit 240 and other components.

[0111] In some embodiments, a second handle 261 may also be provided on the first shell 260. The second handle 261 is provided on the side of the first shell 260 facing the first window to provide a force point, so that the user can use the second handle 261 to drive the first computing module 200 to be inserted into the accommodating cavity 110 of the chassis 100.

[0112] refer to Figure 7 and Figure 9 In some embodiments, the first storage unit 230 may include a second housing 231 , a plurality of first storage members 233 , a first connecting member 234 , and a second connecting member 235 .

[0113] The second housing 231 has a second cavity 232 therein. The second housing 231 is an independent structure for accommodating a plurality of first storage elements 233 and can be made of metal or plastic. A closed or semi-closed second cavity 232 is formed therein to protect the first storage elements 233 from the external environment.

[0114] A plurality of first storage elements 233 are disposed in the second cavity 232. The first storage element 233 may be a hard disk for storage, and the number of the first storage elements 233 may be four.

[0115] The first connecting member 234 is used to detachably connect the second shell 231 to the first shell 260 , and the second connecting member 235 is used to detachably connect the second shell 231 , the first shell 260 and the chassis 100 .

[0116] Among them, the first connecting member 234 can refer to a component that realizes a detachable connection between the second shell 231 and the first shell 260, and can specifically adopt a screw, a snap or an overlapping structure to facilitate the separate disassembly of the second shell 231 for maintenance or replacement.

[0117] Among them, the second connecting member 235 can refer to a component that simultaneously fixes the second shell 231, the first shell 260 and the chassis 100. Specifically, it can adopt a bracket with threaded holes, an I-nail or a pin structure to ensure the stability of the second shell 231 in the chassis 100 while allowing overall disassembly.

[0118] In a specific implementation, the second shell 231 forms a detachable connection with the first shell 260 through the first connector 234, so that the first storage unit 230 can be installed or removed independently of other components of the first shell 260. The second connector 235 further connects the second shell 231 to the chassis 100 to form a double fixed structure. For example, during maintenance, the lock of the chassis 100 by the second connector 235 can be released first, and then the second shell 231 can be separated from the first shell 260 through the first connector 234, so that the first storage unit 230 can be taken out separately. This design makes it possible to replace or upgrade the storage module without disassembling the entire first computing module 200, reducing the complexity of maintenance. In addition, the independent cavity layout of the second shell 231 can optimize the internal space of the chassis 100 and avoid interference with other components.

[0119] Through the above technical solution, the present application solves the problem of difficult maintenance of server storage modules, improves the replacement efficiency of the first storage unit 230, reduces operation and maintenance costs, and optimizes the internal space utilization of the chassis 100 through modular design, thereby enhancing the overall reliability of the server.

[0120] refer to Figure 5 、 Figure 9 and Figure 10 In some embodiments, the second storage unit 250 may include a third housing 251 , a plurality of second storage elements 253 , and a third connecting element 254 .

[0121] The third housing 251 is disposed between two adjacent expansion cards 221 , and a third cavity 252 is defined inside the second housing 231 .

[0122] Among them, the third shell 251 can refer to a supporting structure for accommodating the second storage unit 253, which can be specifically implemented by combining a metal frame and a plastic cover. The third cavity 252 formed therein provides physical protection for the second storage unit 253.

[0123] A plurality of second storage elements 253 are disposed in the third cavity 252 , and the number of the second storage elements 253 may be eight.

[0124] The second storage element 253 may refer to a hardware device for data storage, which may be implemented as a mechanical hard disk or a solid-state hard disk, and the storage density may be increased by an array arrangement.

[0125] The third connecting member 254 is disposed on a side of the third housing 251 facing the second housing 231 . The third connecting member 254 is configured to be detachably connected to the second housing 231 .

[0126] Among them, the third connecting member 254 can refer to a mechanical structure that realizes a detachable connection between the third shell 251 and the second shell 231, which can be specifically realized by a snap-on method, so that the connection stability between the second storage unit 250 and the first storage unit 230 is higher, and the connection stability between the internal structures of the first computing module 200 can be improved.

[0127] Through the above technical solution, this application achieves spatially optimized integration of the storage unit and the computing unit, effectively increasing storage density within the limited space of the chassis 100 while simplifying maintenance procedures through modular design. The detachable connection structure enables independent maintenance of the second storage unit 250 without affecting other functional modules, enhancing the maintainability and expansion flexibility of the system. The mechanical connection design between the second housing 231 and the third housing 251 also improves the overall structural stability, ensuring that the first storage unit 230 and the second storage unit 250 remain securely fixed when the device vibrates or moves.

[0128] In some embodiments, the third housing 251 may be provided with an input / output component 255 , which may be used for data transmission between the second storage element 253 and an external device.

[0129] refer to Figure 5 、 Figure 11 and Figure 12 In some embodiments, the second computing module 300 may include a fourth housing 310 and a second computing unit 330 .

[0130] The fourth housing 310 has a fourth cavity 320 therein, disposed within the recessed portion 210 and located on the side of the first computing unit 240 facing the top of the chassis 100. The fourth housing 310 may be a container structure for accommodating the second computing unit 330 and may be made of metal or plastic. The fourth cavity 320 is formed within the housing, and a mounting opening is provided on the side facing the top of the chassis 100.

[0131] In some embodiments, the fourth shell 310 can be fixedly connected to the first shell 260 at a position corresponding to the lower recess 210 by snaps or screws to achieve stable installation of the second computing module 300.

