Server

Abstract

The invention discloses a server, and relates to the technical field of servers, the server comprises at least one mainboard and at least one backboard, each mainboard comprises a central processing unit, the central processing units are mutually independently arranged, each backboard is provided with at least one hard disk, each hard disk is based on a disk sequence of the hard disks, and each hard disk is provided with at least one hard disk. According to the technical scheme, the central processing unit is connected with the nearest central processing unit, so that the central processing units can independently operate and execute the functions of data storage and data operation, the problem of load imbalance existing in the multiple central processing units in the server can be avoided, and the operation efficiency in a multi-processor mode is improved.

Description

Technical Field

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

[0002] As the core unit for executing data operations and control in a server, the Central Processing Unit (CPU) has high-speed computing capabilities, long-term operation capabilities, and powerful data throughput capabilities. Related servers mainly meet the growing data processing requirements by setting two or more central processing units on the motherboard.

[0003] However, the above multi-processor mode is prone to cause problems such as load imbalance in the server, resulting in data loss, reduced computing efficiency, wasted data processing resources, slow heat dissipation efficiency, and even system downtime in the server, affecting the working efficiency and stability of the server. Summary of the Invention

[0004] This application provides a server to at least solve the problem of load imbalance of the central processing units on the motherboard of the server in the related art.

[0005] This application provides a server, including at least one motherboard and at least one backplane;

[0006] Each motherboard includes a central processing unit, and each central processing unit is independently set;

[0007] At least one hard disk is provided on each backplane;

[0008] For each backplane, based on the disk order of the hard disks on the backplane, each hard disk is sequentially connected to the target central processing unit; wherein, the target central processing unit is used to indicate the central processing unit with the smallest distance to the hard disk.

[0009] Through this application, since each motherboard on the server includes a central processing unit, each central processing unit is independently set, and at least one hard disk is provided on each backplane, and each hard disk is connected to the central processing unit with the closest distance based on the disk order of the hard disks, enabling each central processing unit to operate independently and perform the functions of data storage and data operation. Therefore, the problem of load imbalance existing in multiple central processing units in the server can be avoided, and the computing efficiency in the multi-processor mode is improved. Description of the Drawings

[0010] To more clearly illustrate the embodiments of this application, the following will briefly introduce the drawings required for the embodiments. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 It is a schematic structural diagram of a server;

[0012] Figure 2 It is one of the schematic structural diagrams of the server provided by the embodiment of the present application;

[0013] Figure 3 It is a schematic application diagram of a central processing unit in a master-slave mode;

[0014] Figure 4 It is a schematic application diagram of the central processing unit provided by the embodiment of the present application;

[0015] Figure 5 It is a schematic connection diagram of a related server;

[0016] Figure 6 It is one of the schematic connection diagrams of the server provided by the embodiment of the present application;

[0017] Figure 7 It is the second schematic connection diagram of the server provided by the embodiment of the present application;

[0018] Figure 8 It is the third schematic connection diagram of the server provided by the embodiment of the present application;

[0019] Figure 9 It is the second schematic structural diagram of the server provided by the embodiment of the present application. Detailed implementation manners

[0020] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present application.

[0021] It should be noted that in the description of the present application, the terms "include", "comprise" or any other variant thereof are intended to cover a non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. The terms "first", "second", etc. in the present application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0022] It should be noted that the orientation or positional relationship indicated by terms such as "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present application. The terms "mounted", "connected", and "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. The terms "parallel", "perpendicular", and "equal" include the described situations and situations similar to the described situations, and the range of the similar situations is within an acceptable deviation range, where the acceptable deviation range is determined by those of ordinary skill in the art considering the measurements being discussed and the errors associated with the measurements of specific quantities (i.e., the limitations of the measurement system). For example, "parallel" includes absolute parallel and approximate parallel, where the acceptable deviation range of approximate parallel can be, for example, within 5° deviation; "perpendicular" includes absolute perpendicular and approximate perpendicular, where the acceptable deviation range of approximate perpendicular can also be, for example, within 5° deviation. "Equal" includes absolute equality and approximate equality, where the acceptable deviation range of approximate equality can be, for example, that the difference between the two equal ones is less than or equal to 5% of either one. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0023] The master-slave mode means that in the case of multiple central processing units (CPUs), one of the central processing units is the master (also called the master CPU), and the central processing units interact with each other to dynamically allocate tasks according to the load condition of the master CPU.

