A data center, server communication method, system, product and medium

Abstract

The application discloses a kind of data center, server communication method, system, product and medium, it is related to the field of server, to solve the problem of idle resource and signal transmission difficulty in multiple host server system, the server system includes multiple computing groups, each computing group includes multiple computing devices, any two computing devices in computing group are interconnected;N exchange chips, exchange chip is configured to identify each computing device connected with itself;First management component is connected with the first transmission interface on exchange chip, first management component is configured to obtain all computing devices identified by exchange chip, allocate independent identification for all computing devices, and configure the routing table of all optical ports on exchange chip based on independent identification, so that exchange chip forwards data according to routing table.The application improves the communication efficiency of multiple host server system, reduces the idle resource phenomenon, and improves the overall performance of data center.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of servers, in particular to a data center, a server communication method, a system, a product and a medium. BACKGROUND

[0002] With the rapid development of cloud computing and informatization, the scale of data centers is expanding, and the demand for servers is increasing. GPUs (Graphics Processing Units) are often used in servers for accelerated computing. In a multi-host server system, a chassis contains multiple hosts, each host is connected to multiple GPUs, and each host can usually only communicate with the GPUs connected to itself. Communication between GPUs under the host needs to be forwarded by the host, which can easily cause resource idling and signal transmission difficulties. This limitation not only affects the scalability of the data center, but also limits the potential of high-performance computing.

[0003] Therefore, how to provide a solution to the above technical problems is a problem that those skilled in the art need to solve at present. SUMMARY

[0004] The purpose of the present application is to provide a data center, a server communication method, a system, a product and a medium, to improve the communication efficiency of a multi-host server system, reduce resource idling, and improve the overall performance of the data center.

[0005] To solve the above technical problems, the present application provides a server system, comprising:

[0006] a plurality of computing groups, each computing group comprising a plurality of computing devices, any two computing devices in the computing group being interconnected; n switching chips, the switching chip being configured to identify each computing device connected thereto, a plurality of optical ports on the i-th switching chip corresponding to the i-th computing device in a plurality of computing groups, i=1, 2, …, n; a first management component connected to a first transmission interface on the switching chip, the first management component being configured to obtain all the computing devices identified by the switching chip, assign independent identifiers to all the computing devices, and configure a routing table for all the optical ports on the switching chip based on the independent identifiers, so that the switching chip forwards data according to the routing table; the routing table comprising a correspondence between each optical port and the independent identifier of the computing device connected thereto.

[0007] Optionally, the server system further comprises a plurality of hosts, the hosts being connected with the at least two computing groups respectively, the hosts being configured to determine a first target computing device in all the computing devices, determine a second target computing device in all the computing devices connected with the host itself, and access the first target computing device through the second target computing device and a switch chip connected with the second target computing device.

[0008] Optionally, the first management component is configured to acquire all the computing devices identified by the switch chip, allocate an address space for all the computing devices, allocate an independent identifier for all the computing devices in the address space, and configure a routing table of all the optical ports on the switch chip based on the independent identifier, so that the switch chip forwards data according to the routing table.

[0009] Optionally, the first management component is further configured to control n switch chips to perform a reset operation.

[0010] Optionally, the first management component is further configured to identify transmission information of each optical port, and adjust the interface specification and / or interface protocol of each optical port based on the transmission information; the transmission information comprises a device type of a computing device connected with the optical port and a data transmission rate of the optical port.

[0011] Optionally, the server system further comprises a switch board, the switch board being provided with at least one switch chip, the switch board being further provided with a plurality of optical modules connected with a plurality of optical ports on the switch chip correspondingly, and the computing devices being connected with the optical ports through the optical modules.

[0012] Optionally, the server system further comprises a first programmable device, the first programmable device being arranged on the switch board, and the first programmable device being configured to perform a reset operation on the optical module arranged on the switch board where the first programmable device is arranged, so that the switch chip arranged on the switch board where the first programmable device is arranged identifies a computing device connected with the optical module.

[0013] Optionally, the first programmable device is configured to perform a reset operation on each optical module connected with the switch chip on the switch board where the first programmable device is arranged when power-on of the switch chip on the switch board where the first programmable device is arranged is completed, and / or perform a reset operation on each optical module connected with the switch chip on the switch board where the first programmable device is arranged when receiving a reset signal sent by the first management component.

[0014] Optionally, the first programmable device is further configured to output a mode control signal; the mode control signal comprises a first control signal corresponding to a host software control mode and a second control signal corresponding to a module hardware control mode; the optical module is configured to adjust the current power mode to a low power mode in response to the first control signal and adjust the current power mode to a high power mode in response to the second control signal.

[0015] Optionally, the optical module is further configured to adjust the current power mode to the high power mode in response to receiving a host switching signal when in the low power mode.

[0016] Optionally, the first programmable device is further configured to perform a corresponding protection operation in response to an alarm signal output by the optical module.

[0017] Optionally, the server system further comprises a second programmable device configured to control power-on of the first management component upon receiving a power-on signal and output an allow power-on signal, and output a trigger signal to the first management component upon receiving a complete power-on signal; the first programmable device is further configured to control power-on of the switching board on which the first programmable device is located upon receiving the allow power-on signal, and output the complete power-on signal upon completion of power-on of the switching board on which the first programmable device is located; the first management component is further configured to obtain all the computing devices identified by the switching chip in response to receiving the trigger signal, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the switching chip based on the independent identifiers, so that the switching chip forwards data according to the routing table.

[0018] Optionally, the second programmable device is further configured to identify in-place information of each of the switching boards and output a logical judgment instruction according to the in-place information; the first management component is further configured to perform a corresponding management operation in response to the logical judgment instruction.

[0019] Optionally, the server system further comprises an interface indication board, the interface indication board comprising a plurality of state indication lamps; the second programmable device is further configured to control the plurality of state indication lamps to perform indication operations according to respective corresponding current state information.

[0020] Optionally, the server system further comprises: a plurality of fan modules; a plurality of first temperature acquisition modules configured to acquire first temperature information of corresponding switching chips; a plurality of second temperature acquisition modules configured to acquire second temperature information of corresponding computing groups; and a second management component configured to regulate rotation speeds of the fan modules based on all the first temperature information and all the second temperature information.

[0021] To solve the above technical problems, the application further provides a data center comprising a plurality of server systems as described in any of the above.

[0022] To solve the above technical problems, the application further provides a server communication method applied to the server system as described in any of the above, comprising: identifying, by a switch chip in the server system, each computing device connected thereto; obtaining, by a first management component in the server system, all the computing devices identified by the switch chip, assigning independent identifiers to all the computing devices, and configuring a routing table of all optical ports on the switch chip based on the independent identifiers; after receiving data sent by a first computing device through any optical port of the switch chip in the server system, transmitting the data to a second computing device through a target optical port according to the routing table.