[0132] The second computing unit 330 is disposed within the fourth housing 310. The second computing unit 330 may be a hardware module used to assist the first computing unit 240 in processing computing tasks. Specifically, the second computing unit 330 may be implemented using a GPU card or an accelerator card, and may be integrated with the fourth housing 310 via a liquid cooling component to address heat dissipation issues associated with the high-power second computing unit 330.

[0133] In some embodiments, the second computing unit 330 may have multiple GPU cards. For example, the second computing unit 330 may have 8 GPU cards.

[0134] In a specific implementation, the fourth housing 310 is nested within the recess 210 above the first computing unit 240, with its bottom contacting the surface of the first computing unit 240 and its top flush with the inner wall of the chassis 100. The second computing unit 330 is secured within the fourth cavity 320 via a slot or connector and connected to the external cooling system via a liquid cooling line. During installation, the fourth housing 310 fits into the recess 210 along the height of the chassis 100, forming a vertical stacked arrangement between the second computing unit 330 and the first computing unit 240.

[0135] Through the above technical solution, the second computing unit 330 is disposed in the fourth housing 310, achieving a modular design. This design facilitates assembly and maintenance, improves production efficiency, and also enhances structural stability.

[0136] In addition, the fourth shell 310 can also protect the second computing unit 330, reduce the probability of the second computing unit 330 colliding with external components, and improve the protection performance of the second computing unit 330.

[0137] In some embodiments, a third handle 311 may be provided on the fourth shell 310 so that the operator can use the third handle 311 to push the second computing module 300 into the accommodating cavity 110 of the chassis 100, thereby improving the convenience of operation.

[0138] refer to Figure 7 、 Figure 13 、 Figure 14 、 Figure 15 、 Figure 16 and Figure 17 In some embodiments, a first power supply unit 400 and a rear plug-in box module 500 may also be included.

[0139] The first power supply unit 400 is disposed near the first window 120 and is located below the first computing module 200 in the second direction.

[0140] Among them, the first power supply unit 400 refers to a module that provides main power input for the server, which can be specifically implemented using a 54V power module. Its arrangement close to the first window 120 can shorten the power supply path, reduce cable length and power loss, and optimize the layout by utilizing the front space of the chassis 100.

[0141] The rear plug-in box module 500 is disposed near the second window 130 and is electrically connected to the first power supply unit 400 . The rear plug-in box module 500 is also electrically connected to the first computing module 200 and the second computing module 300 .

[0142] The rear plug-in box module 500 may refer to a modular component integrating power distribution and signal transmission functions, which is disposed near the second window 130 to facilitate connection with an external power supply system and simplify the complexity of internal wiring.

[0143] The above technical solution achieves centralized power management and distribution. The first power supply unit 400 is positioned near the first window 120, facilitating access and maintenance. The rear plug-in module 500, positioned near the second window 130, is electrically connected to the first power supply unit 400 and, in turn, to the first computing module 200 and the second computing module 300. This layout streamlines power distribution, reduces the length and complexity of power cables, and improves the stability and reliability of the power supply. This design also facilitates maintenance and replacement of the power modules, enhancing the server's maintainability.

[0144] In some embodiments, the first power supply unit 400 and the rear plug-in box module 500 can be plugged in by blind plugging, which can reduce the connection harness between the first power supply unit 400 and the rear plug-in box module 500 and the complexity of the internal wiring of the chassis 100.

[0145] refer to Figures 13 to 15 In some embodiments, the rear plug-in box module 500 may include a fifth housing 510 , a mid-back plate unit 530 , and a power distribution unit 540 .

[0146] The fifth housing 510 has a fifth cavity 520 therein, and a first port 511 and a second port 512 disposed opposite to each other along a first direction and communicating with the fifth cavity 520 . The first port 511 faces the second window 130 relative to the second port 512 .

[0147] Among them, the fifth shell 510 can refer to a supporting structure for accommodating the internal components of the rear plug-in box module 500, which can be specifically implemented by combining a metal frame and a sheet metal shell. The internal fifth cavity 520 is used to support the mid-backplane unit 530 and the power distribution unit 540.

[0148] The middle backplane unit 530 is arranged in the fifth cavity 520 and close to the first port 511. Along the second direction, there is a first gap 513 between the middle backplane unit 530 and the inner bottom wall of the fifth shell 510. The middle backplane unit 530 is electrically connected to the first computing module 200 and the second computing module 300 respectively.

[0149] The mid-backplane unit 530 may be a component that integrates power and signal transmission functions and is positioned near the first port 511 to facilitate connection to an external power supply unit. The first gap 513 may be the space between the bottom of the mid-backplane unit 530 and the bottom of the fifth housing 510, and may be used to accommodate the power distribution unit 540.

[0150] The power distribution unit 540 is inserted into the first compartment 513 along the first direction from the first port 511 and electrically connected to the mid-backplane unit 530 . The power distribution unit 540 may include a first power connector 541 for plugging with the first power supply unit 400 .

[0151] Among them, the power distribution unit 540 can refer to a functional module that realizes power conversion and distribution, and can adopt an independent drawer structure with a plug-in connector, so that the power distribution unit 540 can be installed in the first compartment 513 in a pull-out manner, and can realize blind plugging between the power distribution unit 540 and the mid-backplane unit 530, thereby reducing the number of wiring harnesses and simplifying the complexity of the internal structure of the chassis 100.