[0024] Single-way: The mainboard includes one central processing unit.

[0025] Dual-way: The mainboard includes two central processing units, and the two central processing units form a system ecosystem.

[0026] Dual-single-way mode means that in a server, two single-way mainboards are placed at the same time, and the central processing units on each mainboard operate independently.

[0027] A network card, also known as a Network Interface Card, is a hardware component of a server, which is responsible for physically connecting to the network and transmitting data.

[0028] The Central Processing Unit (CPU), as the core unit for performing data operations and control in a server, has high-speed computing capabilities, long-term operating capabilities, and strong data throughput capabilities. Related servers meet the growing data processing requirements by setting two or more central processing units on the motherboard.

[0029] However, the above multi-processor mode easily leads to problems such as unbalanced CPU load, data loss, server downtime, reduced computing efficiency, waste of data processing resources, and slow heat dissipation efficiency in the server, thus affecting the working efficiency and stability of the server.

[0030] Figure 1 It is a schematic diagram of the structure of the related server.

[0031] See Figure 1 , there are 2 central servers (including central server 1-0 and central server 1-1) installed on the motherboard in the server, and the central server 1-0 and the central server 1-1 are communicatively connected. In this way, the central server 1-0 and the central server 1-1 can work in a master-slave mode. Taking the central server 1-0 as the master server as an example, during the operation of the server, the load of the central server 1-0 is much higher than that of the central server 1-1. When the server needs to execute more data operation tasks, the load of the central server 1-0 will still increase, easily leading to data loss and even server downtime. Moreover, in the case of the above unbalanced load, the computing resources of the central server 1-1 are left unused and wasted, and the solution path between the central server 1-1 and the central server 1-0 does not change, resulting in a reduction in the computing efficiency of the central server 1-0.

[0032] At the same time, in the above master-slave mode, the cables between the central server 1-0, the central server 1-1 and each component on the server motherboard cross each other, affecting the heat dissipation of the central server 1-0 and the central server 1-1. When the load of the central server 1-0 is large, it is easy to cause problems such as server downtime or failure, affecting the normal operation of the server.

[0033] To solve the above problems, the present application provides a server, which includes at least one main board and at least one backplane. Each main board includes a central processing unit, and the central processing units are independently arranged. At least one hard disk is provided on each backplane. For each backplane, based on the disk order of the hard disks on the backplane, each hard disk is sequentially connected to the target central processing unit, and the target central processing unit is used to indicate the central processing unit with the smallest distance from the hard disk. In this way, each central processing unit can operate independently and perform the functions of data storage and data operation, avoiding the problem of load imbalance of multiple central processing units in the master-slave mode and improving the operation efficiency in the multi-processor mode.

[0034] To enable those skilled in the art of this technical field to better understand the solution of the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] The server includes at least one main board and at least one backplane;

[0036] Each main board includes a central processing unit, and the central processing units are independently arranged;

[0037] At least one hard disk is provided on each backplane;

[0038] For each backplane, based on the disk order of the hard disks on the backplane, each hard disk is sequentially connected to the target central processing unit; wherein, the target central processing unit is used to indicate the central processing unit with the smallest distance from the hard disk.

[0039] Optionally, the main board of the server is a structure with a first end face and a second end face.

[0040] Optionally, each main board on the server can be adjacent to the backplane or not adjacent to the backplane, and each main board and each backplane are horizontally placed in the server.

[0041] Optionally, each main board on the server includes at least one central processing unit, and the central processing units are independently arranged and do not communicate with each other, and each central processing unit operates independently.

[0042] Optionally, at least one hard disk is provided on each backplane of the server, and the hard disks are horizontally arranged on the backplane.

[0043] Optionally, the disk order of the hard disks refers to the arrangement order of the hard disks in the system, which is usually used to identify the positions and identification methods of multiple hard disks in the system.

[0044] Optionally, the disk order of the hard disks in the server starts from 0 and is sequentially sorted based on the order from right to left.

