Network card connection circuit and server

Abstract

The invention discloses a network card connection circuit and a server, and relates to the technical field of network card connection, the network card connection circuit comprises a first riser card and a second riser card, the first riser card is used for receiving a first output signal sent by a first processor and separating a communication signal and a power supply voltage in the first output signal, and the second riser card is used for receiving a second output signal sent by a second processor; the second adapter card is arranged on the second processor and is used for receiving and integrating the first communication signal and the first power supply voltage to obtain a second output signal, and / or receiving a third output signal sent by the second processor and outputting the third output signal to the second processor; and the second output signal or the third output signal is sent through the target network card, so that when the resource of the first processor is insufficient, the target network card corresponding to the second processor can be multiplexed through the first riser card and the second riser card, the technical problem of communication transmission delay of the OCP network card when the CPU resource is insufficient can be solved, and the user experience is improved. And the technical effect of cross-CPU resource allocation of the target network card is achieved.

Description

Technical Field

[0001] This application relates to the technical field of network card connection, and particularly to a network card connection circuit and a server. Background Art

[0002] Under the promotion and organization of the OCP (Open Compute Project, a technical association alliance for servers), more and more enterprises and institutions have an increasingly strong demand for OCP NIC (Network Interface Card) network cards in cloud computing server systems. This requires server solution providers and complete machine manufacturers to perfectly adapt to the compatibility of OCP NIC network cards.

[0003] In related technologies, OCP network cards and CPUs (Central Processing Units) are arranged in a one-to-one correspondence, that is, each OCP network card is connected to the corresponding CPU through the corresponding link and adapter card, and each CPU performs information transmission through the connected OCP network card. However, when the CPU resources are overloaded, it is easy to cause communication transmission delays of the OCP network card, reducing the user experience. Summary of the Invention

[0004] This application provides a network card connection circuit and a server to at least solve the problem of communication transmission delay of the OCP network card when the CPU resources are insufficient in related technologies.

[0005] This application provides a network card connection circuit, including: a first adapter card, the input end of the first adapter card is connected to the network card interface of the first processor, and is used to receive the first output signal sent by the first processor, and separate the communication signal and the power supply voltage in the first output signal to obtain a first communication signal and a first power supply voltage; a second adapter card disposed on the second processor, the first input end of the second adapter card is connected to the output end of the first adapter card, the second input end of the second adapter card is connected to the data interface of the second processor, and the output end of the second adapter card is connected to the target network card, and is used to receive and integrate the first communication signal and the first power supply voltage to obtain a second output signal, and / or receive the third output signal sent by the second processor, and send the second output signal or the third output signal through the target network card.

[0006] This application also provides a server, including: a plurality of processors, where the plurality of processors include a second processor and a first processor; the above-mentioned network card connection circuit.

[0007] With this application, since the network card interface of the first processor can be connected to the target network card corresponding to the second processor through the first adapter card and the second adapter card, not only can it be used for the communication transmission of the second processor through the target network card, but also for the communication transmission requirements of the first processor. At this time, when the resources of the first processor are insufficient, the target network card corresponding to the second processor can be multiplexed through the first adapter card and the second adapter card, so as to perform communication transmission based on the target network card. Therefore, the technical problem of communication transmission delay of the OCP network card when the CPU resources are insufficient can be solved, and the technical effect of dynamically allocating the target network card across CPU resources and improving the load balance can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0009] Figure 1 Schematic diagram of a network card connection circuit provided by an embodiment of the present application; Figure 2 Schematic diagram of a network card connection circuit provided by a specific embodiment of the present application; Figure 3 Schematic diagram of a network card connection circuit provided by another specific embodiment of the present application; Figure 4 Schematic diagram of the connection of a server provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0010] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the 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. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the protection scope of the present application.

[0011] It should be noted that in the description of the present application, the terms "including", "comprising" or any other variation 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 explicitly 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 not to describe a specific order or sequence.

[0012] To enable those skilled in the art of the present technology to better understand the solution of the present application, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0013] Under the promotion and organization of the OCP Association, the demand for OCP NIC network cards in cloud computing server systems by more and more enterprises and institutions is increasing day by day. This requires server solution providers and complete machine manufacturers to perfectly adapt to the compatibility of OCP NIC network cards.

