Dual-channel heat dissipation computing power server

Abstract

The utility model discloses a double -channel heat dissipation computing power server, including shell and node module, two node modules are in -line nested in the shell, the node module is equipped with mesh towards the shell outside, the node module is close to the first fan of inboard wall of shell, the first fan side wall with node module side wall is arranged with the first angle, two node modules the side wall with shell forms two heat dissipation space, the shell is equipped with heat dissipation window with heat dissipation space position. This kind of double -channel heat dissipation computing power server, through setting node module, heat dissipation window, mainboard area and GPU extension area, make the computing power server when operating, the front node module uses the air duct structure of front air intake, side air exhaust, the rear node module uses the air duct structure of rear air intake, side air exhaust, effectively avoid heat accumulation, improve heat dissipation efficiency greatly.

Description

Technical Field

[0001] This utility model relates to the field of server equipment technology, specifically a dual-channel heat dissipation computing server. Background Technology

[0002] With the rapid development of information technology, especially the high demand for edge computing power, computing servers, as core node devices for data processing and storage, face increasingly higher requirements for performance and stability. Particularly in fields such as artificial intelligence, big data processing, and cloud computing, servers need to possess high-performance computing capabilities and excellent heat dissipation to ensure stable operation over extended periods. However, traditional server designs often fall short in terms of heat dissipation, especially during high-density deployments and heavy-load operation, where heat dissipation issues become particularly prominent.

[0003] Existing servers typically employ a single-airflow cooling design, meaning all heat is exhausted through a fan at the rear of the chassis. This design can easily lead to heat buildup and low cooling efficiency when handling high-heat components that operate for extended periods, thus affecting the server's stability and lifespan.

[0004] Therefore, it is particularly important to design a server structure that can achieve efficient heat dissipation and easy maintenance within the standard server frame height. Utility Model Content

[0005] The purpose of this invention is to provide a dual-channel heat dissipation computing server to solve the problems of low heat dissipation efficiency and inconvenient maintenance of traditional servers mentioned in the background art.

[0006] To achieve the above objectives, this utility model provides the following technical solution: a dual-channel heat dissipation computing server, comprising a shell and node modules, two node modules being nested side-by-side within the shell, each node module having mesh openings facing outwards from the shell, a first fan being positioned near the inner wall of the shell on each node module, the side wall containing the first fan being arranged at a first angle to the side wall of the node module, the side walls containing the two node modules forming two heat dissipation spaces with the shell, and heat dissipation windows being provided on the shell corresponding to the positions of the heat dissipation spaces.

[0007] Preferably, the first angle is 45 degrees, and the cross-section of the heat dissipation space is triangular.

[0008] Preferably, the two node modules are arranged in a rotationally symmetrical manner, and the heat dissipation space is also arranged in a rotationally symmetrical manner.

[0009] Preferably, the heat dissipation windows are located on three sides of the corresponding heat dissipation space of the outer casing.

[0010] Preferably, the two node modules are slidably assembled inside the housing, with the ends of the two node modules in physical contact.

[0011] Preferably, it also includes a panel, the outer wall of which is provided with a second fan. The airflow channel flows from the second fan and the mesh to the node module, then flows from the first fan to the heat dissipation space, and finally flows out through the heat dissipation window of the outer shell.

[0012] Preferably, the left side of the inner cavity of the two node modules is provided with a motherboard area, the right side of the motherboard area is provided with a GPU expansion area, the motherboard is installed inside the motherboard area, the outer wall of the motherboard is equipped with a CPU, memory module and hard drive, the GPU expansion area is provided with a GPU graphics card, the connection between the GPU graphics card and the motherboard is provided with a PCIe flexible cable, and the inner wall of the motherboard area near the panel is provided with a power and control module.

[0013] Preferably, the mesh air intake channel is located in the motherboard area, the second fan air intake channel is located in the GPU expansion area, a baffle is provided at the separation point between the motherboard area and the GPU expansion area, and an air passage is provided on the outer wall of the baffle.

[0014] Compared with the prior art, the beneficial effects of this utility model are:

[0015] 1. By setting up node modules, heat dissipation windows, motherboard area and GPU expansion area, the front node modules use a front air intake and side air exhaust airflow structure, and the rear node modules use a rear air intake and side air exhaust airflow structure, which effectively avoids heat accumulation and greatly improves heat dissipation efficiency.

[0016] 2. By setting the node modules as removable independent operating units, supporting hot-swapping and mutual non-interference, the availability and maintenance efficiency of the server are improved. The device adopts a symmetrical double-slot structure, with each slot corresponding to a node module, which facilitates future expansion and upgrades. The hot-swappable and sliding rail structure makes it easy to install and remove the node modules, significantly reducing the difficulty of maintenance. Attached Figure Description

[0017] Figure 1 This is a three-dimensional structural diagram of the present invention;

[0018] Figure 2 This is a top view of the structure of this utility model;

[0019] Figure 3 This is a front view structural diagram of the present invention;

[0020] Figure 4 This is a schematic diagram of the cross-sectional structure of the node module in this utility model.