[0045] Optionally, for each backplane, based on the disk order of the hard disks on the backplane, each hard disk is sequentially connected to the target central processing unit with the smallest distance from the hard disk.

[0046] Exemplarily, in the case where the server includes 4 mainboards and 1 backplane, and 4 hard disks are provided on the backplane, the disk orders of the hard disks from right to left are 0, 1, 2, and 3 in sequence. Each mainboard includes a central processing unit, and the mainboards are arranged from right to left on the server in sequence. Based on this, the central processing units (such as central processing unit 0, central processing unit 1, central processing unit 2, and central processing unit 3) on the server are arranged from right to left in sequence. The hard disk with disk order 0 is connected to the nearest central processing unit 0, the hard disk with disk order 1 can be connected to the nearest central processing unit 1, the hard disk with disk order 2 can be connected to the nearest central processing unit 2, and the hard disk with disk order 3 can be connected to the nearest central processing unit 3.

[0047] Figure 2 It is one of the structural schematic diagrams of the server provided by the embodiment of the present application.

[0048] See Figure 2 , the server includes 2 mainboards (mainboard 1 and mainboard 2 respectively), each mainboard includes a central processing unit (central processing unit 2-0 on mainboard 1 and central processing unit 2-1 on mainboard 2 respectively), and central processing unit 2-0 and central processing unit 2-1 are independently set and do not communicate with each other.

[0049] The server also includes 1 backplane, and 4 hard disks are provided on the backplane, which are hard disk 0, hard disk 1, hard disk 2, and hard disk 3 from right to left in sequence.

[0050] Among them, hard disk 0 and hard disk 1 provided on the backplane are connected to central processing unit 2-0, and hard disk 2 and hard disk 3 provided on backplane 1 are connected to central processing unit 2-1.

[0051] In this way, each hard disk provided on the backplane is connected to the central processing unit with the smallest distance from it, which can enable each central processing unit to operate independently, achieve load balancing, and at the same time, make the cables on the server not intersect or intersect less, thereby improving the heat dissipation efficiency of each central processing unit.

[0052] Figure 3 It is the application schematic diagram of the central processing unit in the master-slave mode.

[0053] See Figure 3, the motherboard on the server includes two central processing units (Central Processing Unit 3-0 and Central Processing Unit 3-1 respectively), and they work in a master-slave mode. The data blocks to be processed include Data Block A and Data Block B, and each data block includes 8 sub-data blocks (Data Block A-1, Data Block A-2, Data Block A-3, Data Block A-4, Data Block A-5, Data Block A-6, Data Block A-7, Data Block A-8 and Data Block B-1, Data Block B-2, Data Block B-3, Data Block B-4, Data Block B-5, Data Block B-6, Data Block B-7, Data Block B-8).

[0054] Four sub-data blocks in Data Block A (including Data Block A-1, Data Block A-3, Data Block A-5, Data Block A-7) and four sub-data blocks in Data Block B (including Data Block B-1, Data Block B-3, Data Block B-5, Data Block B7) are respectively assigned to Central Processing Unit 3-0 for processing. The remaining four sub-data blocks in Data Block A (including Data Block A-2, Data Block A-4, Data Block A6, Data Block A-8) and the remaining four sub-data blocks in Data Block B (including Data Block B-2, Data Block B-4, Data Block B-6, Data Block B-8) are assigned to Central Processing Unit 3-1 for processing. In the master-slave mode, taking Central Processing Unit 3-0 as the master processor, Central Processing Unit 3-1 needs to wait for Central Processing Unit 3-0 to complete the processing before performing data processing operations. Due to the large load of Central Processing Unit 3-0 and the low data processing efficiency, the time-consuming for Central Processing Unit 3-1 to execute waiting and data processing is relatively long, which affects the operating efficiency of the server.

[0055] Figure 4 It is a schematic diagram of the application of the central processing unit provided by the embodiment of the present application.