[0152] In some embodiments, if the first power supply unit 400 is a 54V power module, the first power connector 541 can be a 54V power connector compatible with the 54V power module.

[0153] Through the above technical solution, a first gap 513 is set between the middle backplane unit 530 of the rear plug-in box module 500 and the fifth shell 510, and the power distribution unit 540 is inserted into this gap. This design realizes a flexible connection between the power distribution unit 540 and the middle backplane unit 530. The power distribution unit 540 is plugged into the first power supply unit 400 through the first power connector 541, realizing efficient transmission and distribution of power. This modular design allows the power distribution unit 540 and the middle backplane unit 530 to be replaced and maintained independently, improving the reliability and maintainability of the system. At the same time, the electrical connection between the middle backplane unit 530 and the first computing module 200 and the second computing module 300 realizes stable transmission of signals and power, improving the overall performance of the server.

[0154] In some embodiments, a fourth handle 514 may also be provided on the rear insertion module, and the fourth handle 514 is provided on the side of the rear insertion module facing the second window 130, so that the operator can use the fourth handle 514 to insert the rear insertion module from the second window 130 into the accommodating cavity 110 of the chassis 100.

[0155] refer to Figures 13 to 17In some embodiments, the mid-backplane unit 530 may include a second power supply unit 531 and a first plug interface, with the first plug interface being disposed toward the second window 130. The power distribution unit 540 may further include a second power connector 542 and a second connector 543, with the second power connector 542 plugging into the second power supply unit 531 and the second connector 543 plugging into the first plug interface.

[0156] The middle backplane unit 530 may further include a backplane 533 , and the second power supply unit 531 and the first plug interface are both disposed on the backplane 533 .

[0157] In some embodiments, the second power supply unit 531 may be a 12V power supply unit, and the second power connector 542 may be a 12V power strip to accommodate the 12V power supply unit. The second connector 543 may be a busbar connector that can be plugged into the first socket.

[0158] The above technical solution enables a blind-plug electrical connection between the power supply on the midplane 533 and the power distribution unit 540, further optimizing power distribution and management. While ensuring a stable power supply, the blind-plug method reduces the number of connecting wires between the power distribution unit 540 and the midplane unit 530, simplifying the complexity of the internal structure of the chassis 100 and facilitating maintenance and replacement of the power distribution unit 540, thereby improving the maintainability of the system.

[0159] In some embodiments, there is an opening 121 corresponding to the second electrical connector at the first window 120, and the opening 121 is connected to the outside so that external airflow can enter the interior of the chassis 100 through the opening 121 at the first window 120, and can blow toward the second electrical connector on the power distribution unit 540 to facilitate heat dissipation of the second electrical connector.

[0160] In some embodiments, an air duct 544 is also provided on the power distribution unit 540, one end of the air duct 544 corresponds to the opening 121, and the other end corresponds to the second electrical connector. This makes it easy to guide the airflow flowing into the chassis 100 from the opening 121 directly to the second electrical connector, so as to improve the heat dissipation effect of the second electrical connector.

[0161] refer to Figures 13 to 17 In some embodiments, the first computing module 200 may include a third connector 270 , and the side of the mid-backplane unit 530 facing the mid-backplane unit 530 may also include a second plug-in port 532 , and the second plug-in port 532 is plugged into the third connector 270 .

[0162] The third connector 270 may be a high-density link cable head, which can be blind-plugged into the second plug-in port 532 on the mid-backplane unit 530 .

[0163] Through the above technical solution, blind insertion between the first computing module 200 and the rear plug-in box module 500 can be achieved, the wiring harness connection between the first computing module 200 and the rear plug-in box module 500 can be reduced, the complexity of the internal structure of the chassis 100 is simplified, and the assembly between the first computing module 200 and the rear plug-in box module 500 is facilitated, which helps to improve the assembly efficiency of the server.

[0164] The second computing module 300 may further include a fourth connector on a side facing the mid-backplane unit 530. The mid-backplane unit 530 may further include a third plug interface, which is plugged into the fourth connector.

[0165] The fourth connector may also be a high-density connector cable head, which can be plugged into the third plug interface on the mid-backplane unit 530 .

[0166] Through the above technical solution, blind insertion between the second computing module 300 and the rear plug-in box module 500 can be achieved, the wiring harness connection between the second computing module 300 and the rear plug-in box module 500 can be reduced, the complexity of the internal structure of the chassis 100 is simplified, and the assembly between the second computing module 300 and the rear plug-in box module 500 is facilitated, which helps to improve the assembly efficiency of the server.

[0167] refer to Figures 13 to 17 In some embodiments, the rear plug-in box module 500 may further include an air cooling unit 550 , which is disposed on a side of the middle back plate unit 530 facing the second window 130 .

[0168] The air cooling unit 550 may refer to a radiator structure for air cooling and heat dissipation for the chassis 100 .

[0169] The first computing module 200 may further include a first computing unit 240 and a first liquid cooling unit 280 . The first liquid cooling unit 280 is used to dissipate heat for the first computing unit 240 .

[0170] In some embodiments, if the first computing module 200 includes a first computing unit 240, the first liquid cooling unit 280 can be used to dissipate heat for the first computing unit 240.

[0171] The second computing module 300 may further include a second computing unit 330 and a second liquid cooling unit 340 . The second liquid cooling unit 340 is used to dissipate heat for the second computing unit 330 .