[0023] To solve the above technical problems, the application further provides a server communication system applied to the server system as described in any of the above, comprising: an identification module configured to identify, by a switch chip in the server system, each computing device connected thereto; an establishment module configured to obtain, by a first management component in the server system, all the computing devices identified by the switch chip, assign independent identifiers to all the computing devices, and configure a routing table of all optical ports on the switch chip based on the independent identifiers; and a routing module configured to, after receiving data sent by a first computing device through any optical port of the switch chip in the server system, transmit the data to a second computing device through a target optical port according to the routing table.

[0024] To solve the above technical problems, the application further provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the server communication method described above.

[0025] To solve the above technical problems, the application further provides a computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the server communication method described above.

[0026] The application provides a server system, which first divides a plurality of computing devices connected under each host into a device group, any two computing devices in the device group are interconnected, communication efficiency between the computing devices under the same host is realized, then n switching chips are set, a plurality of optical ports on the i switching chip correspond to the i computing device in the plurality of computing groups connected, interconnection between the device groups is realized through the switching chip, a first management component is responsible for acquiring all the computing devices identified by the switching chip, and independently assigns an identifier to the devices, and the assignment mode ensures the uniqueness of each computing device in the network, based on the independent identifier of the computing device, the first management component configures a routing table of all the optical ports on the switching chip. The routing table contains the corresponding relationship between each optical port and the independent identifier of the computing device connected thereto, so that the switching chip can efficiently forward data according to the routing table, and through direct connection and routing forwarding of the switching chip, the host can access any computing device in the server system, so that the communication of the multi-host server system is more efficient, the resource idle phenomenon is reduced, and the overall performance of the data center is improved.

[0027] The application also provides a data center, a server communication method, a server communication system, a computer program product and a computer readable storage medium. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the application, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0029] Figure 1 A structural schematic diagram of a server system provided by the application.

[0030] Figure 2 A schematic diagram of interconnection of a plurality of computing devices provided by the application.

[0031] Figure 3 A structural schematic diagram of a switching board provided by the application.

[0032] Figure 4 A structural schematic diagram of another server system provided by the application. DETAILED DESCRIPTION

[0033] The core of the application is to provide a data center, a server communication method, a system, a product and a medium, to improve the communication efficiency of a multi-host server system, reduce the resource idle phenomenon, and improve the overall performance of the data center.

[0034] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0035] In a first aspect, referring to Figure 1 The present application provides a server system, comprising:

[0036] a plurality of computing groups 1, each computing group 1 comprising a plurality of computing devices 11, any two computing devices 11 in the computing group 1 being interconnected.

[0037] n switching chips SW, the switching chip SW being configured to identify each computing device 11 connected thereto, a plurality of optical ports on the i th switching chip SW corresponding to the i th computing device 11 in the plurality of computing groups 1, i = 1, 2, …, n.

[0038] a first management component 21 connected to a first transmission interface on the switching chip SW, the first management component 21 being configured to obtain all computing devices 11 identified by the switching chip SW, assign independent identifiers to all computing devices 11, and configure a routing table for all optical ports on the switching chip SW based on the independent identifiers, so that the switching chip SW forwards data according to the routing table; the routing table comprising a correspondence between each optical port and the independent identifier of the computing device 11 connected thereto.

[0039] In the present embodiment, the multi-host server system comprises a plurality of computing motherboards, each computing motherboard can connect a plurality of computing devices 11, the plurality of computing devices 11 connected to each computing motherboard is divided into at least one computing group 1, and the plurality of computing devices 11 in the same computing group 1 are interconnected two by two, so that the computing devices 11 in the computing group 1 can interact two by two without passing through the host on the computing motherboard. The computing devices 11 in the computing group 1 can be interconnected through a first communication bus, which includes but is not limited to a PCIe (Peripheral Component Interconnect Express, Peripheral Component Interconnect Express) bus.

[0040] The server system further comprises n switch chips SW, each of which comprises a plurality of optical ports, and each of the plurality of computing devices 11 in the computing group 1 is connected to an optical port on the plurality of switch chips SW. In order to improve the utilization rate of the optical port resources on the switch chip SW, as an optional embodiment, when the computing devices 11 connected under the same computing mainboard are divided, the division can be performed according to the number of switch chips SW, for example, there are four switch chips SW, and the four computing devices 11 connected under the same computing mainboard can be divided into one computing group 1, the first computing device 11 in the computing group 1 is connected to any optical port of the first switch chip, the second computing device 11 in the computing group 1 is connected to any optical port of the second switch chip, the third computing device 11 in the computing group 1 is connected to any optical port of the third switch chip, and the fourth computing device 11 in the computing group 1 is connected to any optical port of the fourth switch chip, so that all the computing devices 11 connected under the same computing mainboard can be connected together through the switch chip SW. It can be understood that the computing device 11 and the switch chip SW can be connected through a second communication bus, and the second communication bus includes but is not limited to a PCIe bus. The first communication bus and the second communication bus can be the same or different, and can be selected according to actual engineering needs. The present embodiment is not limited here. Figure 1 Only four computing groups 1 and two switch chips SW are shown in the figure, and each computing group is shown with two computing devices 11. The specific architecture can be arranged according to actual engineering needs.

[0041] Referring to Figure 2 As shown in the figure, S0-S7 refers to eight GPUs (i.e. the computing device 11 in the present embodiment) connected to a single computing mainboard, and there are eight computing mainboards in total, so there are eight groups S0-S7 in the figure. Among them, S0, S3, S4 and S7 are interconnected inside the host, and these four GPUs can be regarded as a whole. The 64 GPUs in the entire architecture can be divided into 16 groups, each group having four GPUs, and data and information transmission between them can be performed through the cables connected to each other. Each of the four GPUs is connected to a different switch chip SW, as shown in Figure 2 As shown in the figure, S0 in the upper left corner is connected to the first switch chip, S4 is connected to the second switch chip, S3 is connected to the third switch chip, and S7 is connected to the fourth switch chip. Thus, each switch chip SW has 16 external interfaces and will be connected to a certain GPU of each GPU group (i.e. the computing group 1 in the present embodiment), so that all the GPUs connected to the mainboard can be connected together through the switch chip SW. The switch chip SW is configured to recognize the computing device 11 connected thereto.

[0042] The first management component 21 is connected with the first transmission interface of each switch chip SW through a third communication bus, which can be an I2C (Inter-Integrated Circuit) bus or the like. The transmission rate of the third communication bus can be less than the transmission rate of the first communication bus and the transmission rate of the second communication bus. The first management component 21 can be an mCPU (micro Central Processing Unit). The first management component 21 can obtain the computing devices 11 recognized by each switch chip SW through the third communication bus, and then assign independent identifiers to each computing device 11. The independent identifiers can be independent numbers or independent addresses assigned. The independent identifiers of the computing devices 11 are ensured to be different from each other. Then, the routing table of all optical ports on each switch chip SW is configured based on the independent identifiers, so as to determine which computing device 11 is connected to each optical port on each switch chip SW. When data is forwarded through the switch chip SW, the routing table can be used to determine which computing device is the target computing device, so as to select the corresponding optical port for data transmission, thereby improving the data transmission efficiency and accuracy.

