Computing device and server cabinet

Abstract

The application provides a kind of computing device and server cabinet, it is related to server technical field, the computing device includes: shell, first internal cold plate, first mainboard and second mainboard, first internal cold plate, first mainboard and second mainboard are located in shell interior respectively, wherein, first mainboard is provided with first central processing unit on side surface, first central processing unit is attached with the first face of first internal cold plate;Second mainboard is provided with second central processing unit on side surface, second central processing unit is attached with the second face of first internal cold plate;First internal cold plate is used to cool first central processing unit and second central processing unit.The method of the application embodiment can improve the computing power of server cabinet.

Description

Technical Field

[0001] This application relates to the field of server technology, and more particularly to a computing device and a server rack. Background Technology

[0002] Server racks, also known as cabinets or racks, are the physical frameworks or structures used to store and organize computer servers and other network equipment.

[0003] In related technologies, server racks can house multiple computing devices, and air-cooling systems can assist in cooling the processors of these devices. However, with the current exponential growth in computing power demand, the computing power requirements of the devices to be installed in server racks are increasing. Furthermore, the fans in air-cooling systems occupy a significant amount of space within the server racks, resulting in a relatively low increase in computing power per rack. Summary of the Invention

[0004] This application provides computing devices and server racks to improve the computing power of the server racks.

[0005] In a first aspect, embodiments of this application provide a computing device, including a casing, a first internal cold plate, a first motherboard, and a second motherboard, wherein the first internal cold plate, the first motherboard, and the second motherboard are respectively located inside the casing, wherein...

[0006] A first central processing unit is disposed on the side of the first motherboard, and the first central processing unit is attached to the first side of the first internal cold plate.

[0007] A second central processing unit is provided on the side of the second motherboard, and the second central processing unit is attached to the second side of the first internal cold plate;

[0008] The first internal cooling plate is used to cool the first central processing unit and the second central processing unit.

[0009] In one possible implementation, the computing device further includes a heating element, a second internal cold plate, and a third internal cold plate, wherein the heating element is located between the second and third internal cold plates.

[0010] The second internal cold plate and the third internal cold plate are used to cool down the heating device;

[0011] The heating element, the second internal cold plate, and the third internal cold plate are located at one end of the first internal cold plate, and the difference between the center height of the heating element and the center height of the first internal cold plate is less than or equal to a preset threshold.

[0012] In one possible implementation,

[0013] The thickness of the first internal cold plate is greater than the thickness of the second internal cold plate;

[0014] The thickness of the first internal cold plate is greater than the thickness of the third internal cold plate;

[0015] The second and third internal cold plates have the same thickness.

[0016] In one possible implementation,

[0017] One end of the first internal cold plate is connected to one end of the second internal cold plate so that the coolant in the first internal cold plate and the second internal cold plate can communicate with each other.

[0018] One end of the first internal cold plate is connected to one end of the third internal cold plate so that the coolant in the first internal cold plate and the third internal cold plate can communicate with each other.

[0019] In one possible implementation, the computing device further includes multiple memory modules, and the first motherboard includes a first side and a second side, wherein...

[0020] M memory modules are laid flat on the first surface of the first motherboard, where M is an integer greater than 1.

[0021] N memory modules from the plurality of memory modules are laid flat on the second surface of the first motherboard, where N is an integer greater than or equal to 1, and the second surface is the surface on which the first central processing unit is mounted.

[0022] In one possible implementation,

[0023] The second internal cold plate is also used to cool at least one memory module in the first motherboard that is disposed facing the second internal cold plate;

[0024] The third internal cold plate is also used to cool at least one memory module in the second motherboard that is disposed facing the third internal cold plate.

[0025] In one possible implementation, the housing includes a first outer cold plate and a second outer cold plate, the first outer cold plate communicating with a first inner cold plate, and the second outer cold plate communicating with the first inner cold plate, wherein...

[0026] The first outer cold plate is used to cool down the multiple memory modules in the first motherboard that are positioned facing the first outer cold plate;

[0027] The second outer cold plate is used to cool down the multiple memory modules in the second motherboard that are positioned facing the second outer cold plate.