[0172] In some embodiments, if the second computing module 300 includes a second computing unit 330 , the second liquid cooling unit 340 can be used to dissipate heat for the second computing unit 330 .

[0173] The above technical solution utilizes the air cooling unit 550, the first liquid cooling unit 280, and the second liquid cooling unit 340 simultaneously, achieving a cooling method that combines both air and liquid cooling. The first computing module 200 and the second computing module 300 are each equipped with a liquid cooling unit, effectively reducing the temperature of key components and improving the stability and reliability of the server under high-load operation. Simultaneously, the air cooling unit 550 of the rear plug-in box module 500 provides sufficient cooling support for components that do not require liquid cooling, ensuring the overall cooling effect of the server. This cooling method not only meets the needs of high-performance computing, but also optimizes cooling efficiency and extends the service life of the server.

[0174] In some embodiments, the second computing module 300 may further include a liquid leakage detection device for detecting liquid leakage in the second liquid cooling unit in the second computing module to ensure that the second liquid cooling unit 340 provides heat dissipation to the second computing unit 330. The liquid leakage detection device includes a liquid leakage detection connector 350 that can be blindly plugged into the midplane unit 530 to facilitate data transmission within the liquid leakage detection device.

[0175] refer to Figure 8 、 Figure 12 、 Figures 13 to 17 In some embodiments, the first liquid cooling unit 280 and the second liquid cooling unit 340 are provided independently of each other.

[0176] The first and second liquid cooling units 280 and 340 are independently configured. This design prevents interference between the cooling systems of the two computing modules. If a cooling issue occurs in one computing module, it will not affect the cooling efficiency of the other, thus improving system reliability. Furthermore, the independent cooling units can be optimized based on the cooling requirements of each computing module, improving cooling efficiency and ensuring server stability during high-performance operation.

[0177] refer to Figure 8 and Figure 17 In some embodiments, the first liquid cooling unit 280 may include a first cold plate 281 and a first joint assembly 282 connected to each other.

[0178] The first cold plate 281 is disposed on a side of the first computing unit 240 facing the top of the chassis 100 and is used to dissipate heat from the first computing unit 240. The first connector assembly 282 is disposed on a side of the first computing unit 240 facing the second window 130 and extends through the rear insert module 500 to the outside of the second window 130.

[0179] In some embodiments, a pipeline connecting the first cold plate 281 and the first joint assembly 282 may be further included, and the pipeline can be used to allow the flow of cooling liquid.

[0180] In some embodiments, when the first connector assembly 282 passes through the rear plug-in box module 500 , there may be an avoidance area on the rear plug-in box module 500 . The first connector assembly 282 passes through the avoidance area to avoid layout interference with other structures on the rear plug-in box module 500 .

[0181] Through the above technical solution, the first cold plate 281 is positioned on the side of the first computing unit 240 facing the top of the chassis 100, providing direct heat dissipation support for key components and effectively reducing their temperature. The first connector assembly 282 extends beyond the second window 130, facilitating connection to an external cooling system and enabling coolant circulation. This design not only improves heat dissipation efficiency but also makes the first computing unit 240's heat dissipation system more flexible and easier to maintain and replace.

[0182] In some embodiments, the first joint assembly 282 may include a first bracket 2821 and a first joint 2822. The first bracket 2821 is arranged on the first shell 260. The first bracket 2821 can be used to fix the first joint 2822 to prevent the first joint 2822 from tilting or shaking, thereby ensuring the stability of the input and output of the coolant.

[0183] refer to Figure 12 and Figure 17 In some embodiments, the second liquid cooling unit 340 may include a second cold plate and a second joint assembly 341 connected to each other.

[0184] The second cold plate is disposed on a side of the second computing unit 330 facing the top of the chassis 100 , and is used to dissipate heat for the second computing unit 330 .

[0185] The second joint assembly 341 is connected to the second cold plate. The second joint assembly 341 is disposed on a side of the second cold plate facing the second window 130 and extends to the outside of the second window 130. The third joint assembly 342 is connected to the second cold plate and spaced apart from the second joint assembly 341. The third joint assembly 342 is disposed on a side of the second cold plate facing the second window 130 and extends to the outside of the second window 130.

[0186] In some embodiments, a pipeline connecting the second cold plate and the second connector assembly 341 and a pipeline connecting the second cold plate and the third connector assembly 342 may also be included. These pipelines can be used to flow coolant. By providing dual pipelines in the second cold plate, the flow rate of coolant within the second cold plate can be increased, improving the heat dissipation capacity of the second cold plate and, in turn, the heat dissipation effect of the second computing module 300.

[0187] In some embodiments, when the second connector assembly 341 and the third connector assembly 342 pass through the rear plug-in box module 500, an avoidance area may exist on the rear plug-in box module 500. The second connector assembly 341 and the third connector assembly 342 pass through the avoidance area, thereby avoiding layout interference with other structures on the rear plug-in box module 500.

[0188] In some embodiments, the second joint assembly 341 may include a second bracket 3411 and a second joint 3412. The second bracket 3411 is arranged on the fourth shell 310. The second bracket 3411 can be used to fix the second joint 3412 to prevent the second joint 3412 from tilting or shaking, thereby ensuring the stability of the input and output of the coolant.

[0189] In some embodiments, the third joint assembly 342 may include a third bracket 3421 and a third joint 3422. The second bracket 3411 is arranged on the fourth shell 310. The third bracket 3421 can be used to fix the third joint 3422 to prevent the third joint 3422 from tilting or shaking, thereby ensuring the stability of the input and output of the coolant.