[0056] See Figure 4 , when the server includes two motherboards and each motherboard includes a central processing unit (Central Processing Unit 4-0 and Central Processing Unit 4-1 respectively), and the data blocks to be processed include Data Block A and Data Block B, and each data block includes 8 sub-data blocks, each sub-data block in Data Block A is assigned to Central Processing Unit 4-0 for processing, and each sub-data block in Data Block B is assigned to Central Processing Unit 4-1 for processing. Central Processing Unit 4-0 and Central Processing Unit 4-1 independently and concurrently run data processing tasks without affecting each other and without waiting. The computing efficiency of the server is equal to that of two independent central processing unit links working simultaneously, and different data can be processed in parallel at the same time. The data reading and writing rate is faster, and it has the ability to run multiple tasks and multiple threads, making the internal resource allocation more average, thereby improving the overall data processing efficiency of the server.

[0057] Optionally, the number of hard disks connected to each central processing unit is inversely proportional to the number of central processing units.

[0058] Optionally, due to limited space on the server, the number of backplanes included in the server is also limited, and the number of hard disks set on the server is restricted. Thus, the more the number of central processing units (CPUs) on the server, the fewer the number of hard disks that each CPU can connect to, that is, the number of hard disks connected to each CPU is inversely proportional to the number of CPUs.

[0059] Optionally, in the case where the server includes two or more motherboards, the number of hard disks connected to each CPU can be equal.

[0060] Exemplarily, in the case where the server includes 2 motherboards, each motherboard includes one CPU, and the server includes 2 backplanes, and 4 hard disks are set on each backplane, each CPU can connect to 2 hard disks, or 4 hard disks.

[0061] Another exemplarily, in the case where the server includes 4 motherboards, each motherboard includes one CPU, and the server includes 2 backplanes, and 4 hard disks are set on each backplane, each CPU can connect to 2 hard disks.

[0062] Optionally, each CPU on the server is connected to at least one hard disk. In the case where the server includes two or more motherboards and each motherboard includes one CPU, the hard disks on each backplane can be evenly distributed among the CPUs. In the case where the server includes only one motherboard and that motherboard includes one CPU, there is one or more hard disks not connected to the CPU via a cable.

[0063] In some application scenarios, in the case where the server includes only one motherboard and the motherboard includes one CPU, the above connection method can meet the business requirements. However, in order to cope with changing market demands and business requirements, the number of independently operating CPUs can be increased or decreased by adding or deleting motherboards on the server, so as to flexibly switch or expand the configuration between single-channel, dual-single-channel, and even multi-single-channel working modes, making the server applicable to more business scenarios, while reducing the production and maintenance costs of the server motherboard.

[0064] Another exemplarily, in the case where the server includes 4 motherboards, each motherboard includes one CPU, and the server includes 1 backplane, and 4 hard disks are set on the backplane, each CPU can connect to 1 hard disk.

[0065] Figure 5 It is a connection schematic diagram of a related server.

[0066] See Figure 5, the motherboard of the server includes two central processing units (Central Processing Unit 5-0 and Central Processing Unit 5-1 respectively), Central Processing Unit 5-0 and Central Processing Unit 5-1 are communicatively connected, and the two central processing units work in a master-slave mode. Among them, Central Processing Unit 5-0 is the master processor, and the positions of Central Processing Unit 5-0 and Central Processing Unit 5-1 are the first end face of the motherboard. The server also includes 3 backplanes, each backplane is provided with 4 hard disks, and the hard disks on the server are sorted in the order from right to left and from top to bottom.

[0067] In the master-slave mode, the hard disks on the server start from disk sequence 0, and the cables are connected in sequence with disk sequences from 0 to 11, so as to be connected to Central Processing Unit 5-0 and Central Processing Unit 5-1. The cable led out from the interface of Central Processing Unit 5-0 needs to be wired from the left side of the server around the second end face to the position of the hard disk on the right side, and the cable led out from the interface of Central Processing Unit 5-1 needs to be wired from the right side of the server around the second end face to the position of the hard disk on the left side, so as to realize the connection between each hard disk and Central Processing Unit 5-0 and Central Processing Unit 5-1.

[0068] It can be understood that, due to working in the master-slave mode, the central processing unit resources on the server are unevenly distributed, and the cables are wired in the above manner, resulting in cable crossovers, which are likely to cause problems such as unbalanced central processing unit load and slow central processing unit heat dissipation.