[0021] In the diagram: 1. Outer shell; 2. Node module; 3. Mesh; 4. First fan; 5. First angle; 6. Heat dissipation space; 7. Heat dissipation window; 8. Front panel; 9. Second fan; 10. Baffle; 11. Air duct; 12. Motherboard area; 13. GPU expansion area; 14. Motherboard; 15. GPU graphics card; 16. PCIe flexible cable; 17. Power supply and control module. Detailed Implementation

[0022] The technical solutions of the present utility model will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present utility model, and not all embodiments. Based on the embodiments of the present utility model, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present utility model.

[0023] Please see Figure 1 , Figure 2 and Figure 4 A dual-channel heat dissipation computing server includes a casing 1 and node modules 2. Two node modules 2 are nested side-by-side within the casing 1, with their ends close together on the same plane. Each node module 2 has mesh openings 3 facing outwards from the casing 1; these mesh openings are multiple rectangular holes arranged side-by-side. A first fan 4 is positioned near the inner wall of the casing 1, and the sidewall containing the first fan 4 forms a first angle 5 with the sidewall of the node module 2. The first angle 5 is preferably 45 degrees, but in other embodiments it can be between 30 and 60 degrees, and in more specific embodiments it can be between 0 and 90 degrees. The sidewalls containing the two node modules 2 and the casing 1 form two heat dissipation spaces 6. Corresponding to the positions of the heat dissipation spaces 6, the casing 1 has heat dissipation windows 7, which are multiple circular holes arranged through the casing 1. Corresponding to the configuration of the first angle 5, the cross-section of the heat dissipation space 6 is triangular. The two node modules 2 are arranged in a rotationally symmetrical manner. Therefore, the heat dissipation space 6 is arranged in a rotationally symmetrical manner. The rotationally symmetrical design means that the nodes are distributed on the shell 1 with the center point of the shell 1 (the center position of the arrangement of the two node modules 2) as the center. The heat dissipation windows 7 are provided on at least one surface of the corresponding heat dissipation space 6 of the shell 1, preferably on the side and the top and bottom surfaces. The two node modules 2 are slidably assembled in the shell 1, and the ends of the two node modules 2 are in physical contact. The shell 1 also includes a panel 8. The outer wall of the panel 8 is provided with a second fan 9. The airflow channel flows from the second fan 9 and the mesh 3 to the node module 2, then flows to the heat dissipation space 6 by the first fan 4, and then flows out through the heat dissipation window 7 of the shell 1.

[0024] By configuring node module 2, heat dissipation window 7, motherboard area 12, and GPU expansion area 13, with a baffle 10 separating the motherboard area 12 and GPU expansion area 13, a basic separation of heat dissipation is achieved. A through air duct 11 is provided on the baffle 10 near the end (first fan) to partially connect the internal airflow. This allows the front node module 2 to use a front intake and side exhaust airflow structure, while the rear node module 2 uses a rear intake and side exhaust airflow structure, effectively preventing heat accumulation and significantly improving heat dissipation efficiency.

[0025] In a more specific implementation, the motherboard area 12 and the GPU expansion area 13 are provided with several heat dissipation holes on the side wall of the casing 1, and some airflow can also passively flow out from the side. However, it is understandable that due to the presence of the first fan, the main heat dissipation channel is still mainly from the heat dissipation space 6.

[0026] This embodiment takes a 2U rackmount chassis as an example. It adopts a front and rear integrated ventilation structure and is designed to install two computing node modules. Each node operates independently and has good heat dissipation performance and maintainability. 2U means two server node modules. This solution adopts a front and rear plug-in form construction. The front and rear are rotationally symmetrical and have the same panel.

[0027] Please see Figure 1 , Figure 3 and Figure 4 The outer casing 1 is made of metal. The left side of the inner cavity of the two node modules houses the motherboard area 12, and the right side of the motherboard area 12 houses the GPU expansion area 13. The motherboard 14 is installed inside the motherboard area 12. The outer wall of the motherboard 14 houses the CPU, memory modules, and hard drive. The GPU expansion area 13 houses the GPU graphics card 15. A PCIe flexible cable 16 connects the GPU graphics card 15 to the motherboard 14. The inner wall of the motherboard area 12 near the panel 8 houses the power and control module 17. The mesh 3 air intake channel is located in the motherboard area 12, and the second fan 9 provides air intake. The channel is located in the GPU expansion area 13. A baffle 10 is provided at the separation between the motherboard area 12 and the GPU expansion area 13. An air channel 11 is opened on the outer wall of the baffle 10. By setting the computing node module 2 as a removable independent operating unit, it supports hot-swapping and does not interfere with each other, thereby improving the availability and maintenance efficiency of the server. The device adopts a left-right symmetrical double-slot structure, with each slot corresponding to a node module 2, which facilitates future expansion and upgrades. The hot-swappable and sliding rail structure facilitates the installation and removal of the node module 2, significantly reducing the maintenance difficulty.