[0043] As an optional embodiment, the optical ports on the switch chip SW for connecting the computing devices 11 are configured in a PCIe EP (Endpoint) mode.

[0044] Exemplarily, it is assumed that there are four switch chips SW, which are a first switch chip, a second switch chip, a third switch chip and a fourth switch chip. Eight GPUs (S0-S7) under each computing mainboard are divided into two computing groups 1, each group containing four GPUs. For example, the first group is S0, S1, S2, S3, and the second group is S4, S5, S6, S7. S0 in the first group is connected to one optical port of the first switch chip, S1 is connected to one optical port of the second switch chip, S2 is connected to one optical port of the third switch chip, and S3 is connected to one optical port of the fourth switch chip. S4 in the second group is connected to one optical port of the first switch chip, S5 is connected to one optical port of the second switch chip, S6 is connected to one optical port of the third switch chip, and S7 is connected to one optical port of the fourth switch chip. Within each computing group 1, the GPUs are interconnected through a first communication bus (such as a PCIe bus). For example, S0, S1, S2, S3 are interconnected through a PCIe bus. The first management component 21 is connected with the switch chip SW through a third communication bus (such as an I2C bus) to obtain information of the computing device 11 recognized by the switch chip SW. The first management component 21 allocates an independent identifier for each GPU, for example: the independent identifier of S0 is G1, the independent identifier of S1 is G2, and so on. The first management component 21 configures a routing table of all optical ports on the switch chip SW according to the independent identifiers, including that one optical port on the switch chip is connected to G1 (S0) and the other optical port is connected to G5 (S4). When G1 (S0) needs to perform data transmission with G5 (S4), the data is first transmitted to the switch chip within the computing group 1 through the first communication bus (PCIe bus), and then transmitted to G5 (S4) through the optical port of the switch chip according to the routing table.

[0045] The multi-host server system architecture in this embodiment can realize efficient and accurate data transmission, and improve the utilization rate of optical port resources on the switch chip SW.

[0046] It can be seen that in the embodiment, first, the plurality of computing devices 11 connected under each host is divided into a device group, any two computing devices 11 in the device group are interconnected, the communication efficiency between the computing devices 11 under the same host is realized, then the n switching chips SW are set, the plurality of optical ports on the i th switching chip SW correspond to the i th computing device 11 in the plurality of computing groups 1, the interconnection between each device group is realized through the switching chip SW, the first management component 21 is responsible for acquiring all the computing devices 11 recognized by the switching chip SW, and independently identifies these devices. This allocation method ensures the uniqueness of each computing device 11 in the network, based on the independent identification of the computing device 11, the first management component 21 configures the routing table of all optical ports on the switching chip SW. The routing table contains the correspondence between each optical port and the independent identification of the computing device 11 connected thereto, so that the switching chip SW can efficiently forward data according to the routing table. Through direct connection and routing forwarding of the switching chip SW, the host can access any computing device 11 in the server system, so that the communication of the multi-host server system is more efficient, the resource idling phenomenon is reduced, and the overall performance of the data center is improved.

[0047] On the basis of the above embodiment:

[0048] In an exemplary embodiment, the server system further includes a plurality of hosts, the host is connected with at least two corresponding computing groups 1, the host is configured to determine a first target computing device in all computing devices 11, determine a second target computing device in all computing devices 11 connected with itself, and access the first target computing device through the second target computing device and the switching chip SW connected with the second target computing device.

[0049] In the embodiment, each computing mainboard includes a host, the host can access the computing devices 11 connected under other hosts through the computing devices 11 connected thereunder, so that each host can read the information of all computing devices 11.

[0050] Suppose it includes two hosts, host A connects the first computing group 1C1 (including computing devices 11 with independent identifications G1-G4) and the second computing group 1C2 (including computing devices 11 with independent identifications G5-G8), host B connects the third computing group 1C3 (including computing devices 11 with independent identifications G9-G12) and the fourth computing group 1C4 (including computing devices 11 with independent identifications G13-G16), the computing devices 11 in each computing group 1 are interconnected through the first communication bus (such as PCIe bus), and the computing devices 11 of each computing group 1 are connected with the corresponding switching chip SW.

[0051] If host A needs to access G10 located in computing group 1C3, host A first determines G5 in computing group 1C2 directly connected to host A as the second target computing device, and then sends a request to the switch chip SW connected to computing group 1C2 through G5. The switch chip SW forwards the request to the switch chip SW connected to computing group 1C3 according to the routing table, and the switch chip SW receiving the request forwards the request to G10. Host B accesses computing device 11 under host A: assuming that host B needs to access G2 located in computing group 1C1, host B first determines G13 in computing group 1C4 directly connected to host B as the second target computing device, and then sends a request to the switch chip SW connected to computing group 1C4 through G13. The switch chip SW forwards the request to the switch chip SW connected to computing group 1C1 according to the routing table, and the switch chip SW receiving the request forwards the request to G2. When G10 receives the request from host A, it can perform a computing task and return the result to host A through the switch chip SW network. Similarly, when G2 receives the request from host B, it can also perform a computing task and return the result to host B through the switch chip SW network.

[0052] In this way, each host can access and read the information of all computing devices 11 in the server system through the computing group 1 and the switch chip SW network directly connected thereto, realizing the sharing of resources and efficient data transmission.

[0053] In an exemplary embodiment, the first management component 21 is configured to obtain all computing devices 11 identified by the switch chip SW, allocate an address space to all computing devices 11, allocate an independent identifier to all computing devices 11 in the address space, and configure the routing table of all optical ports on the switch chip SW based on the independent identifier, so that the switch chip SW forwards data according to the routing table.

[0054] In this embodiment, the first management component 21 allocates a unified address space to all computing devices 11 in the server system. This address space can be virtual or physical, but it must be large enough to allocate an independent identifier to all devices. In the address space, the first management component 21 allocates a unique independent identifier to each computing device 11. This identifier can be a MAC (Media Access Control Address) address, an IP (Internet Protocol Address) address, or a custom number, as long as the identifier of each device is unique.

[0055] For example, the first computing device 11 is assigned an identity of 0x01, the second computing device 11 is assigned an identity of 0x02, the sixteenth computing device 11 is assigned an identity of 0x10, and so on, until each computing device 11 is assigned an individual identity.

[0056] The first management component 21 configures the routing table of all optical ports on the switch chip SW according to the assigned individual identities. The routing table will indicate how each switch chip SW forwards data according to the destination identity of the data packet. Taking the computing device 11 as a GPU for example, the optical port 0 of the first switch chip is connected to GPU 01 (identity 0x01), the optical port 1 of the first switch chip is connected to GPU 05 (identity 0x05), and the routing table entry can be as follows: destination identity 0x01→optical port 0, destination identity 0x05→optical port 1, and so on. When the switch chip SW receives a data packet, it will check the destination identity of the data packet and determine which optical port should be used to forward the data according to the routing table. For example, if the first switch chip receives a data packet with a destination identity of 0x02, it will look up the routing table and find that the data should be forwarded through the optical port connected to GPU 2.