[0028] In one possible implementation, the memory module includes a memory motherboard and multiple memory chips, each memory chip corresponding to two data channels, which are used for data transmission within the memory module.

[0029] In one possible implementation, the computing device further includes a first microcontroller and a second microcontroller, wherein the first microcontroller is disposed on a second surface of the first motherboard, and the second microcontroller is disposed on a second surface of the second motherboard, wherein...

[0030] The first microcontroller contacts a first side of the first internal cold plate, and the second microcontroller contacts a second side of the first internal cold plate. The first side and the second side are symmetrical surfaces of the first internal cold plate.

[0031] Secondly, embodiments of this application provide a server rack, the server rack including a power management device, any one or more computing devices and switching nodes as described in the above embodiments, the switching nodes being connected to the multiple computing devices, wherein the power management device is used to provide power to the server rack; the switching nodes are used for interconnection between internal and external networks.

[0032] The computing device and server rack provided in this application embodiment include a chassis, a first internal cold plate, a first motherboard, and a second motherboard, all located inside the chassis. A first central processing unit (CPU) is disposed on the side of the first motherboard, and the CPU is attached to a first side of the first internal cold plate. A second CPU is disposed on the side of the second motherboard, and the CPU is attached to a second side of the first internal cold plate. The first internal cold plate is used to cool the first and second CPUs. The cold plate can replace the fan for heat dissipation, reducing the size of the chassis and allowing more computing devices to be installed in the server rack, thus increasing the computing density of the server rack. Attached Figure Description

[0033] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0034] Figure 1 A schematic diagram illustrating the application scenarios provided in the embodiments of this application;

[0035] Figure 2 This application provides a schematic diagram of the architecture of a computing device.

[0036] Figure 3A A schematic diagram of the layout of a first motherboard provided in an embodiment of this application;

[0037] Figure 3B A schematic diagram of the layout of a second motherboard provided in an embodiment of this application;

[0038] Figure 4 A schematic diagram of another first motherboard layout provided in an embodiment of this application;

[0039] Figure 5 This is a system topology diagram of a server rack provided in an embodiment of this application.

[0040] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0042] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and are not limited in number; for example, a first object can be one or more.

[0043] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0044] The terms "at least one," "at least one of," etc., used in the specification and claims of this application refer to any one, any two, or a combination of two or more of the included items. For example, at least one of a, b, and c can mean: "a," "b," "c," "a and b," "a and c," "b and c," and "a, b, and c," where a, b, and c can be single or multiple. Similarly, "at least two" refers to two or more items, and its meaning is similar to that of "at least one."

[0045] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0046] First, let me explain the terms used in this application:

[0047] Data Center: A data center is a specially designed facility for the centralized storage of computer systems and their related components, such as telecommunications and storage systems. It provides support such as power, cooling, security, and network connectivity to ensure the proper functioning of the equipment. A data center can contain multiple server racks, providing a centralized environment for managing and operating a large number of servers.

[0048] Server racks: Server racks, also known as server enclosures, are frames used to install and organize servers and other network equipment. They provide physical support, ventilation, and security. Server racks are typically placed in data centers and can house multiple server chassis, optimizing space utilization and equipment management.

[0049] Server chassis: A server chassis is the outer shell of a server, containing all internal components such as the motherboard, CPU, memory, hard drive, and power supply. It provides physical protection and structural support. The computing devices described in this application may include server chassis and servers.

[0050] Central Processing Unit (CPU): The CPU is the core processing unit of a computer, responsible for executing instructions and processing data. It is one of the most important components of a server, determining the server's computing power. The CPU is installed on the motherboard inside the server chassis and performs the server's computing tasks.

[0051] Cold plate: A sealed heat sink that can accommodate liquid flow. It is usually attached to the heat-generating device through a thermally conductive interface material, thereby removing the heat from the heat-generating device.

[0052] Multiple server racks can be placed in a data center, providing centralized computing resources and services by organizing and managing these racks. Each server rack can contain multiple computing devices, which are the core computing devices in the data center. These devices are responsible for processing data, running applications, and providing various services (such as web services and database services). These computing devices are also called computing nodes or computing units.