[0028] In summary, the heat dissipation optimization data of this solution compared to the traditional solution are as follows:

[0029] Test environment description:

[0030] Software: AIDA64, FurMark (run for 10 minutes);

[0031] Ambient temperature: Room temperature 26°C;

[0032] Load: CPU single stress test / GPU single stress test / simultaneous dual stress test;

[0033]

[0034] Therefore, it can be seen that the solution does not affect the operation of other node modules through hot-swappable design, making it suitable for edge AI scenarios; the independent dual air ducts for air cooling prevent GPU heat from being conducted to the motherboard; and the space is compressed to a standard 2U height, maintaining a good performance-density ratio.

[0035] Working principle: When staff need to use the server, the Mini-ITX motherboard 8 is fixed on the support pillar in the motherboard area 12. The CPU, memory modules, and hard drive are installed, and the heatsink is installed, ensuring a tight fit with the CPU. Then, in the GPU expansion area 13, the GPU graphics card 15 is installed and connected to the motherboard 14 via the PCIe flexible cable 16. The GPU graphics card 15 is reinforced with a mounting bracket, and a 6-pin or 8-pin power cable is connected. Finally, the modular power supply board is installed inside the chassis, connecting the ATX 24-pin main power supply and the CPU. An 8-pin auxiliary power supply is connected to the motherboard 14, and the GPU graphics card 15 power supply cable is also connected. An intelligent control interface board is installed to support remote management and monitoring. A cable tie is used to organize the power supply and data cables, and they are gathered between the motherboard 14 and the air duct to ensure internal cleanliness and facilitate air circulation. High-power second fans 9 (model 9038) are installed in the GPU expansion area 13. The front node module 2 is for front intake and side exhaust, and the rear node module 2 is for rear intake and side exhaust. The assembled computing node module 2 is pushed into the slot along the slide rail and fixed to the hot-swappable bracket to ensure stability. The assembled server is placed on a standard rack, and the power cord and network cable are connected. The power is turned on, and the two node modules 2 can start independently. The above is the working process of the entire device. All contents not described in detail in this manual are existing technologies known to those skilled in the art.

[0036] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dual-channel heat-dissipation computing server comprising a shell (1) and a node module (2), characterized in that, The two node modules (2) are nested side by side inside the shell (1). The node modules (2) have mesh holes (3) facing the outside of the shell (1). The node modules (2) are provided with a first fan (4) near the inner side wall of the shell (1). The side wall where the first fan (4) is located is arranged at a first angle (5) with the side wall of the node module (2). The side walls where the two node modules (2) are located and the shell (1) form two heat dissipation spaces (6). The shell (1) is provided with heat dissipation windows (7) corresponding to the heat dissipation spaces (6).

2. The dual-channel heat dissipation computing power server according to claim 1, wherein: The heat dissipation space (6) has a triangular cross-section.

3. The dual-channel heat-dissipation computing server of claim 1, wherein: The first angle (5) is forty-five degrees.

4. The dual-channel heat-dissipation computing server of claim 1, wherein: The two node modules (2) are arranged in a rotationally symmetrical manner, and the heat dissipation space (6) is arranged in a rotationally symmetrical manner.

5. The dual-channel heat-dissipation computing server of claim 1, wherein: The heat dissipation window (7) is provided on at least one surface of the corresponding heat dissipation space (6) of the outer shell (1).

6. The dual-channel heat-dissipation computing server of claim 1, wherein: The two node modules (2) are slidably assembled inside the housing (1), and the ends of the two node modules (2) are in physical contact.

7. The dual-channel heat-dissipation computing server of claim 1, wherein: It also includes a panel (8), which is equipped with a second fan (9). The airflow channel flows from the second fan (9) and the mesh (3) to the node module (2), then flows from the first fan (4) to the heat dissipation space (6), and then flows out through the heat dissipation window (7) of the outer shell (1).

8. The dual-channel heat-dissipation computing server of claim 7, wherein: The left side of the inner cavity of the two node modules (2) is provided with a motherboard area (12), and the right side of the motherboard area (12) is provided with a GPU expansion area (13).

9. The dual-channel heat-dissipation computing server of claim 8, wherein: The mesh (3) air intake channel is located in the motherboard area (12), and the second fan (9) air intake channel is located in the GPU expansion area (13). A baffle (10) is provided at the separation point between the motherboard area (12) and the GPU expansion area (13). A through air passage (11) is provided on the baffle (10) near the end.

10. The dual-channel heat-dissipation computing server of claim 9, wherein: The motherboard area (12) and GPU expansion area (13) are provided with heat dissipation holes on the outer shell (1).

Images