[0190] refer to Figure 17 In some embodiments, the air cooling unit 550 may include a mounting plate and a plurality of fans 551. The mounting plate is disposed on a side of the mid-back plate unit 530 facing the second window 130, and the mounting plate can provide a mounting point and structural support for the fans 551.

[0191] A plurality of fans 551 are mounted on a side of the mounting plate facing away from the mid-back plate unit 530, and the fans 551 are electrically connected to the mid-back plate unit 530. The fans 551 are arranged along a third direction perpendicular to the first direction and the second direction.

[0192] By providing a mounting plate and multiple fans 551, air cooling is achieved for the rear plug-in box module 500. Fans 551 are arranged along the third direction, providing sufficient cooling support for components that do not require liquid cooling, ensuring effective cooling for the entire server. This coexistence of air and liquid cooling not only meets the needs of high-performance computing, but also optimizes cooling efficiency and extends the server's service life. Furthermore, the arrangement of fans 551 ensures more uniform heat dissipation, improving system stability.

[0193] Furthermore, the fans 551 are staggered with the connector assemblies, so that the number of fans 551 is not sacrificed while satisfying the water supply needs of multiple cold plates at different positions in the chassis 100, and sufficient heat dissipation support is provided for non-liquid cooling components.

[0194] In some implementations, there may be five fans 551 , and the five fans 551 are arranged along the third direction.

[0195] refer to Figure 8 、 Figure 12 and Figure 17 In some embodiments, the rear insert box module 500 may further include a first baffle 560 . The first baffle 560 is positioned adjacent to the second window 130 . The first baffle 560 and one of the fans 551 are arranged along the second direction. The first baffle 560 can be positioned in the aforementioned avoidance area. The first baffle 560 may include a first retaining hole 561 through which the first connector assembly 282 passes. The first baffle 560 is used to secure the first connector assembly 282 that passes through the first retaining hole 561 .

[0196] The first baffle 560 may be a plastic plate or a metal plate. For example, the first baffle 560 may be a metal plate, which can improve the structural strength and durability of the first baffle 560 .

[0197] The first stopper 560 and the corresponding first retaining hole 561 secure the first connector assembly 282. This design prevents the first connector assembly 282 from loosening or shifting during operation, ensuring stable coolant flow and improving the reliability of the cooling system. Furthermore, the provision of the first stopper 560 facilitates maintenance and replacement of the first connector assembly 282, enhancing the maintainability of the system.

[0198] refer to Figure 8 、 Figure 12 and Figure 17 In some embodiments, the rear plug-in box module 500 may further include a second baffle 570 and a third baffle 580. The second baffle 570 and the third baffle 580 are arranged near the second window 130. The second baffle 570 and the third baffle 580 are arranged at intervals along the third direction. The second baffle 570 and the third baffle 580 are both arranged along the second direction with the fan 551.

[0199] The material of the second baffle 570 and the third baffle 580 can be the same as that of the first baffle 560. For example, the second baffle 570 and the third baffle 580 can be metal plates, which can improve the structural strength and durability of the second baffle 570 and the third baffle 580.

[0200] The second blocking piece 570 may include a second limiting hole 571 through which the second connector assembly 341 can pass. The second blocking piece 570 is used to fix the second connector assembly 341 passing through the second limiting hole 571 .

[0201] The third blocking piece 580 may include a third limiting hole 581 through which the third connector assembly 342 can pass. The third blocking piece 580 is used to fix the third connector assembly 342 passing through the third limiting hole 581 .

[0202] The second and third baffles 570, 580, and corresponding retaining holes secure the second and third connector assemblies 341, 342. This design prevents the second and third connector assemblies 341, 342 from loosening or shifting during operation, ensuring stable coolant flow and improving the reliability of the cooling system. Furthermore, the provision of the second and third baffles 570, 580 facilitates maintenance and replacement of the second and third connector assemblies 341, 342, enhancing the system's maintainability.

[0203] refer to Figure 3 、 Figure 5 and Figure 18 In some embodiments, the server may further include a first sliding mechanism 600 . A portion of the first sliding mechanism 600 is disposed on the chassis 100 , and another portion is disposed on the first computing module 200 . The first sliding mechanism 600 is used to slide the first computing module 200 into or out of the accommodating cavity 110 .

[0204] The first sliding mechanism 600 may refer to a structure for guiding the first computing module 200 to be inserted into the receiving cavity 110 of the chassis 100. For example, the first sliding mechanism 600 may be a combination of a track and a roller, or a combination of a slide groove and a slider.

[0205] Through the above technical solution, by providing a first sliding mechanism 600 between the chassis 100 and the first computing module 200, the traditional fixed installation method is replaced with a sliding installation method. This design utilizes the principle of relatively low sliding friction, allowing the first computing module 200 to slide easily into and out of the accommodating cavity 110 of the chassis 100. This greatly improves the efficiency of installing and removing the first computing module 200, reducing the workload and time cost during maintenance. This design also makes maintenance of the first computing module 200 more convenient, as it can be performed without disassembling other components, improving the maintainability of the system.

[0206] refer to Figure 3 、 Figure 5 and Figure 18 In some embodiments, the chassis 100 may include a first side panel 140 and a second side panel 150 that are opposite to each other along a third direction, where the third direction is perpendicular to the first and second directions. The third direction may be the width direction of the chassis 100 , and the first side panel 140 and the second side panel 150 are two opposite side panels in the width direction of the chassis 100 .