[0053] In related technologies, air-cooling systems can assist in heat dissipation for the processors of computing devices. These systems use fans to drive airflow and remove heat from the processors. However, current computing power demands are growing exponentially, and the computing power requirements of the computing devices that need to be installed in server racks are increasing. Air-cooling system fans occupy a significant amount of space for these computing devices, resulting in fewer computing devices that can be installed in the server racks and thus a lower overall computing power density.

[0054] The computing device provided in this application embodiment can use a cold plate instead of a fan for heat dissipation, reducing the size of the computing device and allowing more computing devices to be installed in the server rack, thereby increasing the computing power density of the server rack.

[0055] Figure 1 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. Please refer to [link / reference]. Figure 1 The server rack 100 may include multiple computing devices 101, multiple switching nodes 102, and at least one power management device 103. Each computing device 101 may occupy a height of 1U. The server rack 100 has a total height of 45U, a width of 540 mm, and a depth of 1200 mm. Here, "U" refers to the height specification of the computing device. 1U represents one rack unit, approximately equal to 1.75 inches (about 4.45 centimeters). In data centers or server racks, the height of equipment is typically measured in U units.

[0056] On the first side of the server rack 100, 42 computing devices 101 can be set up, and on the second side of the server rack 100, 38 computing devices 101 can be set up. Among them, 6 switching nodes 102 can be set up on the second side of the server rack 100, and two switching nodes 102 can be set up in a height of 1U.

[0057] A power management device 103 is installed at the bottom of the first side and the bottom of the second side of the server rack 100. The power management device 103 can be used to provide centralized power to the server rack 100, and can convert 380V alternating current (VC) power to 54V direct current (DC) power.

[0058] Server rack 100 also includes a cable tree, a power distribution system, and a manifold for coolant supply. The cable tree organizes and manages the cable structure within the rack, providing clear cable paths and simplifying cable installation, maintenance, and replacement. The power distribution system distributes power from the rack's main power input to individual devices (e.g., computing devices). The manifold distributes coolant to the various cold plates within the rack (e.g., the first internal cold plate). The cable tree, power distribution system, and manifold can be located in the middle of the first and second sides of server rack 100.

[0059] The computing device provided in this application embodiment may include a chassis, a first internal cold plate, a first motherboard, and a second motherboard, all located inside the chassis. A first central processing unit (CPU) is disposed on the side of the first motherboard, and the CPU is attached to the first surface of the first internal cold plate. A second CPU is disposed on the side of the second motherboard, and the CPU is attached to the second surface of the first internal cold plate. The first internal cold plate is used to cool the first and second CPUs. The cold plate can replace the fan for heat dissipation, reducing the size of the chassis and allowing more computing devices to be installed in the server rack, thus increasing the computing density of the server rack.

[0060] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0061] Figure 2 This is a schematic diagram of the architecture of a computing device provided in an embodiment of this application. Please refer to [link / reference]. Figure 2 The computing device may include a casing, a first internal cold plate, a first motherboard, and a second motherboard. The first internal cold plate, the first motherboard, and the second motherboard are located inside the casing.

[0062] A first central processing unit (CPU) is mounted on the side of the first motherboard. A first CPU socket can be installed on the first motherboard, through which the first CPU can be inserted into the first motherboard. The first CPU is fitted into the first side of the first internal cold plate.

[0063] A second central processing unit (CPU) is located on the side of the second motherboard. A second CPU socket can be installed on the second motherboard, through which the second CPU can be inserted. The second CPU is fitted into the second side of the first internal cold plate.

[0064] The first internal cold plate can be connected to a main liquid supply pipeline, which can distribute coolant to the first internal cold plate, enabling the flow of liquid inside the first internal cold plate. The first internal cold plate can be used to cool the first central processing unit and the second central processing unit. In this embodiment, the specific structure and working principle of the first internal cold plate are not described in detail.

[0065] Cooling can be achieved by replacing fans with cold plates, reducing the size of the chassis and allowing more computing devices to be installed in the server rack, thus increasing the computing power density of the server rack.

[0066] In some embodiments, the computing device further includes a heat-generating device, a second internal cold plate, and a third internal cold plate, with the heat-generating device positioned between the second and third internal cold plates, which can be used to cool the heat-generating device.