[0014] In the related art, a connection solution between a server and an OCP network card, or a connection solution in which a CPU and an OCP network card are bound one by one is adopted. Specifically as follows: In the connection solution between a server and an OCP network card, an adaptation design of an OCP NIC network card with a fixed upstream PCIe (Peripheral Component Interconnect Express, high-speed serial computer expansion bus standard) link and bandwidth is adopted on the motherboard of the server. By placing a 4C+ / 4C (Connector) connector of the fixed OCP NIC network card on the motherboard, the upstream PCIe signal of the connector is connected to the fixed CPU or bridge chip to provide a fixed PCIe link and bandwidth for the OCP NIC network card. However, although this standard design of providing a fixed PCIe link to the OCP NIC card based on the motherboard is effective, it can only be compatible with and adapt to a limited type of OCP NIC network cards, and the types of compatible OCP NIC network cards are limited, which cannot meet the compatibility requirements of diverse OCP NIC network cards. Taking the use of the server OCP NIC network card by a customer as an example, when the operation and maintenance personnel are not fully familiar with the internal design of the server in the computer room and the types of peripheral OCP NIC network cards, when inserting a type A OCP NIC network card into a server that only supports type B OCP NIC network cards, at this time, the server will have a problem of incompatibility with the externally inserted OCP NIC network card, resulting in server downtime and the OCP NIC network card not working properly, which is not conducive to the user experience.

[0015] In the connection solution in which a CPU and an OCP network card are bound one by one, the OCP network card is connected to a single CPU (such as CPU1) through a single link and an adapter card. When the server is only configured with a single CPU or the resources of CPU1 are overloaded, the latency increases, reducing the user experience.

[0016] To solve at least one of the above technical problems, the present application proposes a network card connection circuit. The network card interface of the first processor (such as CPU1) can be connected to the target network card corresponding to the second processor (such as CPU0) through the first transfer circuit and the second transfer circuit. Thus, not only can it be used for communication transmission for CPU0 through the target network card, but also for the communication transmission requirements of CPU1. At this time, when the resources of CPU1 are insufficient, the target network card at CPU0 can be multiplexed through the first transfer card and the second transfer card, so as to perform communication transmission based on the target network card. Therefore, the technical problem of communication transmission delay of the OCP network card when the CPU resources are insufficient can be solved, the technical effect of dynamically allocating the target network card across CPU resources and improving the load balancing can be achieved, and the problem of hardware-level resource exclusivity can be solved.

[0017] Taking the target network card as an OCP network card as an example, the network card connection circuit of the embodiment of the present application will be described in detail below.

[0018] Figure 1 It is a schematic diagram of a network card connection circuit provided by an embodiment of the present application.

[0019] As Figure 1 shown, the network card connection circuit 100 of the embodiment of the present application includes: a first transfer card 10 and a second transfer card 20 provided on the second processor.

[0020] Among them, the input end of the first transfer card 10 is respectively connected to the network card interface of the first processor, and is used to receive the first output signal sent by the first processor, and separate the communication signal and the power supply voltage in the first output signal to obtain the first communication signal and the first power supply voltage. The first input end of the second transfer card 20 is respectively connected to the output end of the first transfer card 10, the second input end of the second transfer card 20 is connected to the data interface of the second processor, and the output end of the second transfer card 20 is connected to the target network card, and is used to receive the first communication signal and the first power supply voltage, and integrate the first communication signal and the first power supply voltage to obtain a second output signal, and / or receive the third output signal sent by the second processor, so as to send the second output signal or the third output signal through the target network card.

[0021] Specifically, the first adapter card 10 is arranged in the area where the corresponding first processor is located. By connecting to the OCP interface of the first processor (i.e., the network card interface of the first processor), it receives the first output signal sent by the first processor, and separates the PCIe signal (i.e., the communication signal) and the 12V power supply voltage from the first output signal. It processes the PCIe signal, such as signal amplification, filtering, etc., to obtain the processed PCIe signal (i.e., the first communication signal), and processes the 12V power supply voltage, such as filtering, etc., to obtain the processed 12V power supply (i.e., the second power supply voltage). By separating the PCIe signal and the 12V power supply of the first output signal into independent lines and transmitting them to the second adapter card 20 respectively, the first adapter card 10 can avoid power noise coupling.

[0022] The second processor can be selected according to the processor load in the server, and the processor with a smaller load is selected as the second processor, which can be specifically selected according to the actual situation. The second adapter card 20 is arranged in the area where the second processor is located and is connected to the motherboard through the Slimline x8 interface. On the one hand, it can integrate the first communication signal and the first power supply voltage to obtain the second output signal and output it through the target network card. That is to say, it completes the data transmission task of the first processor through the target network card set at the second processor. On the other hand, it can receive the data sending task (i.e., the third output signal) sent by the second processor based on the Slimline x8 interface and output it through the target network card. It should be noted that the second output signal needs to meet the PCIe signal requirements of the second processor. For example, the second power supply voltage is converted to 3.3V / 5V to adapt to the PCIe power supply requirements of the second processor, and then it is sent out based on the OCP interface of the second processor, that is, the output end of the second adapter card 20.

[0023] In addition, the transmission of the second output signal or the third output signal through the target network card can be achieved through the line on / off control inside the second adapter card 20, or it can also be achieved through motherboard control, and there is no specific limitation.