[0057] In this way, the first management component 21 ensures that the communication between all computing devices 11 in the server system is efficient and accurate, and each switch chip SW can quickly forward data according to the configured routing table without the need to pass through the host processor, thereby realizing P2P (Peer-to-Peer) communication between GPU EP devices, reducing latency and improving the overall performance of the system.

[0058] In an exemplary embodiment, the first management component 21 is further configured to control the n switch chips SW to perform a reset operation.

[0059] In this embodiment, the first management component 21 first initializes a reset command, which will be sent to all switch chips SW to instruct them to perform a reset operation. Specifically, the first management component 21 sends the reset command to each switch chip SW through a control interface (such as I2C, SPI (Serial Peripheral Interface), network interface, etc.). This process can be performed in parallel or sequentially, and the actual engineering needs can be set, which is not limited in this embodiment.

[0060] Upon receiving the reset command, each switch chip SW sends an acknowledgement signal to the first management component 21, indicating that it has received the reset command. Upon receiving the reset command, each switch chip SW begins performing the reset operation. The reset operation includes, but is not limited to, the following operations: disconnecting all current data connections and sessions; resetting internal states, including routing tables, caches, counters, etc.; resetting hardware registers to default states; and possibly including hardware-level resets, such as power cycling.

[0061] The first management component 21 monitors the reset process of each switch chip SW to ensure that the reset operation is performed correctly. If an error occurs during the reset process or a switch chip SW does not respond, the first management component 21 can take further measures, such as resending the reset command or recording error information. Once the switch chip SW completes the reset operation, it sends a reset completion signal to the first management component 21. After the reset is completed, the first management component 21 may need to reconfigure the switch chip SW, including reallocating address space, unique identification, and configuring routing tables, to ensure that the switch chip SW can work normally again.

[0062] In an exemplary embodiment, the first management component 21 is further configured to identify transmission information of each optical port, and adjust the interface specification and / or interface protocol of each optical port based on the transmission information. The transmission information includes the device type of the computing device 11 connected to the optical port and the data transmission rate of the optical port.

[0063] In this embodiment, the first management component 21 first monitors the device type of the computing device 11 connected to each optical port and the data transmission rate. This information can be obtained by querying the device or by reporting from the switch chip SW. Based on the monitored transmission information, the first management component 21 analyzes the data transmission requirements of each optical port. For example, if a high-performance GPU is connected to an optical port, it may require a higher data transmission rate and a specific interface protocol. The first management component 21 determines which interface specification and / or interface protocol (PCIe or CXL (Compute Express Link) and the like) of each optical port needs to be adjusted according to the transmission requirements. This may include adjusting the data transmission rate, modifying the signal encoding method, changing the error checking mechanism, etc. The first management component 21 sends commands to the switch chip SW to perform specific adjustment operations, which may involve modifying the hardware register settings of the switch chip SW or updating the firmware configuration. This process ensures that each optical port is optimized according to the characteristics of the computing device 11 connected to it, thereby improving the performance and efficiency of the entire system.

[0064] In an example embodiment, the server system further comprises a switching board, the switching board is provided with at least one switching chip SW, and the switching board is further provided with a plurality of optical modules 32 connected with a plurality of optical ports on the switching chip SW in a one-to-one manner, and the computing device 11 is connected with the optical port through the optical module 32.

[0065] In the embodiment, the hardware topology of the switching chip SW is as shown in Figure 3 The switching chip SW includes but is not limited to 18 PCIe interfaces of X8, of which 16 PCIe interfaces of X8 are used as optical ports, Figure 3 The port numbers of PE0-PE7 are all 0-7, each optical port is connected to a QSFP-DD connector, and each QSFP-DD connector is connected with the computing device 11 through the external optical module 32. The remaining two X8 PCIe interfaces are used for communication with other switching chips SW and are used as reserved interfaces. The clock signal of the switching chip SW is from the clock generator clk generator on the board card.

[0066] In the embodiment, at least one switching chip SW is arranged on each switching board, and the switching chips SW are arranged according to a certain layout, 16 optical ports on each switching chip SW are connected to a QSFP-DD (Quad Small Form Factor Pluggable-Double Density, double density four-channel small form factor pluggable package) connector on the switching board, and the QSFP-DD connector is connected with the computing device 11 (such as GPU) through the optical module 32. The optical module 32 is responsible for converting electrical signals into optical signals to realize high-speed data transmission.

[0067] In an example embodiment, referring to Figure 3 , the server system further comprises a first programmable device 31, the first programmable device 31 is arranged on the switching board, and the first programmable device 31 is configured to perform a reset operation on the optical module 32 arranged on the switching board where the first programmable device 31 is arranged, so that the switching chip SW arranged on the switching board where the first programmable device 31 is arranged recognizes the computing device 11 connected with the optical module 32.

[0068] In the embodiment, the switching board can be further provided with the first programmable device 31, the first programmable device 31 includes but is not limited to a CPLD (Complex Programmable Logic Device, complex programmable logic device), the first programmable device 31 controls the optical module 32 on the switching board where the first programmable device 31 is arranged to reset, and once the optical module 32 is reset, the first programmable device 31 informs the switching chip SW to re-detect the computing device 11 connected with the optical module 32.

[0069] It can be understood that after the system is started or the optical module 32 is replaced, the reset operation can ensure that the optical module 32 returns to a known initial state, if the optical module 32 encounters errors or exceptions in previous work, the reset operation can help clear these error states, so that the optical module 32 can resume normal work, in some cases, the connection between the optical module 32 and the computing device 11 may be disconnected due to software or hardware problems, and the reset operation can reestablish these connections. The reset operation can make the switch chip SW re-detect the presence and state of the optical module 32, so that the system correctly identifies all hardware components.

[0070] In an exemplary embodiment, the first programmable device 31 is configured to perform a reset operation on each optical module 32 connected to the switch chip SW on the switch board on which the first programmable device 31 is located when the switch chip SW on the switch board on which the first programmable device 31 is located is powered on, and / or when a reset signal sent by the first management component 21 is received.

[0071] In this embodiment, the management board on which the first management component 21 is arranged sends a reset signal (SW_PERST) to the switch chip SW for resetting the switch chip SW, and in addition, 16 groups of RESET signals (reset signals) are connected to the first programmable device 31. After the first programmable device 31 receives the RESET signal, the first programmable device 31 sends the RESET signal to the 16 QSFP-DD optical modules. The first programmable device 31 has two mechanisms for sending the RESET signal, one is self-control, which sends the RESET signal to the QSFP-DD optical module after the switch chip SW is powered on and the clock is stable. This process ensures that the optical module 32 is in the correct initial state before the switch chip SW is ready. The other is to wait for the RESET signal sent by the management board and then send it to the 16 QSFP-DD optical modules. This mechanism allows the system administrator or the first management component 21 to remotely control the reset of the optical module 32.

[0072] In an exemplary embodiment, the first programmable device 31 is further configured to output a mode control signal; the mode control signal includes a first control signal corresponding to a host software control mode and a second control signal corresponding to a module hardware control mode; and the optical module 32 is configured to adjust the current power mode to a low power mode in response to the first control signal, and adjust the current power mode to a high power mode in response to the second control signal.