[0067] Please see Figure 2 The heating element, the second internal cold plate, and the third internal cold plate can be located at one end of the first internal cold plate, and the difference between the center height of the heating element and the center height of the first internal cold plate is less than or equal to a preset threshold.

[0068] Heat-generating components can be open computing projects (OCP) or hard drives, and other heat-generating devices besides the CPU and memory modules. For example, a hard drive can be an M.2 hard drive, where "M.2" is an interface standard, originally called Next Generation Form Factor (NGFF). The M.2 interface can be used to connect solid state drives (SSDs) and other devices.

[0069] While cooling the first and second central processing units through the first internal cold plate, the second and third internal cold plates can also be used to cool heat-generating components, thereby improving the heat dissipation efficiency of the computing device.

[0070] Because the CPU has high power, it generates a lot of heat. Therefore, the thickness of the first internal cold plate can be greater than that of the second and third internal cold plates.

[0071] To facilitate production and standardize specifications, the thickness of the second and third internal cold plates can be the same.

[0072] For example, the thickness of the first internal cold plate can be 10mm, the thickness of the second internal cold plate can be 4mm, and the thickness of the third internal cold plate can be 4mm.

[0073] The thickness of the second and third internal cold plates is less than that of the first internal cold plate, which reduces the height occupied by the second and third internal cold plates on the computing device. This allows the computing device to install more memory modules in a limited space, thereby improving the storage capacity of the computing device.

[0074] In some possible embodiments, one end of the first internal cold plate is connected to one end of the second internal cold plate so that the coolant in the first internal cold plate and the second internal cold plate can communicate with each other; one end of the first internal cold plate is connected to one end of the third internal cold plate so that the coolant in the first internal cold plate and the third internal cold plate can communicate with each other.

[0075] Please see Figure 2 The right ends of the second and third internal cold plates are aligned, and the heating element is located between the second and third internal cold plates. The thickness of the first internal cold plate is greater than that of the second internal cold plate, and the thickness of the first internal cold plate is greater than that of the third internal cold plate. The right end of the second internal cold plate is connected to the left end of the first internal cold plate, and the right end of the third internal cold plate is connected to the left end of the first internal cold plate.

[0076] When the main supply line provides coolant to the first internal cold plate connection, the coolant can flow through the connection part inside the first internal cold plate, the second internal cold plate and the third internal cold plate, which can simplify the supply path and reduce the space occupied by the supply path.

[0077] Please see Figure 2 The computing device also includes multiple memory modules. The first motherboard includes a first side and a second side. M memory modules are laid flat on the first side of the first motherboard, where M is an integer greater than 1. N memory modules are laid flat on the second side of the first motherboard, where N is an integer greater than or equal to 1. The second side of the first motherboard is the side where the first central processing unit is located.

[0078] The height occupied by the memory module is less than that of the first central processing unit, which allows space to be placed between the second internal cold plate and the first motherboard. This space can be used to lay the memory module flat on the first motherboard, increasing the number of memory modules that can be placed on the first motherboard.

[0079] The memory module can be connected to the first motherboard via the Compression Attached Memory Module 2 (CAMM2) connector.

[0080] Compared to vertically inserted dual in-line memory module (DIMM) connectors, the CAMM2 connector reduces the height occupied in computing devices, thereby reducing the overall height of the computing device.

[0081] The second internal cold plate is also used to cool at least one memory module in the first motherboard that is positioned facing the second internal cold plate. The third internal cold plate is also used to cool at least one memory module in the second motherboard that is positioned facing the third internal cold plate.

[0082] Please see Figure 2 Because the thickness of the first internal cold plate is greater than that of the second and third internal cold plates, and because the first CPU, inserted into the first motherboard via the first CPU socket, occupies a certain height, and the second CPU, inserted into the second motherboard via the second CPU socket, occupies a certain height, the combined height of the second internal cold plate, heat-generating components, and the third internal cold plate at one end of the first internal cold plate is less than the combined height of the first CPU socket, the first CPU, the first internal cold plate, the second CPU, and the second CPU socket. Therefore, when N memory modules are placed on the second side of the first motherboard, they can be placed between the first motherboard and the second internal cold plate. When N memory modules are placed on the first side of the second motherboard, they can be placed between the second motherboard and the third internal cold plate. This allows the second internal cold plate to cool at least one memory module on the first motherboard that is positioned on the second internal cold plate, and the third internal cold plate to cool at least one memory module on the second motherboard that is positioned towards the third internal cold plate.