[0024] Through the circuit connection between the first adapter card 10 and the second adapter card 20, this embodiment can complete the communication transmission of the first processor based on the target network card set in the second processor. At the same time, based on the transmission strategy of separating communication and power supply, interference is reduced and the transmission effect is ensured. The second adapter card 20 can, based on communication switching, realize the signal transmission from the OCP interface to the first processor and the second processor. The PCIe channel is seamlessly switched from the first processor to the second processor through the hardware of the adapter card without software intervention, realizing the dynamic allocation of OCP network card cross-processor resources based on dual adapter cards, solving the load balancing problem in the related technology when the processor resources are insufficient, and improving the application flexibility of the target network card. This circuit can also be applied to the dynamic resource allocation of PCIe peripherals such as GPUs (Graphics Processing Units) and NVMe (Non-Volatile Memory Express) devices.

[0025] Further, the network card connection circuit 100 may include multiple first adapter cards 10. At this time, each first adapter card 10 corresponds to a first processor, and is used to separate communication and power supply for the first output signal of the corresponding first processor, and the output ends are respectively connected to the corresponding ports of the first input end of the second adapter card 20. The ports of the first input end of the second adapter card 20 identify the inputs of different processors, so as to perform signal integration and output, thereby realizing the dynamic allocation of resources between multiple processors.

[0026] Combined Figure 2 As shown, in some embodiments of the present application, the first adapter card 10 includes: a signal relay module 11, the input end of the signal relay module 11 is connected to the communication port of the input end of the first adapter card 10, and the output end of the signal relay module 11 is adapted to be connected to the communication port of the first input end of the second adapter card 20, and is used to receive the communication signal in the first output signal through the communication port of the input end of the first adapter card, and generate a first communication signal based on the communication signal in the first output signal; a power supply separation module 12, the input end of the power supply separation module 12 is connected to the power supply port of the input end of the first adapter card 10, and the output end of the power supply separation module 12 is adapted to be connected to the power supply port of the first input end of the second adapter card 20, and is used to receive the power supply voltage in the first output signal through the power supply port of the first adapter card 10, and generate a first power supply voltage based on the power supply voltage in the first output signal.

[0027] Specifically, the signal relay module 11 is a signal relay node from the first processor to the target network card, which is used to receive, amplify and forward signals, so as to achieve the functions of expanding the signal coverage, enhancing the transmission reliability and compensating for signal attenuation, so as to solve the signal attenuation problem in long-distance transmission and ensure the stability and reliability of long-distance communication. The signal relay module 11 processes the communication signal separated from the first output signal to obtain the first communication signal for long-distance transmission, and sends the second output signal to the communication port of the first input end of the second adapter card 20 through the communication line. The power supply separation module 12 processes the power supply voltage separated from the first output signal to obtain the first power supply voltage for transmission, and transmits it to the power supply port of the first input end of the second adapter card 20 through the power line. Thus, by separating the 12V power supply and the PCIe signal into independent lines, power supply noise coupling is avoided. In some embodiments of the present application, the signal relay module 11 includes: a differential signal amplification unit, the input end of the differential signal amplification unit is connected to the communication port of the input end of the first adapter card 10, and is used to amplify the communication signal in the first output signal to generate the first communication signal.

[0028] Specifically, the differential signal is a signal transmission method, which transmits signals through two signal lines, one of which is the positive signal and the other is the negative signal, and the two have opposite phases. This signal transmission method has the advantages of strong anti-interference ability and good signal integrity, and is suitable for high-speed signal transmission. The differential signal amplification unit includes at least one differential signal amplifier, which is used to amplify the received communication signal to generate the first communication signal, so as to enhance the signal quality and enable it to be transmitted more stably and reliably over a longer distance.

[0029] For example, the gain value of the differential signal amplification unit can be set to 8dB@5GHz, that is, at a frequency of 5GHz, the gain of the signal is 8 decibels. Gain is an index to measure the degree of signal amplification. A higher gain can make the signal be amplified more effectively during transmission, thereby enhancing the strength and quality of the signal. An 8dB gain is a relatively reasonable value in high-speed signal transmission, which can effectively improve the transmission performance of the signal without introducing too much noise and distortion. Specifically, it can be set according to the actual situation. The gain can be adjusted by changing the resistance value of the external resistor, which can accurately control the degree of signal amplification, so as to achieve the best signal transmission effect, enabling users to flexibly adjust the gain size according to the actual application requirements and system characteristics, providing greater flexibility and adaptability for system design, and being able to better meet the signal transmission requirements in different scenarios.