[0073] In an exemplary embodiment, the optical module 32 is further configured to adjust the current power mode to a high power mode in response to receiving a host switching signal when in a low power mode.

[0074] In this embodiment, the signal sent to the optical module 32 also includes a LPM (Low Power Mode) signal (i.e. the mode control signal in this embodiment), which allows the host to define whether the QSFP-DD optical module is initialized in a host software control mode (LPM set to high) or a module hardware control mode (LPM set to low). In the host software control mode, the optical module 32 should remain in low power mode until the software can transition to a high power mode defined in the QSFP-DD management interface specification. In the module hardware control mode, the optical module 32 can transition to a high power mode immediately after the management interface initialization.

[0075] In the low power mode, the optical module 32 consumes less power, typically used for initialization or standby state, and in the high power mode, the optical module 32 reaches full power operation, suitable for data transmission. The host software control mode allows system administrators to dynamically adjust the power mode of the optical module 32 according to actual needs, providing higher system flexibility and manageability. When switching from low power mode to high power mode for data transmission is required, the optical module 32 can quickly respond to the host switching signal, ensuring the timeliness and efficiency of data transmission. In the module hardware control mode, the optical module 32 can transition to a high power mode immediately after the management interface initialization, which helps to ensure the stability and safety of the system. This power mode control mechanism provides more efficient energy management, better device performance and longer device life for data centers and network equipment, while enhancing the overall reliability and flexibility of the system.

[0076] In an exemplary embodiment, the first programmable device 31 is further configured to perform corresponding protection operations in response to the alarm signal output by the optical module 32.

[0077] In this embodiment, the optical module 32 and the first programmable device 31 also transmit the optical module 32's present signal (PRSNT) and the optical module 32's alarm signal (INTL) between them. When the first programmable device 31 receives the optical module 32's alarm signal, it performs corresponding protection operations. Specifically, when the optical module 32 detects internal faults or abnormal conditions, it triggers an alarm signal (INTL), which is sent to the first programmable device 31. After receiving the alarm signal, the first programmable device 31 will perform a series of protection operations according to the preset program, which includes but is not limited to resetting the optical module 32 to try to restore its function, removing the affected optical module 32 from service to avoid further data transmission errors, notifying the management system or host about the alarm event for further diagnosis and maintenance, switching to a backup optical module 32 or path to maintain the continuity of data transmission.

[0078] Figure 3The exchange board also includes power connectors, management connectors, sensors, replaceable switches, bidirectional switches, arbitration modules, etc.

[0079] In an exemplary embodiment, referring to Figure 4 , the server system further includes:

[0080] The second programmable device 22 is configured to control the first management component 21 to be powered on when receiving a power-on signal, and output an allowed power-on signal, and output a trigger signal to the first management component 21 after receiving a complete power-on signal; the first programmable device 31 is further configured to control the exchange board where the first programmable device 31 is located to be powered on after receiving the allowed power-on signal, and output the complete power-on signal when the exchange board where the first programmable device 31 is located is powered on; and the first management component 21 is further configured to, in response to receiving the trigger signal, obtain all computing devices 11 recognized by the switch chip SW, assign independent identifiers to all computing devices 11, and configure a routing table of all optical ports on the switch chip SW based on the independent identifiers, so that the switch chip SW forwards data according to the routing table.

[0081] In the embodiment, the second programmable device 22 is also disposed on the management board, and the second programmable device 22 includes but is not limited to a CPLD. The second programmable device 22, the first programmable device 31, and the first management component 21 cooperate to ensure that the exchange board and the switch chip SW are correctly powered on and configured, so as to forward data.

[0082] Specifically, when the system is powered on, the second programmable device 22 receives a power-on signal, controls the power-on of other components on the management board, such as the power-on of the first management component 21, and outputs an allowed power-on signal. The first programmable logic device receives the allowed power-on signal, controls the power-on of the switching board where it is located, and outputs a completed power-on signal when the power-on of the switching board is completed. The second programmable device 22 receives the completed power-on signal, and the second programmable device 22 outputs a trigger signal to the first management component 21. The first management component 21 receives the trigger signal output by the second programmable device 22 and performs the operation of obtaining the identification of the computing device 11 by the switching chip SW. In this way, it can be ensured that the system components are powered on in the correct order, and the damage of hardware caused by improper power-on sequence can be avoided. The first management component 21 is responsible for configuration and management, which simplifies the complexity of the system. The cooperation between multiple programmable devices improves the reliability and stability of the system. After the power-on of the switching chip SW is completed, the first management component 21 obtains the computing device 11 identified by the switching chip SW, ensures that the obtained device information is accurate and correct, and ensures that the device identification is performed after the switching chip SW is completely powered on and stably operated, which can avoid identification errors or device damage caused by unstable power supply. Moreover, by performing device identification after the power-on is completed, it can be ensured that all components in the system are in a known and synchronized state. Before the switching chip SW is completely powered on, the device identification is not performed, which can avoid operations during chip initialization, reduce operation conflicts and potential errors. The power-on logic of the present embodiment ensures the stability and reliability of the system during the startup and configuration process, reduces the possibility of errors, and improves the overall performance and user experience of the system.

[0083] In an exemplary embodiment, the second programmable device 22 is further configured to identify the in-place information of each switching board, and output a logical decision instruction according to the in-place information; and the first management component 21 is further configured to perform a corresponding management operation in response to the logical decision instruction.

[0084] In this embodiment, the second programmable device 22 is configured to perform logical judgment, specifically, the in-place information of each switching board can be judged. The second programmable device 22 identifies whether each switching board is in place (i.e., whether it has been correctly inserted and connected to the system) through a hardware detection mechanism (such as a switch, a sensor, or an EEPROM (Electrically Erasable Programmable Read-Only Memory) on an I2C bus). According to the in-place information of the switching board, the second programmable device 22 outputs logical judgment instructions. These instructions can be simple in-place / out-of-place states, or more complex logical combinations, indicating how the system responds to different configuration situations. The first management component 21 receives the logical judgment instructions from the second programmable device 22 and performs management operations, including but not limited to: if the switching board is in place, performing normal configuration and management operations; if the switching board is not in place, one of the following operations can be performed: ignoring the switching board, not configuring it, recording error information, and notifying the system administrator, trying to re-detect the in-place state of the switching board, adjusting the system configuration to adapt to the absence of the switching board. This embodiment improves the flexibility and robustness of the multi-host server system.