[0083] In some possible embodiments, the casing of the computing device may include a first outer cold plate and a second outer cold plate, the first outer cold plate being connected to a first inner cold plate and the second outer cold plate being connected to the first inner cold plate, wherein the first outer cold plate is used to cool down a plurality of memory modules in the first motherboard that are disposed facing the first outer cold plate; and the second outer cold plate is used to cool down a plurality of memory modules in the second motherboard that are disposed facing the second outer cold plate.

[0084] Please see Figure 2 M memory modules are laid flat on the first side of the first motherboard, which faces the first outer cold plate. The M memory modules of the first motherboard can be cooled by the first outer cold plate. Similarly, M memory modules are laid flat on the first side of the second motherboard, which faces the second outer cold plate. The M memory modules of the second motherboard can be cooled by the second outer cold plate.

[0085] Since the power consumption of the memory module is lower than that of the central processing unit, the heat generated by the memory module is lower than that generated by the central processing unit. Therefore, the thickness of the first outer cold plate and the second outer cold plate can be less than the thickness of the first inner cold plate.

[0086] In addition, one end of the first outer cold plate is connected to one end of the first inner cold plate so that the coolant in the first outer cold plate and the first inner cold plate can communicate with each other; one end of the second outer cold plate is connected to one end of the first inner cold plate so that the coolant in the second outer cold plate and the first inner cold plate can communicate with each other.

[0087] In some possible embodiments, the first outer cold plate and the second outer cold plate can be connected to the main liquid supply pipeline respectively, and the coolant can be directly supplied to the first outer cold plate and the second outer cold plate through the main liquid supply pipeline.

[0088] The chassis can be replaced by the first and second outer cold plates, realizing a "chassis-free" design for computing devices, which can improve the efficiency of cooling the computing devices.

[0089] In some embodiments, the memory module may include a memory motherboard and multiple memory chips, with the multiple memory chips corresponding to two data channels.

[0090] Memory chips are the basic storage chips used to manufacture memory modules. Memory chips are semiconductor chips responsible for storing and retrieving data.

[0091] Memory chips can be either data bit memory chips or error correction code (ECC) memory chips. Data bit memory chips can be used to store the actual data bits, while ECC memory chips can be used to store error correction codes.

[0092] A data channel can be used for data transfer within a memory module. One data channel can correspond to half of the memory chips in a given memory module. For example, assuming a memory module has 40 memory chips and 2 data channels, data channel 1 corresponds to 20 memory chips, including 16 data-bit chips and 4 ECC chips. If each data-bit chip has a memory size of 8GB, then each data channel has a memory size of 128GB. Data channel 2 corresponds to the other 20 memory chips.

[0093] In some embodiments, multiple memory chips can be arranged on the front and back of the memory module. For example, with 40 memory chips, 20 memory chips can be arranged on each side of the memory module, and a memory module can have 40 memory chips.

[0094] Setting up two data channels in the memory module can improve the data transfer efficiency of the memory module and at the same time improve the data processing efficiency of the computing device.

[0095] The computing device also includes a first microcontroller and a second microcontroller. The first microcontroller is disposed on a second side of the first motherboard, and the second microcontroller is disposed on a second side of the second motherboard. The first microcontroller is in contact with one side of the first internal cold plate, and the second microcontroller is in contact with the other side of the first internal cold plate.

[0096] To facilitate understanding, the following will be combined with... Figure 3A and Figure 3B The layout diagrams of the first motherboard and the second motherboard provided in the embodiments of this application will be described.

[0097] Figure 3A This is a schematic diagram of the layout of a first motherboard provided in an embodiment of this application. Please refer to... Figure 3A , Figure 3A These are schematic diagrams of the first and second sides of the first motherboard. The first side is the side of the first motherboard where the first central processing unit (CPU) is not located, and the second side is the side of the first motherboard where the CPU is located.