[0030] This embodiment uses the differential signal amplification unit to optimize the signal transmission performance, ensure the accurate and fast transmission of data, and thus improve the communication transmission efficiency and reliability. In addition, the signal relay module 11 may further include an adaptive equalization compensation unit to support the adaptive equalization compensation function, including CTLE (Continue Time Linear Equalizer) and DFE (Decision Feedback Equalizer). CTLE can compensate for the high-frequency attenuation caused by the transmission line characteristics during signal transmission and restore the high-frequency components of the signal; DFE can adjust the current signal according to the historical information of the received signal to further optimize the signal quality. The adaptive equalization compensation function can automatically adjust the equalization parameters according to the actual situation of signal transmission, so as to better adapt to different transmission environments and improve the reliability and stability of signal transmission.

[0031] In some embodiments of the present application, the signal relay module 11 further includes: an electrostatic protection unit. One end of the electrostatic protection unit is respectively connected to the communication port of the input end of the first adapter card and the input end of the differential signal amplification unit, and the other end of the electrostatic protection unit is grounded.

[0032] Specifically, electrostatic discharge refers to the phenomenon of instantaneous high-voltage discharge due to electrostatic accumulation. This kind of discharge may damage electronic devices, especially at the input end, because the input end is directly connected to the external signal source or user operation interface and is more vulnerable to external static electricity interference and impact. To prevent the damage of electrostatic discharge to the internal circuit and ensure the stable operation and reliability of the circuit. This embodiment sets an electrostatic protection unit. Specifically, a bidirectional TVS (Transient Voltage Suppressor) diode array can be deployed at the input end of the signal relay module 11, and a protection level of ±15 kV can be achieved, providing electrostatic protection ability for the electronic device, effectively reducing the risk of electrostatic damage to the circuit, and ensuring the normal operation and reliability of the circuit.

[0033] In some embodiments of the present application, the signal relay module 11 further includes: an impedance matching unit. One end of the impedance matching unit is connected to the output end of the differential signal amplification unit, and the other end of the impedance matching unit serves as the output end of the signal relay module.

[0034] Specifically, when a signal propagates on a communication line, if the characteristic impedance of the communication line does not match the impedance of the signal source and the load, signal reflection will occur, resulting in signal distortion and transmission loss. Therefore, this embodiment adds an impedance matching unit in the signal relay module 11. By setting an impedance matching network, it is ensured that the first communication signal can be transmitted to the second adapter card 20 efficiently and completely, reducing the reflection and interference of the first communication signal and improving the quality and reliability of signal transmission. For example, the impedance matching unit deploys a 100Ω differential resistor (±1% accuracy) at the output end of the signal relay module 11, and the PCB (Printed Circuit Board) traces strictly maintain a 100Ω differential impedance (4mil line width, 8mil spacing). Among them, deploying a 100Ω differential resistor at the output end of the signal relay module 11 is to achieve impedance matching of differential signals, so that the differential signals (i.e., the first communication signals) can be well matched at the end of the communication line, reduce signal reflection, and ensure signal integrity; the accuracy requirement of the differential resistor is ±1% accuracy, which means that the actual value of the differential resistor should be between 99Ω and 101Ω. A high-precision resistor can more accurately achieve impedance matching, thereby improving the quality and stability of signal transmission. In addition, differential impedance refers to the impedance characteristics of the differential signal transmission line to the signal, which is related to the geometric structure, dielectric material, etc. of the transmission line. Therefore, in PCB design, a specific differential impedance value is achieved through precise trace design. For the 100Ω differential impedance corresponding to this embodiment, factors such as the line width, spacing, and dielectric thickness of the transmission line need to be considered. The specific line width of the PCB trace used is 4mil, and the spacing is 8mil, which can be determined according to the differential impedance calculation formula and the dielectric material characteristics of the PCB to ensure that a 100Ω differential impedance is achieved under the given PCB structure, which can ensure the impedance consistency of differential signals during transmission and reduce reflection and distortion during signal transmission.

[0035] In some embodiments of the present application, the output end of the signal relay module 11 is adapted to be connected to the communication port of the first input end of the second adapter card 20 through a shielded cable.

[0036] That is to say, the first adapter card 10 and the second adapter card 20 are interconnected through a customized cable. The signal relay module 11 transmits PCIe differential signals through a shielded cable, and the additional shielding layer of the shielded cable reduces interference. In some embodiments of the present application, the power supply separation module 12 includes: a filtering unit and a noise isolation unit, which are used to filter and isolate the power supply voltage in the first output signal to generate a first power supply voltage.

[0037] Specifically, the connection order of the filtering unit and the noise isolation unit is not limited. The separated power supply voltage can be filtered first by the filtering unit, and then the filtered power supply voltage can be subjected to noise isolation processing by the noise isolation unit to obtain the first power supply voltage; or the separated power supply voltage can be subjected to noise isolation processing first by the noise isolation unit, and then the power supply voltage after noise isolation processing can be filtered by the filtering unit to obtain the first power supply voltage. Among them, filtering is used to filter out high-frequency noise, and the noise isolation processing may include high-frequency noise suppression and transient overvoltage suppression protection, etc.