[0085] In an exemplary embodiment, the server system further comprises: a plurality of fan modules; a plurality of first temperature acquisition modules, the first temperature acquisition modules being configured to acquire first temperature information of corresponding switching chips SW; a plurality of second temperature acquisition modules, the second temperature acquisition modules being configured to acquire second temperature information of corresponding computing groups 1; and a second management component 23 configured to control the rotation speed of the fan modules based on all the first temperature information and all the second temperature information, as shown in Figure 4

[0086] ​In this embodiment, the second management component 23 can be a BMC (Baseboard Management Controller), the second management component 23, the first management component 21 and the second programmable device 22 are arranged on the same management board, the fan modules can be arranged on the same fan board, each fan module can be independently controlled, each switch chip SW can correspond to a first temperature acquisition module, or a first temperature acquisition module can be arranged on each switch board, each computing group 1 can correspond to a second temperature acquisition module, or a first temperature acquisition module can be arranged on each computing mainboard. The first temperature acquisition module periodically acquires temperature information of the switch chip SW, and the second temperature acquisition module periodically acquires temperature information of the computing group 1. The second management component 23 collects temperature data from all first and second temperature acquisition modules, analyzes all temperature information, and adjusts the speed of the fan module according to a preset strategy. For example, if the temperature of a certain switch chip SW or computing group 1 exceeds a threshold value, the second management component 23 will increase the speed of the corresponding fan module to enhance the cooling effect.

[0087] This embodiment can dynamically adjust the cooling system according to the real-time thermal load of the server, thereby improving energy efficiency and prolonging the service life of the hardware.

[0088] In an exemplary embodiment, with reference to Figure 4 , the server system further comprises an interface indication board, the interface indication board comprising a plurality of status indication lights; the second programmable device 22 is further configured to control the plurality of status indication lights to perform indication operations according to respective corresponding current state information.

[0089] In this embodiment, the IO (Input / Output) board (i.e. the interface indication board in this embodiment) provides an interface for the multi-host server system to the outside, including but not limited to 1 Power button, 1 UID (Unique Identifier) button, 4 LED (Light Emitting Diode) lights for indicating system status (system status indication light, BMC status indication light, power status indication light, and fan status indication light respectively), 2 network RJ45 (Registered Jack-45) interfaces, 1 debug RJ45 interface, and 1 USB (Universal Serial Bus) interface.

[0090] The Power button is used to start or shut down the server system, and the UID button is used to identify the server for quick positioning among multiple servers in a rack. The LED lights include system status indicator lights, BMC status indicator lights, power status indicator lights, and fan status indicator lights for indicating the status of different components. The network RJ45 interface is used to connect to the network, which can include a management network interface and a data network interface, and the Debug RJ45 interface is used for diagnostic and debugging purposes. The USB interface is used to connect external devices such as a keyboard, mouse, or USB storage device. Specifically, the second programmable device 22 controls the on-off and flashing of the LED lights according to the status information of each part of the system to indicate different system states. The second programmable device 22 obtains status information from each component of the server system, such as the power module, fan module, BMC, etc.

[0091] Figure 4 In the server system, there is one management board, two switching boards, one interface indication board, and one fan board, each switching board has 16 external QSFP-DD interfaces, a total of 32 QSFP-DD interfaces, the management board includes a first management component 21 (mCPU), a second management component 23 (BMC), and a second programmable device 22 (CPLD-B), the switching board is composed of a switching chip SW, a first programmable device 31 (CPLD-S), 2 MCIO (Mini Cool Edge IO) interfaces, and 16 QSFP-DD connectors, the switching chip SW is responsible for interconnecting and data transmission of external computing devices 11, the CPLD-S is dedicated to QSFP-DD optical module configuration, and the MCIO module is reserved for data transmission of the two switching boards. The mCPU provides management for the switching chip SW; the BMC is used for out-of-band management, the CPLD-B is used for power-on and power-off timing control, logic judgment, and light indication, etc. in the whole machine. In combination with Figure 3The management board transmits management signals, including UART (Universal Asynchronous Receiver / Transmitter) signals, I2C signals, and PERST signals, to the switching board through a Manage Connector. The source of the UART signals is the BMC and the mCPU on the management board, which are connected to the switching chip SW for debugging and information reading of the chip. The I2C signals include one I2C signal from the mCPU and two I2C signals from the BMC, and the three groups of I2C signals are connected to the CPLD for communication. In addition, one group of I2C signals from the BMC is connected to the switching chip SW for management and communication, and another group of I2C signals is used for ID identification. This group of I2C signals is connected to the switching chip SW, and 16 groups of signals are connected to 16 QSFP-DD optical modules. In addition, the management board provides power for the switching board through the power connector, providing 3.3V and 12V power for the switching board, and controlling the power-on sequence of other components through the CPLD-S. The server system is arranged in a server case.

[0092] In a second aspect, the present application also provides a data center comprising a plurality of server systems according to any one of the preceding server systems.

[0093] The data center in this embodiment includes at least two cases, each case having one management board, two switching boards, one IO board, and one fan board. The entire architecture has four switching boards, each with 16 external QSFP-DD interfaces, for a total of 64 QSFP-DD interfaces. The switching chip is selected, and the GPUs belonging to different mainboards are connected to the same switching chip using optical modules. The interconnection topology of the system allows each host to read the information of all GPUs.

[0094] In a third aspect, the present application also provides a server communication method applied to the server system according to any one of the preceding server systems. The server communication method comprises: identifying each computing device connected to the server system through the switching chip in the server system; obtaining all computing devices identified by the switching chip through the first management component in the server system, assigning independent identifiers to all computing devices, and configuring the routing table of all optical ports on the switching chip based on the independent identifiers; and after receiving data sent by the first computing device through any optical port of the switching chip in the server system, transmitting the data to the second computing device through the target optical port according to the routing table.

[0095] In an exemplary embodiment, the server system further comprises a plurality of hosts connected to at least two computing groups corresponding to the hosts, and the server communication method further comprises:

[0096] The host is configured to determine a first target computing device among all the computing devices, determine a second target computing device among all the computing devices connected to itself, access the first target computing device by using the second target computing device and a switch chip connected to the second target computing device.

[0097] In an example embodiment, the server communication method further comprises:

[0098] The first management component acquires all the computing devices identified by the switch chip, allocates an address space to all the computing devices, allocates an independent identifier to all the computing devices in the address space, and configures a routing table of all the optical ports on the switch chip based on the independent identifier, so that the switch chip forwards data according to the routing table.

[0099] In an example embodiment, the server communication method further comprises:

[0100] The first management component performs a reset operation on the n switch chips.

[0101] In an example embodiment, the server communication method further comprises:

[0102] The first management component identifies transmission information of each optical port, and adjusts the interface specification and / or interface protocol of each optical port based on the transmission information; the transmission information includes the device type of the computing device connected to the optical port and the data transmission rate of the optical port.

[0103] In an example embodiment, the server system further comprises a switch board, the switch board is provided with at least one switch chip, and the switch board is further provided with a plurality of optical modules corresponding to a plurality of optical ports on the switch chip, and the computing device is connected to the optical port through the optical module.

[0104] In an example embodiment, the server system further comprises a first programmable device, and the first programmable device is arranged on the switch board; the server communication method further comprises: performing a reset operation on the optical module arranged on the switch board where the first programmable device is located by the first programmable device, so that the switch chip arranged on the switch board where the first programmable device is located identifies the computing device connected to the optical module.