[0069] Moreover, the cables used for connection on the backplane include but are not limited to one or more of Serial General Purpose Interface (SGPIO) cables, Inter-Integrated Circuit (I2C), power cables, data cables, etc. Among them, the power cables and data cables are both cable groups with larger wire diameters and greater hardness. The crossovers of the above-mentioned various cables are not conducive to operations such as server production and maintenance.

[0070] Figure 6 This is one of the connection schematic diagrams of the server provided by the embodiment of the present application.

[0071] See Figure 6, the server includes two mainboards, and each mainboard includes a central processing unit (the central processing unit 6-0 on mainboard 1 and the central processing unit 6-1 on mainboard 2 respectively). At this time, the server supports the dual single-way working mode, that is, each central processing unit is set independently, does not communicate with each other, and works independently. This avoids the problem of uneven load on the central processing unit, makes the resource allocation of the central processing unit 6-0 and the central processing unit 6-1 on the server average, and improves the data processing efficiency of each central processing unit. Mainboard 1 and mainboard 2 are both horizontally placed on the first end face of the server. There are also 3 backplanes on the first end face of the server, and each backplane is provided with 4 hard disks. The hard disks on the server are sorted in sequence from right to left and from top to bottom. Among them, the central processing unit 6-0 is closer to the hard disks with smaller disk numbers on the backplane, and the central processing unit 6-1 is closer to the hard disks with larger disk numbers on the backplane. Based on this, the hard disk 0 on backplane 1, the hard disk 4 on backplane 2, and the hard disk 8 on backplane 3 are all connected to the central processing unit 6-0 through cables, and the hard disk 3 on backplane 1, the hard disk 7 on backplane 2, and the hard disk 11 on backplane 3 are all connected to the central processing unit 6-1 through cables. In this way, the hard disks set on each backplane can be connected to the central processing unit closest to the hard disk, reducing the situation of cable crossing. While reducing the length of the cables and improving the heat dissipation efficiency of each central processing unit, it also realizes the anti-fooling design, preventing the user from accidentally connecting wrongly due to confusion, carelessness or poor hands-on ability during operation, thus avoiding threatening the performance and safety of the connector. The above wiring method is simplified and streamlined, reducing the probability of cable connection errors in the server, thereby reducing the production and maintenance costs of the server.

[0072] According to Figure 5 and Figure 6 it can be seen that Figure 6 In the connection method of the server shown in

[0073] the length of the cables is shorter, thus reducing the production cost of the mainboard. And when the central processing unit is connected to the cables, the cables do not need to be routed around the second end face to the other side to be connected to the hard disk, which is convenient for the maintenance of the mainboard and improves the heat dissipation efficiency of the central processing unit. It realizes that while improving the balance degree of the central processing unit resource allocation, it reduces the production cost and repair efficiency of the server.

[0073] It can be understood that Figure 6 only takes the connection of each central processing unit to one hard disk set on each backplane as an example for illustration. In actual applications, such as Figure 6The central processing units 6-0 and 6-1 can also be connected to two hard disks provided on each backplane. For example, the central processing unit 6-0 is connected to the hard disks 0 and 1 provided on the backplane 1, the hard disks 4 and 5 provided on the backplane 2, and the hard disks 8 and 9 provided on the backplane 3. The central processing unit 6-1 is connected to the hard disks 2 and 3 provided on the backplane 1, the hard disks 6 and 7 provided on the backplane 2, and the hard disks 10 and 11 provided on the backplane 3.

[0074] It can be understood that when the server includes at least one mainboard, each mainboard includes a central processing unit, and the central processing units are independently set and do not communicate with each other and operate independently, the server can implement a working mode with flexible switching between the single-way mode and the dual single-way mode. That is to say, when there is a need to expand the central processing unit in the server, the central processing unit can be added or deleted by adding or deleting the mainboard on the server, which can flexibly support the balanced resource allocation of each central processing unit, and can improve the overall performance, computing density, load response of the server, and server resource scheduling, etc.

[0075] Moreover, the layout and connection method of the above server do not require point-to-point (Ultra Path Interconnect, UPI) interconnection and do not require cross-non-uniform memory access architecture (Non Uniform Memory Access, NUMA) access. Based on the design of flexible switching between single-way / dual single-way, the explosion radius is short when the central processing unit fails, the impact range is small when it fails, and the data migration time is short, thus improving the stability of the server.