[0038] For example, the filtering unit can adopt a π-type filter, including a 10 μF ceramic capacitor, a 10 Ω @ 100 MHz bead, and a 10 μF tantalum capacitor, with an insertion loss of -40 dB @ 100 MHz. Among them, the 10 μF ceramic capacitor is a temperature-compensated ceramic capacitor, which has good temperature stability and high-frequency characteristics, and is mainly used to filter high-frequency noise. The capacitance of 10 μF can provide a low impedance in the high-frequency band, thus effectively bypassing high-frequency noise signals. The 10 μF tantalum capacitor has a high capacitance density and good stability, and is suitable for low-frequency filtering. It can further filter out low-frequency noise in the power supply. When used in combination with the ceramic capacitor, it can achieve a filtering effect in a wider frequency range. The bead is a high-frequency inductance element, and its impedance increases with the increase of frequency. The 10 Ω @ 100 MHz bead is a bead with an impedance of 10 Ω at 100 MHz, which can effectively suppress the passage of high-frequency noise. In the π-type filter, the bead provides a high impedance in the high-frequency band to prevent high-frequency noise from entering the subsequent circuit. Insertion loss refers to the attenuation degree of the signal after passing through the filter. The insertion loss of -40 dB @ 100 MHz means that at a frequency of 100 MHz, the filter can attenuate the noise signal to 1 / 10000 of the original, ensuring that the π-type filter has a strong noise suppression ability at 100 MHz, can effectively reduce the high-frequency noise in the power supply, and improve the purity of the power supply.

[0039] The noise isolation unit can include a high-impedance bead and a transient voltage suppressor. Among them, the high-impedance bead can provide a high impedance in the high-frequency band, thus effectively suppressing the propagation of high-frequency noise, reducing noise coupling and propagation. Through the high-impedance bead, the noise on the power line can be further isolated to prevent noise from spreading from one circuit part to another, thereby improving the anti-interference ability of the entire system. The transient voltage suppressor is used to suppress transient overvoltage (such as lightning strikes, arcs, etc.). It can conduct quickly when overvoltage appears, clamp the overvoltage at a safe level, and protect the subsequent circuit from damage. It can clamp the overvoltage within a safe range in a short time, thereby protecting the circuit from overvoltage damage.

[0040] The filtering unit and the noise isolation unit in this embodiment can effectively reduce power supply noise, improve the quality and stability of the power supply, and thus provide a reliable guarantee for the normal operation of electronic devices. The first power supply voltage output by the power supply separation module 12 is transmitted through the power line, separating the OCP power supply of the first processor from the signal transmission of the second processor, so as to avoid conflicts with the motherboard power supply.

[0041] Furthermore, the power ground of the power supply separation module 12 and the signal ground of the signal relay module 11 are connected at a single point through a 0Ω resistor, achieving effective isolation between the power ground and the signal ground, reducing the interference of power noise on the signal line, and improving the integrity and reliability of the signal. In this embodiment, the 0Ω resistor is essentially a conductor, which plays a role of connecting the power ground and the signal ground in the circuit, ensuring that the two are electrically connected. Additionally, although the resistance value of the 0Ω resistor is close to zero, it still has certain parasitic inductance and parasitic capacitance under high-frequency signals, and these parasitic parameters can isolate high-frequency noise to a certain extent, preventing high-frequency noise from propagating between the power ground and the signal ground through the grounding path. In some hot-pluggable scenarios, the 0Ω resistor can limit the impact of instantaneous current, protecting the circuit from damage. During the circuit design and debugging phase, the 0Ω resistor can be easily removed or replaced to facilitate testing and optimizing the grounding strategy of the circuit. In some embodiments of the present application, the power supply separation module 12 further includes: a power supply protection unit for cutting off the input of the power supply voltage in response to a network card migration request.

[0042] Specifically, network card migration refers to the situation in some multi-processor or distributed systems where the network card may migrate from one processor to another, which may be due to load balancing, failover, or other system management strategies. When a network card migration request is detected, it means that the network card is about to switch from one processor to another, which may cause a sudden change in power demand.