[0105] In an example embodiment, the server communication method further comprises:

[0106] The first programmable device performs a reset operation on each optical module connected to the switch chip on the switch board where the first programmable device is located by powering on the switch chip on the switch board where the first programmable device is located, and / or receiving a reset signal sent by the first management component.

[0107] In an example embodiment, the server communication method further comprises:

[0108] outputting, by the first programmable device, a mode control signal; the mode control signal comprises a first control signal corresponding to a host software control mode and a second control signal corresponding to a module hardware control mode; adjusting, by the optical module, the current power mode to a low power mode in response to the first control signal and adjusting the current power mode to a high power mode in response to the second control signal.

[0109] In an example embodiment, the server communication method further comprises:

[0110] adjusting, by the optical module, the current power mode to the high power mode in response to receiving the host switching signal while in the low power mode.

[0111] In an example embodiment, the server communication method further comprises:

[0112] performing, by the first programmable device, a corresponding protection operation in response to an alarm signal output by the optical module.

[0113] In an example embodiment, the server system further comprises a second programmable device, and the server communication method further comprises:

[0114] controlling, by the second programmable device, the first management component to power on in response to receiving a power-on signal, and outputting an allow power-on signal, and outputting a trigger signal to the first management component in response to receiving a complete power-on signal; controlling, by the first programmable device, the exchange board on which the first programmable device is located to power on in response to receiving the allow power-on signal, and outputting the complete power-on signal when the exchange board on which the first programmable device is located powers on completely; and obtaining, by the first management component, all computing devices recognized by the exchange chip in response to receiving the trigger signal, assigning independent identifiers to all computing devices, and configuring a routing table of all optical ports on the exchange chip based on the independent identifiers, so that the exchange chip forwards data according to the routing table.

[0115] In an example embodiment, the server communication method further comprises:

[0116] identifying, by the second programmable device, in-place information of each exchange board, and outputting a logical judgment instruction based on the in-place information; and performing, by the first management component, a corresponding management operation in response to the logical judgment instruction.

[0117] In an example embodiment, the server system further comprises an interface indication board comprising a plurality of state indication lights; and the server communication method further comprises: controlling, by the second programmable device, the plurality of state indication lights to perform indication operations according to respective corresponding current state information.

[0118] In an example embodiment, the server system further comprises: a second management component, a plurality of fan modules, a plurality of first temperature acquisition modules, a plurality of second temperature acquisition modules; and the server communication method further comprises: acquiring, by the first temperature acquisition modules, first temperature information of the corresponding switch chips; acquiring, by the second temperature acquisition modules, second temperature information of the corresponding computing groups; and regulating, by the second management component, the rotating speeds of the fan modules based on all the first temperature information and all the second temperature information.

[0119] In a fourth aspect, the present application further provides a server communication system applied to the server system of any one of the above aspects, comprising: an identification module configured to identify, by a switch chip in the server system, each computing device connected thereto; an establishment module configured to obtain, by a first management component in the server system, all the computing devices identified by the switch chip, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the switch chip based on the independent identifiers; and a routing module configured to, after receiving data sent by a first computing device through any optical port of the switch chip in the server system, transmit the data to a second computing device through a target optical port according to the routing table.

[0120] In an example embodiment, the server system further comprises a plurality of hosts, each host being connected to at least two computing groups, and the server communication system is further configured to: determine a first target computing device among all the computing devices, determine a second target computing device among all the computing devices connected to the host, and access the first target computing device by using the second target computing device and a switch chip connected to the second target computing device.

[0121] In an example embodiment, the server communication system is further configured to: obtain, by the first management component, all the computing devices identified by the switch chip, assign an address space to all the computing devices, assign independent identifiers to all the computing devices within the address space, and configure a routing table of all the optical ports on the switch chip based on the independent identifiers, so that the switch chip forwards data according to the routing table.

[0122] In an example embodiment, the server communication system is further configured to: perform, by the first management component, a reset operation on the n switch chips.

[0123] In an example embodiment, the server communication system is further configured to: identify, by the first management component, transmission information of each optical port, and adjust an interface specification and / or an interface protocol of each optical port based on the transmission information; and the transmission information comprises a device type of a computing device connected to the optical port and a data transmission rate of the optical port.

[0124] In an example embodiment, the server system further comprises a switch board, the switch board is provided with at least one switch chip, and the switch board is further provided with a plurality of optical modules connected with a plurality of optical ports on the switch chip correspondingly, and the computing device is connected with the optical port through the optical module.

[0125] In an example embodiment, the server system further comprises a first programmable device, the first programmable device is arranged on the switch board; and the server communication system is further configured to: reset the optical module arranged on the switch board where the server communication system is located through the first programmable device, so that the switch chip arranged on the switch board where the server communication system is located identifies the computing device connected with the optical module.

[0126] In an example embodiment, the server communication system is further configured to: power on the switch chip on the switch board where the server communication system is located through the first programmable device, reset each optical module connected with the switch chip on the switch board where the server communication system is located, and / or reset each optical module connected with the switch chip on the switch board where the server communication system is located in response to receiving the reset signal sent by the first management component.

[0127] In an example embodiment, the server communication system is further configured to: output a mode control signal through the first programmable device; the mode control signal comprises a first control signal corresponding to a host software control mode and a second control signal corresponding to a module hardware control mode; and the optical module is configured to adjust the current power mode to a low power mode in response to the first control signal and adjust the current power mode to a high power mode in response to the second control signal.

[0128] In an example embodiment, the server communication system is further configured to: in response to receiving a host switching signal while the optical module is in the low power mode, adjust the current power mode to the high power mode through the optical module.

[0129] In an example embodiment, the server communication system is further configured to: perform a corresponding protection operation in response to an alarm signal output by the optical module through the first programmable device.

[0130] In an example embodiment, the server system further comprises a second programmable device, and the server communication system is further configured to: control the first management component to be powered on and output an allowed power-on signal in response to receiving a power-on signal through the second programmable device, output a trigger signal to the first management component after receiving a power-on completion signal; control the switch board where the server communication system is located to be powered on through the first programmable device after receiving the allowed power-on signal, and output the power-on completion signal after the power-on of the switch board where the server communication system is located is completed; and in response to receiving the trigger signal, the first management component is configured to acquire all the computing devices identified by the switch chip, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the switch chip based on the independent identifiers, so that the switch chip forwards data according to the routing table.

[0131] In an example embodiment, the server communication system is further configured to: identify, by the second programmable device, the in-place information of each switching board, output a logical decision instruction according to the in-place information, and perform a corresponding management operation by the first management component in response to the logical decision instruction.

[0132] In an example embodiment, the server system further comprises an interface indication board, the interface indication board comprising a plurality of state indication lamps, and the server communication system is further configured to: control, by the second programmable device, the plurality of state indication lamps to perform an indication operation according to the respective corresponding current state information.

[0133] In an example embodiment, the server system further comprises: a second management component, a plurality of fan modules, a plurality of first temperature acquisition modules, and a plurality of second temperature acquisition modules, and the server communication system is further configured to: acquire, by the first temperature acquisition modules, first temperature information of the corresponding switching chips, acquire, by the second temperature acquisition modules, second temperature information of the corresponding computing groups, and control, by the second management component, the rotating speeds of the fan modules based on all the first temperature information and all the second temperature information.