[0076] Figure 7 This is the second connection schematic diagram of the server provided by the embodiment of the present application.

[0077] See Figure 7 , the server includes a mainboard, the mainboard includes a central processing unit 7-0, and the server also includes 3 backplanes, and 4 hard disks are provided on each backplane. The central processing unit 7-0 on the server is connected to the hard disks 0 and 1 on the backplane 1, the hard disks 4 and 5 on the backplane 2, and the hard disks 8 and 9 on the backplane 3. At this time, the server supports the single mode.

[0078] When it is necessary to add a central processing unit, a mainboard can be added to the server shown in Figure 7 , and the central processing unit on the mainboard is connected to one or more hard disks (including hard disks 2, 3, 6, 7, 10, 11), which can realize updating the server to work in the dual single-way mode, balancing the resource allocation of the two central processing units, improving the data processing efficiency, and reducing the situation of cable crossing, thereby improving the heat dissipation efficiency of the central server.

[0079] Figure 8 This is the third connection schematic diagram of the server provided by the embodiments of the present application.

[0080] Refer to Figure 8 , the server includes 4 mainboards, and each mainboard includes a central processing unit (central processing units 8-0, 8-1, 8-2, and 8-3 respectively). The server also includes 2 backplanes, and each backplane is provided with 4 hard disks. The central processing unit 8-0 on the server is connected to the hard disk 0 on the backplane 1 and the hard disk 4 on the backplane 2. The central processing unit 8-1 on the server is connected to the hard disk 1 on the backplane 1 and the hard disk 5 on the backplane 2. The central processing unit 8-2 on the server is connected to the hard disk 2 on the backplane 1 and the hard disk 6 on the backplane 2. The central processing unit 8-3 on the server is connected to the hard disk 3 on the backplane 1 and the hard disk 7 on the backplane 2. At this time, the server supports the multi-single path mode.

[0081] Optionally, the above central processing units 8-0, 8-1, 8-2, and 8-3 are all equipped with corresponding memory modules, riser brackets, riser cards, network cards and other devices, and each central processing unit can operate independently, thereby improving the data processing efficiency of the server.

[0082] Optionally, each mainboard further includes at least one riser bracket, and the riser brackets correspond to the central processing units one by one, and the riser brackets are adjacent to the corresponding central processing units and are connected to the corresponding central processing units through cables;

[0083] Each riser bracket includes a plurality of slots, and each slot is used to place a riser card;

[0084] The riser card includes at least one card slot, and each card slot is used to place a preset hardware device to integrate each preset hardware device onto the mainboard.

[0085] Optionally, each mainboard includes at least one riser bracket, and each riser bracket corresponds to the central processing unit on the above mainboard one by one, and each riser bracket is adjacent to the corresponding central processing unit, and each riser bracket is connected to the corresponding central processing unit through a cable, so that the length of the cable connecting the riser bracket and the corresponding central processing unit can be reduced.

[0086] Optionally, each riser bracket includes a plurality of slots, and each slot is used to place a riser card. The riser card is a structure including at least one card slot, and each card slot is used to place a preset hardware device, so as to realize the connection relationship between the preset hardware device and the central processing unit on the corresponding mainboard through the riser card and the riser bracket, and integrate each preset hardware device onto the corresponding mainboard.

[0087] Exemplarily, the preset hardware devices include at least one of, but are not limited to, network cards, hard disk drives, network adapters, graphics cards, network cards, etc.

[0088] Through the vertical plate brackets, backplanes, etc. connected to the central processing unit as described above, resource balanced allocation and installation of components such as hard disks and network cards of each central processing unit can be achieved, making the overall operation of the server more stable and the computing power faster. It can avoid problems such as resource imbalance, device jamming, downtime, and data loss caused by excessive load on some central processing units.

[0089] Optionally, the preset hardware devices include at least one network card, and each network card is connected to the corresponding central processing unit through a vertical plate card, a vertical plate bracket, and a cable.