[0098] A Complex Programmable Logic Device (CPLD) and two memory modules are set on the first side of the first motherboard, and the first central processing unit is inserted into the second side of the first motherboard between the two memory modules.

[0099] The first CPU and four memory modules are located on the second side of the first motherboard. Six CPU voltage regulators (VRs) are also located on the second side of the first motherboard; the CPU VRs are responsible for providing stable and appropriate voltage to the first CPU. A PCIe bus is also located on the second side of the first motherboard, which can connect heat-generating components via high-density connectors. The first Baseboard Management Controller (BMC) is also located on the second side of the first motherboard; the BMC is responsible for managing and monitoring the status of the system hardware.

[0100] Figure 3B This is a schematic diagram of the layout of a second motherboard provided in an embodiment of this application. Please refer to... Figure 3B , Figure 3B These are schematic diagrams of the first and second sides of the second motherboard. The layout of the first side of the second motherboard can be found in [reference needed]. Figure 3A The first side diagram of the first motherboard is shown below and will not be repeated here. The layout of the second side diagram of the second motherboard can be found in [reference needed]. Figure 3AThe second side diagram of the first motherboard shows that the second BMC of the second motherboard is offset from the first BMC of the first motherboard, so that the first BMC contacts the first side of the first internal cold plate, and the second BMC contacts the second side of the second internal cold plate.

[0101] The staggered placement of the first and second BMCs ensures no structural interference between them, thus reducing the height of the computing device.

[0102] The improved computing device of this application can have a first motherboard with the first central processing unit (CPU) on one side and a second motherboard with the second CPU on the other side, with a first internal cold plate, a second internal cold plate, a third internal cold plate, and heat-generating devices sandwiched in between, and the first BMC and the second BMC are staggered, resembling a "sandwich" structure. This can reduce the height of the computing device, allowing more computing devices to be installed in the server rack, thereby increasing the computing power density of the server rack.

[0103] The computing device provided in this application embodiment can use a cooling plate instead of a fan for heat dissipation, reducing the size of the chassis and allowing more computing devices to be installed in the server rack, thereby increasing the computing power density of the server rack, effectively reducing the floor space occupied by the server rack in the data center, and improving the utilization rate of the data center rack space.

[0104] Figure 4 This is a schematic diagram of another first motherboard layout provided in an embodiment of this application. Please refer to... Figure 4 , Figure 4 These are schematic diagrams of the first and second sides of the first motherboard. The layout of the first side schematic diagram of the first motherboard is shown below. Figure 3A Based on the first schematic diagram, two more memory modules are added, giving the first motherboard a total of eight memory modules. Each memory module has two channels, enabling the CPU to support up to 16 channels. The layout of the second schematic diagram of the first motherboard can be found in [reference needed]. Figure 3A The second side diagram of the first motherboard is shown here and will not be described again.

[0105] In one possible embodiment, the server rack may include a power management device, any number of computing devices as described in the above embodiments, and a switching node, wherein the switching node is connected to the multiple computing devices, and the power management device is used to provide power to the server rack; the switching node is used for interconnection between the internal network and the external network.

[0106] The internal space of general-purpose computing devices can be compressed by improving memory modules and large cold plate heat dissipation, so as to achieve ultra-high density device node configuration. The simplicity and efficiency of computing device design can be ensured by centralized power supply through power management devices and improved chassis structure of computing devices. At the same time, the unique "sandwich" structure, multi-layer cold plate and chassis-less design are used to compress the internal space of computing devices to the extreme, so that the server rack can achieve high density and liquid cooling heat dissipation.

[0107] Figure 5 This is a system topology diagram of a server rack provided for an embodiment of this application. Please refer to [link / reference]. Figure 5 The server rack includes multiple computing devices, each of which may include a first motherboard and a second motherboard. The first motherboard is equipped with a first central processing unit and a first heat-generating device. The first central processing unit and the first heat-generating device can be connected to a second central processing unit and a second heat-generating device via the second motherboard.