[0134] In a fifth aspect, the present application provides a computer program product, comprising computer programs / instructions, which, when executed by a processor, implement the above-mentioned server communication method.

[0135] For the computer program product provided by the present application, please refer to the above-mentioned embodiments, and the present application will not be repeated here.

[0136] The computer program product provided by the present application has the same beneficial effects as the above-mentioned server communication method.

[0137] In a sixth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the above-mentioned server communication method.

[0138] The computer readable storage medium can include: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0139] For the computer readable storage medium provided by the present application, please refer to the above-mentioned embodiments, and the present application will not be repeated here.

[0140] The computer readable storage medium provided by the present application has the same beneficial effects as the above-mentioned server communication method.

[0141] It is also noted that, in this disclosure, relational terms such as first and second, and the like, can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0142] The above description of disclosed embodiments provides enabling concepts for practicing or using the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A server system, characterized by The server system comprises: a plurality of computing groups, each of which comprises a plurality of computing devices, any two of which are interconnected through a first communication bus; n exchange chips, which are configured to identify each of the computing devices connected thereto, a plurality of optical ports on the ith exchange chip corresponding to the ith computing device in the plurality of computing groups, i = 1, 2, …, n; the exchange chips and the computing devices are connected through a second communication bus; each of the computing groups comprises n computing devices; a first management component independent of the exchange chips, which is connected to a first transmission interface on each of the n exchange chips through n third communication buses, and is configured to acquire all the computing devices identified by the exchange chips, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the exchange chips based on the independent identifiers, so that the exchange chips forward data according to the routing table; the routing table comprises a correspondence between each of the optical ports and the independent identifier of the computing device connected thereto; the transmission rate of the third communication bus is less than that of the first communication bus and that of the second communication bus; the server system further comprises a plurality of hosts, each of which is connected to at least two computing groups, and is configured to determine a first target computing device among all the computing devices, determine a second target computing device among all the computing devices connected thereto, and access the first target computing device through the second target computing device and the exchange chip connected to the second target computing device.

2. The server system of claim 1, wherein, The first management component is configured to acquire all the computing devices identified by the exchange chips, assign an address space to all the computing devices, assign independent identifiers to all the computing devices within the address space, and configure a routing table of all the optical ports on the exchange chips based on the independent identifiers, so that the exchange chips forward data according to the routing table.

3. The server system of claim 1, wherein, The first management component is further configured to control the n exchange chips to perform a reset operation.

4. The server system of claim 1, wherein, The first management component is further configured to identify transmission information of each of the optical ports, and adjust the interface specification and / or interface protocol of each of the optical ports based on the transmission information; the transmission information comprises the device type of the computing device connected to the optical port and the data transmission rate of the optical port.

5. The server system of claim 1, wherein, The server system further comprises a switching board, which is provided with at least one of the exchange chips, and is further provided with a plurality of optical modules corresponding to a plurality of the optical ports on the exchange chip, and the computing devices are connected to the optical ports through the optical modules.

6. The server system of claim 5, wherein, The server system further comprises a first programmable device, which is provided on the switching board, and is configured to perform a reset operation on the optical modules provided on the switching board where the first programmable device is located, so that the exchange chip provided on the switching board where the first programmable device is located identifies the computing devices connected to the optical modules.

7. The server system of claim 6, wherein, The first programmable device is configured to reset each optical module connected to the switching chip on the switching board where the first programmable device is located when the switching chip on the switching board where the first programmable device is located is powered on, and / or reset each optical module connected to the switching chip on the switching board where the first programmable device is located when receiving a reset signal sent by the first management component.

8. The server system of claim 6, wherein, The first programmable device is further configured to output a mode control signal, and the mode control signal includes a first control signal corresponding to a host software control mode and a second control signal corresponding to a module hardware control mode. The optical module is configured to adjust the current power mode to a low power mode in response to the first control signal, and adjust the current power mode to a high power mode in response to the second control signal.

9. The server system of claim 8, wherein, The optical module is further configured to adjust the current power mode to the high power mode in response to receiving a host switching signal when in the low power mode.

10. The server system of claim 6, wherein, The first programmable device is further configured to perform a corresponding protection operation in response to an alarm signal output by the optical module.

11. The server system of claim 6, wherein, The server system further comprises: A second programmable device configured to control the first management component to be powered on when receiving a power-on signal, and output an enable power-on signal, and output a trigger signal to the first management component when receiving a power-on completion signal. The first programmable device is further configured to control the switching board where the first programmable device is located to be powered on after receiving the enable power-on signal, and output the power-on completion signal when the switching board where the first programmable device is located is powered on. The first management component is further configured to obtain all the computing devices identified by the switching chip in response to receiving the trigger signal, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the switching chip based on the independent identifiers, so that the switching chip forwards data according to the routing table.

12. The server system of claim 11, wherein, The second programmable device is further configured to identify in-place information of each switching board, and output a logical judgment instruction according to the in-place information. The first management component is further configured to perform a corresponding management operation in response to the logical judgment instruction.

13. The server system of claim 11, wherein, The server system further comprises an interface indication board, and the interface indication board comprises a plurality of state indication lamps. The second programmable device is further configured to control the plurality of state indication lamps to indicate the respective corresponding current state information.

14. The server system of any of claims 1-13, wherein, The server system further comprises: A plurality of fan modules; A plurality of first temperature acquisition modules configured to acquire first temperature information of the corresponding switching chip; A plurality of second temperature acquisition modules configured to acquire second temperature information of the corresponding computing group; A second management component configured to regulate the rotation speed of the fan modules based on all the first temperature information and all the second temperature information.

15. A data center, characterized by, The server system comprises a plurality of server systems according to any one of claims 1-14.

16. A server communication method, comprising: The server communication method is applied to the server system according to any one of claims 1-14, and the server communication method comprises: Identifying each computing device connected to the switching chip in the server system; obtain all the computing devices identified by the switch chip in the server system through a first management component in the server system, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the switch chip based on the independent identifiers; after receiving data sent by a first computing device through any optical port of the switch chip in the server system, transmit the data to a second computing device through a target optical port according to the routing table.

17. A server communication system, characterized by The server communication system is applied to the server system as claimed in any one of claims 1-14, and the server communication system comprises: an identification module, configured to identify each computing device connected to the server system through a switch chip in the server system; a configuration module, configured to obtain all the computing devices identified by the switch chip in the server system through a first management component in the server system, assign independent identifiers to all the computing devices, and configure a routing table of all the optical ports on the switch chip based on the independent identifiers; a routing module, configured to, after receiving data sent by a first computing device through any optical port of the switch chip in the server system, transmit the data to a second computing device through a target optical port according to the routing table.

18. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to realize the server communication method as claimed in claim 16.

19. A computer-readable storage medium, characterized in that, The computer program / instruction is executed by the processor to realize the server communication method as claimed in claim 16.

Images