[0090] Optionally, the preset hardware devices include at least one network card. Each network card is placed on a card slot on the vertical plate card, and the connection relationship between the preset hardware device and the corresponding central processing unit on the motherboard is achieved through the vertical plate card, the vertical plate bracket where the vertical plate card is located, and the cable connecting the vertical plate bracket to the central processing unit, so that the central processing unit can access the network for data exchange.

[0091] Optionally, each motherboard includes a first end face, and each hard disk is connected to the target central processing unit through a cable, and the cable is placed on the first end face of the motherboard.

[0092] Optionally, the motherboard can indicate a structure with a first end face and a second end face for placing the central processing unit and other components (such as IC chips, resistors, capacitors, inductors, etc.). Each hard disk on the motherboard is connected to the target central processing unit closest to the hard disk through a cable. In this way, the cable connecting each hard disk to the target central processing unit is shorter, and there will be no line crossing, which will not affect the heat dissipation of the central processing unit, thereby improving the processing efficiency of the central processing unit. And each cable is placed on the first end face of the motherboard, which is convenient for timely maintenance when the central processing unit or hard disk fails.

[0093] Optionally, each motherboard includes at least one memory module, and at least one memory module corresponds to at least one central processing unit one by one, and each central processing unit is communicatively connected to the corresponding memory module.

[0094] Optionally, each motherboard includes at least one memory module. At least one memory module on each motherboard corresponds to at least one central processing unit on the motherboard one by one, and each central processing unit on each motherboard is communicatively connected to the corresponding memory module through a cable. In this way, each memory module can perform functions such as storing data corresponding to the central processing unit corresponding to the memory module, facilitating the independent operation of each central processing unit.

[0095] Optionally, the server further includes at least one power supply for supplying power to the central processing units on each motherboard.

[0096] Figure 9 This is the second schematic diagram of the server provided by the embodiment of the present application.

[0097] Refer to Figure 9 , there are two mainboards and one power supply on the server. Each mainboard includes a central processing unit (central processing unit 9-0 and central processing unit 9-1 respectively), a memory module, and a riser card bracket. Each riser card bracket includes a slot, and a riser card is placed on the slot of the riser card bracket, and a network card is placed on the card slot of the riser card. Among them, each central processing unit is connected to the memory module and the riser card bracket on the mainboard where the central processing unit is located through a cable. The power supply is connected to each central processing unit through a cable to supply power to each central processing unit. Based on the connection relationship between each central processing unit and the corresponding memory module and riser card bracket, the network card can be integrated on the corresponding mainboard to realize the independent operation of each central processing unit.

[0098] By setting one central processing unit for each mainboard, it is possible to realize the communication between all resources to be allocated and the central processing unit, such as peripheral component interconnect express (PCIe) resources, Dual-Inline-Memory-Modules (DIMM) resources, Mini Cool Edge IO (MCIO) resources, Virtual Reality (VR) resources, complex programmable logic device CPLD, etc., so that each mainboard can form a complete ecological link. By configuring corresponding components such as memory modules for each central processing unit, the resources can be evenly divided, so that the system chain of each central processing unit can independently become a system ecosystem, reducing the length of the cable and reducing the communication threads between the central processing unit and each component, improving the security and computing speed of the central processing unit system, avoiding the problem of cross-Socket mutual access. When the server supports a system ecosystem with two or more central processing units, the server as a whole can synchronously execute the input and output of multiple commands, improving the data processing efficiency of the server.

[0099] By placing functional components such as capacitors, resistors, and inductors on the second end face of the mainboard, the area of the mainboard can be reduced, and during the process of manufacturing and assembling the server, damage to each component located on the second end face can be avoided.

[0100] Optionally, each mainboard further includes at least one resistor, at least one capacitor, and at least one inductor, and each resistor, each capacitor, and each inductor are placed on the second end face of the mainboard.

[0101] Optionally, each main board further includes at least one resistor, at least one capacitor, at least one inductor, and a second end face. The resistors, capacitors, and inductors are all components used to implement functions such as data transmission and data calculation on the server. Each resistor, capacitor, and inductor is placed on the second end face of the main board and is communicatively connected to the central processing unit.