[0207] The first sliding mechanism 600 may include a first pulley assembly, a second pulley assembly 610, a first slide rail 620, and a second slide rail. The first pulley assembly is disposed on the side of the first side plate 140 facing the accommodating cavity 110, and the second pulley assembly 610 is disposed on the side of the second side plate 150 facing the accommodating cavity 110.

[0208] The first slide rail 620 is arranged on the surface of the first computing module 200 facing the first side panel 140. The first slide rail 620 cooperates with the first roller group. The first roller group can roll along the first direction in the first slide rail 620 and drive the first computing module 200 to slide in or out of the accommodating cavity 110.

[0209] In some embodiments, the first roller assembly may include a plurality of rollers spaced apart along a first direction, and each roller is capable of rotating relative to the first computing module 200 .

[0210] If the first computing module 200 includes a first housing 260 , the first roller assembly can be disposed on a side of the first housing 260 facing the first side plate 140 .

[0211] The second slide rail is arranged on the surface of the first computing module 200 facing the second side plate 150. The second slide rail cooperates with the second roller group. The second roller group can roll along the first direction in the second slide rail and drive the first computing module 200 to slide in or out of the accommodating cavity 110.

[0212] In some embodiments, the second roller assembly may include a plurality of rollers spaced apart along the first direction, and each roller is capable of rotating relative to the first computing module 200 .

[0213] If the first computing module 200 includes a first housing 260 , the second roller assembly can be disposed on a side of the first housing 260 facing the second side plate 150 .

[0214] Through the above technical solution, roller assemblies are provided on both sides of the chassis 100, and slide rails are provided on both sides of the first computing module 200, forming a sliding fit. The roller assemblies convert sliding friction into rolling friction, which is much smaller than sliding friction, thereby reducing the resistance of the first computing module 200 during the sliding process.

[0215] The sliding performance of the first computing module 200 is further optimized, making it smoother and reducing wear when sliding in or out of the chassis 100. At the same time, this design also improves the stability and reliability of the system and avoids component damage or jamming caused by excessive friction.

[0216] refer to Figure 3 、 Figure 11 and Figure 18In some embodiments, the server may further include a second sliding mechanism 700 , a portion of which is disposed on the chassis 100 , and another portion of which is disposed on the second computing module 300 . The second sliding mechanism 700 is used to slide the second computing module 300 into or out of the accommodating cavity 110 .

[0217] The second sliding mechanism 700 may refer to a structure for guiding the second computing module 300 to be inserted into the receiving cavity 110 of the chassis 100. For example, the second sliding mechanism 700 may be a combination of a track and a roller, or a combination of a slide groove and a slider.

[0218] Through the above technical solution, a second sliding mechanism 700 is provided between the chassis 100 and the second computing module 300. Similar to the first sliding mechanism 600, this mechanism enables the second computing module 300 to slide in and out. This design also utilizes the principle that rolling friction is less than sliding friction, reducing the resistance of the second computing module 300 during the sliding process. This makes the installation and removal of the second computing module 300 more convenient, further improving the maintainability of the entire server system. At the same time, this design also makes the maintenance of the second computing module 300 more independent, and it can be operated without disassembling the first computing module 200, thereby improving maintenance efficiency.

[0219] refer to Figure 3 、 Figure 11 and Figure 18 In some embodiments, the second sliding mechanism 700 may include a third pulley set, a fourth pulley set 710 , a third slide rail 720 , and a fourth slide rail.

[0220] The third pulley set is disposed on a side of the first side plate 140 facing the interior of the accommodating cavity 110 , and the fourth pulley set 710 is disposed on a side of the second side plate 150 facing the interior of the accommodating cavity 110 .

[0221] The third slide rail 720 is arranged on the surface of the second computing module 300 facing the first side panel 140. The third slide rail 720 cooperates with the third roller group. The third roller group can roll along the first direction in the third slide rail 720 and drive the second computing module 300 to slide in or out of the accommodating cavity 110.

[0222] In some embodiments, the third roller group may include a plurality of rollers spaced apart along the first direction, and each roller is capable of rotating relative to the second computing module 300 .

[0223] If the second computing module 300 includes a fourth housing 310 , the third roller assembly can be disposed on a side of the fourth housing 310 facing the first side plate 140 .

[0224] The fourth slide rail is arranged on the surface of the second computing module 300 facing the second side plate 150. The fourth slide rail cooperates with the fourth roller group. The fourth roller group can roll along the first direction in the fourth slide rail and drive the second computing module 300 to slide in or out of the accommodating cavity 110.

[0225] In some embodiments, the fourth roller group may include a plurality of rollers spaced apart along the first direction, and each roller is capable of rotating relative to the second computing module 300 .

[0226] If the second computing module 300 includes a fourth housing 310 , the fourth roller assembly can be disposed on a side of the fourth housing 310 facing the second side plate 150 .

[0227] Through the above technical solution, a third roller assembly and a fourth roller assembly are respectively provided on either side of the chassis 100, while a third slide rail 720 and a fourth slide rail are provided on either side of the second computing module 300, forming a sliding fit. This design, similar to the first sliding mechanism 600, further optimizes the sliding performance of the second computing module 300. This makes the second computing module 300 more stable and smooth when sliding in or out of the chassis 100, reducing component wear caused by sliding friction. This design also improves the overall reliability of the system and ensures the stability of the second computing module 300 during operation.