[0108] The first and second central processing units (CPUs) can communicate via high-speed interconnect technology. The first CPU can connect to the switching node via a first network interface controller (NIC) and a first interface. The second CPU can connect to the switching node via a second NIC and a second interface.

[0109] The first central processing unit (CPU) can be connected to the first NIC via serial-to-parallel converters (serders), and the first NIC is connected to the first interface via the serders. The second CPU can be connected to the second NIC via serders, and the second NIC is connected to the second interface via the serders.

[0110] When the first and second interfaces can be connected to the interaction node via cables, multiple cables can be protected by a cable tray since multiple computing devices require multiple cables to be connected to the interaction node.

[0111] Server racks can exchange information with upper-layer switches through the high-density optical module interfaces (Quad Small Form-factor Pluggable Double Density, QSFP-DD) of the switching nodes.

[0112] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0113] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A computing device, characterized in that, The system includes a casing, a first internal cooling plate, a first motherboard, and a second motherboard, wherein the first internal cooling plate, the first motherboard, and the second motherboard are respectively located inside the casing. A first central processing unit is disposed on the side of the first motherboard, and the first central processing unit is attached to the first side of the first internal cold plate. A second central processing unit is provided on the side of the second motherboard, and the second central processing unit is attached to the second side of the first internal cold plate; The first internal cold plate is used to cool the first central processing unit and the second central processing unit; The computing device further includes multiple memory modules, and the first motherboard includes a first side and a second side. The memory modules are connected to the first motherboard via a CAMM2 connector. M memory modules are laid flat on the first surface of the first motherboard, where M is an integer greater than 1; N memory modules are laid flat on the second surface of the first motherboard, where N is an integer greater than or equal to 1, and the second surface is the surface on which the first central processing unit is disposed. The housing includes a first outer cold plate and a second outer cold plate, wherein the first outer cold plate is in communication with the first inner cold plate, and the second outer cold plate is in communication with the first inner cold plate. The first outer cold plate is used to cool down the multiple memory modules in the first motherboard that are positioned facing the first outer cold plate; the second outer cold plate is used to cool down the multiple memory modules in the second motherboard that are positioned facing the second outer cold plate. The computing device further includes a heating element, a second internal cold plate, and a third internal cold plate, wherein the heating element is located between the second and third internal cold plates. The second internal cold plate and the third internal cold plate are used to cool down the heating device; The heating element, the second internal cold plate, and the third internal cold plate are located at one end of the first internal cold plate, and the difference between the center height of the heating element and the center height of the first internal cold plate is less than or equal to a preset threshold. The second internal cold plate is also used to cool at least one memory module in the first motherboard that is disposed facing the second internal cold plate; The third internal cold plate is also used to cool at least one memory module in the second motherboard that is disposed facing the third internal cold plate.

2. The computing device according to claim 1, characterized in that, The thickness of the first internal cold plate is greater than the thickness of the second internal cold plate; The thickness of the first internal cold plate is greater than the thickness of the third internal cold plate; The second and third internal cold plates have the same thickness.

3. The computing device according to claim 1, characterized in that, One end of the first internal cold plate is connected to one end of the second internal cold plate so that the coolant in the first internal cold plate and the second internal cold plate can communicate with each other. One end of the first internal cold plate is connected to one end of the third internal cold plate so that the coolant in the first internal cold plate and the third internal cold plate can communicate with each other.

4. The computing device according to claims 2-3, characterized in that, The memory module includes a memory motherboard and multiple memory chips. The multiple memory chips correspond to two data channels, which are used for data transmission within the memory module.

5. The computing device according to any one of claims 1-3, characterized in that, The computing device further includes a first microcontroller and a second microcontroller, wherein the first microcontroller is disposed on a second surface of the first motherboard, and the second microcontroller is disposed on a second surface of the second motherboard. The first microcontroller contacts a first side of the first internal cold plate, and the second microcontroller contacts a second side of the first internal cold plate. The first side and the second side are symmetrical surfaces of the first internal cold plate.

6. A server rack, characterized in that, The server rack includes a power management device, a computing device as described in any one of claims 1-5, and a switching node, wherein the switching node is connected to the plurality of computing devices, and the power management device is used to provide power to the server rack; the switching node is used for interconnection between internal and external networks.

Images