[0043] The power supply separation module 12 of the first adapter card 10 can implement intelligent power distribution management using an electronic fuse. By monitoring the current of the OCP interface (network card interface) of the first processor, it determines whether a network card migration request is received. When a network card migration request is detected, the 12V main power supply is cut off within 10ms, and the 3.3V standby power supply is retained. At the same time, a back-to-back MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) is integrated to prevent the power supply on the second processor side from flowing back to the first processor. By cutting off the input of the power supply voltage, it can ensure that during the network card migration process, power overload or damage is prevented, ensuring the stable operation of the system. At the same time, maintaining the 3.3V standby power supply can prevent the system from completely powering off, so that it can quickly resume operation after the main power supply is restored, reducing the time and complexity of system restart. Additionally, the back-to-back MOSFET is a circuit structure composed of two MOSFETs, which can prevent the power supply on the second processor side from flowing back to the first processor during the power supply switching process. This kind of backflow may cause power short circuit or other faults, so the power supply isolation is achieved through the back-to-back MOSFET.

[0044] In some embodiments of the present application, the second adapter card 20 includes: a signal reshaping module 21, an input end of the signal reshaping module 21 is connected to a communication port of a first input end of the second adapter card 20, and an output end of the signal reshaping module 21 is adapted to be connected to a communication port of an output end of the second adapter card 20, configured to receive a first communication signal and perform signal loss compensation on the first communication signal to generate a second communication signal; a voltage conversion module 22, an input end of the voltage conversion module 22 is connected to a power supply port of a first input end of the second adapter card 20, and an output end of the voltage conversion module 22 is adapted to be connected to a power supply port of an output end of the second adapter card 20, configured to receive a first power supply voltage and perform voltage conversion on the first power supply voltage to generate a second power supply voltage; wherein, the second communication signal and the second power supply voltage are integrated at the output end of the second adapter card 20 to obtain a second output signal.

[0045] Specifically, the signal reshaping module 21 is configured to compensate for the signal attenuation of the first communication signal due to long-distance transmission to obtain the second communication signal. For example, a signal relay chip can be used to perform signal attenuation compensation on the received first communication signal and output the second communication signal. The voltage conversion module 22 is configured to convert the first power supply voltage to a target voltage adapted to the target network card to adapt to the PCIe power supply requirements of the second processor, so as to integrate with the second communication signal at the network card interface of the second adapter card 20 (i.e., the output end of the second adapter card 20) to generate a second output signal, and output the signal through the target network card.

[0046] In this embodiment, the communication signal and the power supply in the first output signal of the first processor are separately isolated and transmitted, and after signal and voltage processing are respectively performed on the second adapter card 20, they are integrated and output. While ensuring the signal quality of the first processor, it is adapted to the target network card to meet the dynamic matching application requirements. In some embodiments of the present application, the signal reshaping module 21 includes: a signal re-driving unit, configured to perform high-frequency gain compensation on the first communication signal based on an initial gain to generate a second communication signal.

[0047] Specifically, the received first communication signal is subjected to high-frequency gain compensation by the signal re-driving unit to offset the attenuation of the high-frequency components during cable transmission. For example, the initial gain is set to 6 dB to compensate for the high-frequency attenuation of the first 8 inches of cable. The initial gain can be set according to the actual situation.

[0048] In this embodiment, the signal re-driving unit can re-drive and condition the received first communication signal to enhance the signal quality and enable it to be transmitted more stably and reliably. In some embodiments of the present application, the signal re-driving unit is further configured to adjust the initial gain based on the signal transmission bit error rate test result, so as to perform high-frequency gain compensation on the first communication signal based on the adjusted initial gain.

[0049] Specifically, the bit error rate (BER) test is a method for evaluating the performance of a digital communication system. It quantifies the signal quality by measuring the ratio of the number of error bits that occur during transmission to the total number of transmitted bits. Specifically, it can be tested through a bit error rate tester (BERT). By generating known test signals and detecting errors in the signals at the receiving end, the signal transmission bit error rate can be calculated. For example, the signal re-driving unit sets the initial gain to 6 dB to compensate for the high-frequency attenuation of the first 8-inch cable, and sets 5 taps to eliminate inter-symbol interference (ISI). According to the BER test result, the gain is dynamically adjusted (-3.5 dB to -6 dB) to further optimize the signal quality.

[0050] This embodiment dynamically adjusts the compensation gain based on the signal transmission bit error rate test result to further improve the transmission quality and reliability of the signal. In some embodiments of the present application, the voltage conversion module 22 includes: a buck unit configured to reduce the first supply voltage to a target voltage to generate a second supply voltage.

[0051] That is to say, in the case where the power supply separation module 12 outputs the first supply voltage through filtering and noise isolation, the buck unit in the voltage conversion module 22 reduces the first supply voltage (12 V) to 3.3 V / 5 V required by the target network card to obtain a second supply voltage for signal integration.

[0052] Combined Figure 3 As shown, in some embodiments of the present application, the second adapter card 20 further includes: a hot-swap button configured to trigger a hot-swap request and send it to the second processor, so that the second processor triggers a hot-swap control process based on the hot-swap request.