[0102] In this way, while reducing the area of the main board, it is possible to avoid the situation of chaotic crossing of connection lines due to excessive components on the main board, thereby reducing the development cost of the main board.

[0103] Optionally, each main board includes at least one through hole. The central processing unit connection cables on each main board pass through the through holes and are connected to each capacitor, each resistor, and each inductor on the main board.

[0104] Optionally, each main board includes at least one through hole. The central processing unit connection cables on each main board pass through the through holes on the main board and are connected to each capacitor, resistor, and inductor located on the second end face of the main board, thereby implementing functions such as data transmission and data calculation of the main board.

[0105] Optionally, each main board further includes at least one chip. The chip is placed on the first end face of the main board. The chips on each main board are connected to the central processing units through cables;

[0106] The chip is used to communicate with the central processing unit and perform data transmission, data processing, and data calculation.

[0107] Optionally, each main board on the server further includes at least one chip. The chip is placed on the first end face of the main board. The pins of the chip on each main board are connected to the central processing unit on the main board through cables. While reducing the length of the cables between the chip and the central processing unit connection, communication with the central processing unit is achieved, enabling the chip to perform operations such as data transmission, data processing, and data calculation based on communication with the corresponding central processing unit.

[0108] By placing the components that need to be wired, as well as key communication components such as chips and memory modules, on the first end face of the main board, which is the same end face as the central processing unit, while improving communication efficiency, it is convenient for each central processing unit to manage and monitor resources, as well as various modules on the system link, etc. In the load balancing mode, each central processing unit can automatically allocate resources according to the load situation and automatically adjust the operating frequency. In this way, in the case of low load, a certain amount of power can be saved to a certain extent, and in the case of high load, higher performance can be provided.

[0109] The above has introduced in detail a server provided by the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present application, several improvements and modifications can still be made to the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. A server, characterized in that, Comprising at least one main board and at least one backplane; Each of the main boards includes a central processing unit, and the central processing units are independently arranged; At least one hard disk is provided on each of the backplanes; For each of the backplanes, based on the disk order of the hard disks on the backplane, the hard disks are sequentially connected to the target central processing unit; wherein, the target central processing unit is used to indicate the central processing unit with the smallest distance from the hard disk.

2. The server according to claim 1, wherein The number of hard disks connected to each central processing unit is inversely proportional to the number of central processing units.

3. The server according to claim 1, wherein Each of the main boards further includes at least one vertical board bracket, the vertical board brackets correspond to the central processing units one by one, and the vertical board bracket is adjacent to the corresponding central processing unit and is connected to the corresponding central processing unit by a cable; Each of the vertical board brackets includes a plurality of slots, and each of the slots is used for placing vertical board cards; The vertical board card includes at least one card slot, and each of the card slots is used for placing a preset hardware device to integrate each of the preset hardware devices onto the main board.

4. The server according to claim 3, wherein The preset hardware device includes at least one network card, and each of the network cards is connected to the corresponding central processing unit through the vertical board card, the vertical board bracket and the cable.

5. The server according to claim 1, wherein Each of the main boards includes a first end face, and each of the hard disks is connected to the target central processing unit by a cable, and the cable is placed on the first end face of the main board.

6. The server according to claim 1, characterized in that, Each of the main boards includes at least one memory module, the at least one memory module corresponds to the at least one central processing unit one by one, and each of the central processing units is communicatively connected to the corresponding memory module.

7. The server according to claim 1, characterized in that Each of the main boards further includes at least one resistor, at least one capacitor, and at least one inductor, and each of the resistors, each of the capacitors, and each of the inductors are placed on the second end face of the main board.

8. The server according to claim 7, wherein Each of the main boards includes at least one through hole, and the central processing unit connection cable on each main board passes through each of the through holes and is connected to each of the capacitors, each of the resistors, and each of the inductors on the main board.

9. The server according to claim 1, characterized in that, Each of the main boards further includes at least one chip, the chip is placed on the first end face of the main board, and the chip on each main board is connected to each of the central processing units by a cable; The chip is used to communicate with the central processing unit and perform data transmission, data processing, and data calculation.

10. The server according to claim 1, characterized in that, The server further includes at least one power supply, and the power supply is used to supply power to each of the central processing units.

Images