[0228] The above is a detailed introduction to a server provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only intended to help understand the method and core ideas of the present application. It should be noted that, for those skilled in the art, without departing from the principles of the present application, several improvements and modifications may be made to the present application, and such improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A server, characterized in that: include: A chassis (100) having an accommodating cavity (110) therein, and a first window (120) and a second window (130) opposite to and connected to the accommodating cavity (110) along a first direction; the first direction is consistent with the extension direction of the front and rear windows of the chassis (100); A first computing module (200) is arranged on a side of the chassis (100) close to the first window (120), and a lower recess (210) is formed on a side of the first computing module (200) facing the top of the chassis (100) along a second direction, wherein the second direction is a height direction of the chassis (100); a second computing module (300) disposed in the chassis (100) and installed in the recessed portion (210), the second computing module (300) being used to assist the first computing module (200) in processing computing tasks; The server further includes a first power supply unit (400) and a rear plug-in box module (500); The rear plug-in box module (500) comprises: a fifth housing (510) having a fifth cavity (520) therein, and a first port (511) and a second port (512) arranged opposite to each other along the first direction and communicating with the fifth cavity (520), wherein the first port (511) faces the second window (130) relative to the second port (512); a middle backplane unit (530) disposed in the fifth cavity (520) and close to the first port (511); along the second direction, a first gap (513) is provided between the middle backplane unit (530) and the inner bottom wall of the fifth housing (510); the middle backplane unit (530) is electrically connected to the first computing module (200) and the second computing module (300); A power distribution unit (540) is inserted into the first compartment (513) along the first direction through the first port (511) and is electrically connected to the mid-backplane unit (530); The power distribution unit (540) comprises a first power connector (541) for plugging into the first power supply unit (400).

2. The server according to claim 1, wherein: The first computing module (200) includes: Extension unit (220); a first storage unit (230), arranged along the second direction on a side of the expansion unit (220) facing the top of the chassis (100); A first computing unit (240), along the first direction, the first computing unit (240) is arranged on a side of the first storage unit (230) facing the second window (130), and the second computing module (300) is arranged on a side of the first computing unit (240) facing the top of the chassis (100); Wherein, along the second direction, the surface of the first computing unit (240) facing the top of the chassis (100) is lower than the surface of the first storage unit (230) facing the top of the chassis (100), so that the expansion unit (220), the first storage unit (230) and the first computing unit (240) together form the recessed portion (210).

3. The server according to claim 2, wherein: The expansion unit (220) comprises a plurality of expansion cards (221) arranged along a third direction, wherein the third direction is perpendicular to the first direction and the second direction; The first computing module (200) further includes: The second storage unit (250) is arranged below the first storage unit (230) and located between two adjacent expansion cards (221).

4. The server according to claim 3, wherein: The first computing module (200) further includes: The first shell (260) has a first cavity inside, and the first cavity is used to accommodate the expansion unit (220), the first storage unit (230), the first computing unit (240) and the second storage unit (250).

5. The server according to claim 4, wherein: The first storage unit (230) comprises: A second shell (231) having a second cavity (232) therein; A plurality of first storage elements (233) are disposed in the second cavity (232); a first connecting member (234) for detachably connecting the second shell (231) to the first shell (260); The second connecting member (235) is used to detachably connect the second shell (231), the first shell (260) and the chassis (100).

6. The server according to claim 5, wherein: The second storage unit (250) comprises: A third housing (251) is disposed between two adjacent expansion cards (221), and a third cavity (252) is provided inside the second housing (231); A plurality of second storage elements (253) are disposed in the third cavity (252); A third connecting member (254) is provided on a side of the third shell (251) facing the second shell (231), and the third connecting member (254) is used for detachably connecting with the second shell (231).

7. The server according to claim 1, wherein: The second computing module (300) includes: a fourth housing (310) having a fourth cavity (320) therein, disposed in the recessed portion (210) and located on a side of the first computing module (200) facing the top of the chassis (100); The second calculation unit (330) is arranged in the fourth housing (310).

8. The server according to claim 1, wherein: Along the second direction, the concave portion (210) has an opening (211) facing the top of the chassis (100), and the second computing module (300) does not protrude from the opening (211) of the concave portion (210).

9. The server according to claim 1, wherein: Also includes: The first power supply unit (400) is disposed near the first window (120) and is located below the first computing module (200) in the second direction; The rear plug-in box module (500) is arranged near the second window (130) and is electrically connected to the first power supply unit (400), and the rear plug-in box module (500) is electrically connected to the first computing module (200) and the second computing module (300) respectively.

10. The server according to claim 9, wherein: The middle backboard unit (530) comprises a second power supply unit (531) and a first plug interface, wherein the first plug interface is arranged toward one side of the second window (130); The power distribution unit (540) further comprises a second power connector (542) and a second connector (543), wherein the second power connector (542) is plugged into the second power supply unit (531), and the second connector (543) is plugged into the first socket.

11. The server according to claim 9, wherein: The first computing module (200) includes a third connector (270); The side of the mid-backplane unit (530) facing the mid-backplane unit (530) further includes a second plug-in interface (532), and the second plug-in interface (532) is plugged into the third connector (270); and / or, The second computing module (300) further includes a fourth connector on a side facing the mid-backplane unit (530); The mid-backplane unit (530) further comprises a third plug-in interface, and the third plug-in interface is plugged into the fourth connector.