[0053] That is to say, the second adapter card 20 supports the hot-swap function. Hot-swap means that when the device is running, hardware components can be safely inserted or removed without affecting the normal operation of the system. For a network card adapter card, the hot-swap function allows users to safely insert or remove the network card without shutting down the system power, thereby improving the availability and maintenance efficiency of the system.

[0054] The second adapter card 20 is provided with a communication interface connected to the communication interface of the second processor, and can communicate with the second processor through the EDID (Extended Display Identification Data) protocol.

[0055] The hot-swap button can be a physical button, which is set on the front panel or side of the second adapter card 20. Users can trigger the hot-swap operation by pressing this button. Specifically, after the user presses the hot-swap button, a hot-swap request is sent to the second processor based on the communication. After receiving the hot-swap request, the second processor executes the following hot-swap control process operations to ensure that the second adapter card realizes the hot-swap function. The specific hot-swap control process is as follows: Cut off the power supply: Cut off the power supply of the network card to ensure that no arc or circuit damage occurs when the network card is inserted or removed; Disable the signal channel: Disable the signal channel of the network card to prevent signal interference or damage; Update the system status: The system updates its internal status to record the insertion or removal operation of the network card; Indicator light prompt: The system uses the indicator light to feedback the status of the hot-swap operation to the user. For example, when the indicator light turns green, it means that the network card can be safely inserted or removed; Insert or remove the network card: The user safely inserts or removes the network card according to the prompt of the indicator light; System recovery: When the network card is inserted, the system will power on again and enable the signal channel to restore the normal operation of the network card.

[0056] This embodiment realizes the hot-swap function based on the hot-swap button, allowing users to safely insert or remove the network card without shutting down the system power. This function ensures that the system or network card will not be damaged during the hot-swap process through a series of security mechanisms (such as power management, signal isolation, and software support). The hot-swap function is of great significance in scenarios such as data centers, servers, and industrial control systems where network cards need to be frequently maintained and replaced, and can improve the availability and maintenance efficiency of the system. Furthermore, the second adapter card 20 also supports the PCIe 4.0 / 5.0 protocol and is connected to other interfaces of the second processor.

[0057] As a specific embodiment of the present application, in combination with Figures 1 - 3 As shown, the hardware deployment of the network card connection circuit 100 is as follows: Insert the first adapter card 10 into the OCP 3.0 interface of the first processor and fix it to the rear window of the server. Connect the output end of the adapter card 1 to the input end of the second adapter card 20 through a customized cable. Insert the second adapter card 20 into the Slimline interface reserved on the motherboard, which is the data interface of the second processor.

[0058] The signal processing flow of the network card connection circuit 100 is as follows: The PCIe signal of the first processor is relayed by the first adapter card 10 and then transmitted to the second adapter card 20 through shielded differential lines. The Re-driver chip of the second adapter card 20 shapes the signal and then sends it to the network card interface of the second processor.

[0059] The power supply management flow of the network card connection circuit 100 is as follows: The first adapter card 10 obtains a 12V power supply from the OCP interface and transmits it to the second adapter card 20 through an independent power supply line. The second adapter card 20 is built-in with a DC-DC (Direct Current - Direct Current) module to convert 12V into 3.3V / 5V required by the target network card.

[0060] In addition, the power supply separation module 12 of the first adapter card 10 uses an eFuse (electronic fuse) to achieve intelligent power distribution management: monitor the current of the OCP interface of the first processor, and when a network card migration request is detected, cut off the 12V main power supply within 10ms, retain the 3.3V standby power supply, and integrate back-to-back MOSFETs to prevent the power supply on the second processor side from flowing back to the first processor.

[0061] The second adapter card 20 is equipped with a Re-driver chip to perform adaptive equalization. Receiver CTLE (Continuous Time Linear Equalization): Initial gain = 6dB, compensate for the high-frequency attenuation of the first 8-inch cable; set 5 taps to eliminate inter-symbol interference (ISI); dynamically adjust according to BER test (-3.5dB to -6dB).

[0062] Thus, in this embodiment, the first adapter card 10 connected to the first processor is used to separate the PCIe signal and the power supply, and the second adapter card 20 connected to the second processor integrates signal reshaping and voltage conversion functions. Through the shielded cable connecting the two adapter cards, the signal and power are transmitted independently.