12. The server according to claim 9, wherein: The rear plug-in box module (500) further includes an air cooling unit (550); The air cooling unit (550) is arranged on a side of the middle back plate unit (530) facing the second window (130); The first computing module (200) further includes a first computing unit (240) and a first liquid cooling unit (280), wherein the first liquid cooling unit (280) is used to dissipate heat for the first computing unit (240); The second computing module (300) further includes a second computing unit (330) and a second liquid cooling unit (340), wherein the second liquid cooling unit (340) is used to dissipate heat for the second computing unit (330).

13. The server according to claim 12, wherein: The first liquid cooling unit (280) and the second liquid cooling unit (340) are arranged independently of each other.

14. The server according to claim 13, wherein: The first liquid cooling unit (280) includes a first cold plate (281) and a first joint assembly (282) connected to each other; The first cold plate (281) is arranged on a side of the first computing unit (240) facing the top of the chassis (100), and the first cold plate (281) is used to dissipate heat for the first computing unit (240); The first joint assembly (282) is arranged on a side of the first computing unit (240) facing the second window (130), and extends through the rear insert box module (500) to the outside of the second window (130).

15. The server according to claim 14, wherein: The second liquid cooling unit (340) comprises: a second cold plate, arranged on a side of the second computing unit (330) facing the top of the chassis (100), the second cold plate being used to dissipate heat for the second computing unit (330); A second joint assembly (341) is connected to the second cold plate, the second joint assembly (341) is arranged on a side of the second cold plate facing the second window (130), and extends through the rear plug-in box module (500) to the outside of the second window (130); A third joint assembly (342) is connected to the second cold plate and spaced apart from the second joint assembly (341). The third joint assembly (342) is arranged on a side of the second cold plate facing the second window (130) and extends through the rear plug-in box module (500) to the outside of the second window (130).

16. The server according to claim 15, wherein: The air cooling unit (550) comprises: a mounting plate, arranged on a side of the middle back plate unit (530) facing the second window (130); A plurality of fans (551) are installed on a side of the installation plate facing away from the middle back plate unit (530), and the fans (551) are electrically connected to the middle back plate unit (530); The plurality of fans (551) are arranged along a third direction, which is perpendicular to the first direction and the second direction.

17. The server according to claim 16, wherein: The rear insert box module (500) further comprises a first baffle (560), the first baffle (560) being arranged close to the second window (130), and the first baffle (560) and one of the fans (551) being arranged along the second direction; The first blocking piece (560) comprises a first limiting hole (561) for the first joint assembly (282) to pass through, and the first blocking piece (560) is used to fix the first joint assembly (282) passing through the first limiting hole (561); and / or, The rear insert box module (500) further comprises a second baffle (570) and a third baffle (580), wherein the second baffle (570) and the third baffle (580) are arranged close to the second window (130), the second baffle (570) and the third baffle (580) are arranged at intervals along the third direction, and the second baffle (570) and the third baffle (580) are arranged along the second direction with the fan (551); The second blocking piece (570) comprises a second limiting hole (571) for the second joint assembly (341) to pass through, and the second blocking piece (570) is used to fix the second joint assembly (341) passing through the second limiting hole (571); The third blocking piece (580) comprises a third limiting hole (581) through which the third joint assembly (342) can pass, and the third blocking piece (580) is used to fix the third joint assembly (342) passing through the third limiting hole (581).

18. The server according to claim 1, wherein: Also includes a first sliding mechanism (600); A portion of the first sliding mechanism (600) is disposed on the chassis (100), and another portion is disposed on the first computing module (200). The first sliding mechanism (600) is used to enable the first computing module (200) to slide into or out of the accommodating cavity (110).

19. The server according to claim 18, wherein: The chassis (100) comprises: a first side plate (140) and a second side plate (150) opposite to each other along a third direction, wherein the third direction is perpendicular to the first direction and the second direction; The first sliding mechanism (600) comprises: a first pulley assembly, disposed on a side of the first side plate (140) facing the interior of the accommodating cavity (110); a second pulley assembly (610), disposed on a side of the second side plate (150) facing the interior of the accommodating cavity (110); A first slide rail (620) is provided on a surface of the first computing module (200) facing the first side panel (140); The second slide rail is arranged on the surface of the first computing module (200) facing the second side plate (150).

20. The server according to claim 19, wherein: Also includes a second sliding mechanism (700); A portion of the second sliding mechanism (700) is disposed on the chassis (100), and another portion is disposed on the second computing module (300). The second sliding mechanism (700) is used to enable the second computing module (300) to slide into or out of the accommodating cavity (110).

21. The server according to claim 20, wherein: The second sliding mechanism (700) comprises: a third pulley assembly, disposed on a side of the first side plate (140) facing the interior of the accommodating cavity (110); a fourth pulley assembly (710), disposed on a side of the second side plate (150) facing the interior of the accommodating cavity (110); a third slide rail (720) disposed on a surface of the second computing module (300) facing the first side panel (140); A fourth slide rail is provided on the surface of the second computing module (300) facing the second side plate (150).

22. The server according to claim 1, wherein: It also includes a first handle (160), which is arranged on the outer side wall of the chassis (100).

23. The server according to claim 1, wherein: The first computing module (200) is a CPU computing module; The second computing module (300) is a GPU computing module.

Citation Information

Patent Citations

  • GPU server

    CN118939091A