[0063] The network card connection circuit 100 based on the dynamic allocation of OCP network card across CPU resources with dual adapter cards can achieve at least the following beneficial effects: Load balancing: Flexibly allocate the OCP network card to the idle CPU to improve resource utilization. Specifically, according to the PCIe channel utilization rate, memory bandwidth occupancy rate, and computing load indicators of each CPU, combined with the characteristics of the NUMA (Non-Uniform Memory Access) topology structure, the optimal matching of the OCP network card device and the target CPU (i.e., the second processor) can be achieved. In the typical business scenario of a dual-processor server, cross-CPU resource allocation can improve the overall resource utilization rate and reduce network latency fluctuations at the same time; Compatibility: At the hardware level, it is compatible with the OCP NIC 3.0 standard specification. The specially designed mechanical structure maintains the original server chassis layout. The pin definition of the adapter card is completely compatible with the standard OCP slot, supporting mainstream server architectures without modifying the motherboard design. Cost advantage: Through the signal-power separation architecture, complex high-speed switching chips and supporting clock buffer circuits are eliminated, achieving a reduction in hardware costs.

[0064] In summary, for the network card connection circuit in the embodiment of the present application, the first adapter card receives the first output signal sent by the corresponding first processor, separates the communication signal and the power supply voltage in the first output signal to obtain the first communication signal and the first power supply voltage, and sends them to the first input end of the second processor through the communication line and the power line respectively. The second adapter card integrates the received first communication signal and the first power supply voltage to obtain the second output signal, and emits the second output signal based on the target network card. The second adapter card is also used to receive the third output signal sent by the second processor to send the third output signal through the target network card for the communication transmission of the second processor. That is to say, the network card interface of the first processor can be connected to the target network card corresponding to the second processor through the first transfer circuit and the second transfer circuit. Thus, it can not only be used for the communication transmission of the second processor through the target network card, but also for the communication transmission requirements of the first processor. At this time, when the resources of the first processor are insufficient, the target network card corresponding to the second processor can be reused through the first adapter card and the second adapter card, so as to perform communication transmission based on the target network card. Therefore, the technical problem of communication transmission delay of the OCP network card when the CPU resources are insufficient can be solved, achieving the technical effect of dynamically allocating the target network card across CPU resources and improving the load balancing.

[0065] Corresponding to the above embodiment, the present application also provides a server.

[0066] As Figure 4 shown, the server 1000 in the embodiment of the present application includes: a plurality of processors, where the plurality of processors include a second processor 200 and a first processor 300; the above-mentioned network card connection circuit 100.

[0067] In the server 1000, multiple network card connection circuits 100 can be set according to requirements. Each network card connection circuit 100 can respectively correspond to a second processor 200 and a first processor 300, or can also correspond to a second processor 200 and multiple first processors 300, without specific limitations.

[0068] In some embodiments of the present application, the second processor 200 is determined based on the utilization rate of each processor in the plurality of processors.

[0069] Specifically, the utilization rate of each processor can be determined by the physical link connection conditions. For example, when there are many devices such as hard disk network cards connected to the processor, it is considered that the CPU utilization rate is high; or through commands in the operating system, the utilization rate values of each CPU can be viewed in real time, and then the processor with a lower utilization rate can be selected as the second processor 200.

[0070] Thus, the server 1000 can achieve load balancing, flexibly allocate the OCP network card to the idle CPU, and improve resource utilization. Specifically, according to the PCIe channel utilization rate, memory bandwidth occupancy rate, and computing load index of each CPU, combined with the characteristics of the NUMA (Non-Uniform Memory Access) topology structure, the matching effect between the target network card and the second processor can be achieved.

[0071] In summary, for the server of the embodiment of the present application, the network card sharing between the first processor and the second processor is realized through the network card connection circuit, that is, the network card interface of the first processor is connected to the target network card at the second processor through the network card connection circuit. Thus, not only can the target network card be used for the communication transmission of the second processor, but also for the communication transmission requirements of the first processor. When the resources of the first processor are insufficient, communication transmission can be based on the target network card, thereby solving the technical problem of communication transmission delay of the OCP network card when the CPU resources are insufficient, achieving the technical effect of dynamically allocating the target network card across CPU resources and improving load balancing.

[0072] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method.

[0073] Those skilled in the art can further realize that, combining the units and algorithm steps of each example described in the embodiments disclosed herein, can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0074] The above has introduced in detail a network card connection circuit and a server provided by this application. Specific examples are used in this text to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and modifications can still be made to this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A network card connection circuit, characterized in that, Including: A first adapter card, the input end of the first adapter card is connected to the network card interface of the first processor, and is used to receive the first output signal sent by the first processor, and separate the communication signal and the power supply voltage in the first output signal to obtain a first communication signal and a first power supply voltage; A second adapter card disposed on the second processor, the first input end of the second adapter card is connected to the output end of the first adapter card, the second input end of the second adapter card is connected to the data interface of the second processor, and the output end of the second adapter card is connected to the target network card, and is used to receive and integrate the first communication signal and the first power supply voltage to obtain a second output signal, and / or receive the third output signal sent by the second processor, and send the second output signal or the third output signal through the target network card.

2. The network card connection circuit according to claim 1, wherein The first adapter card includes: A signal relay module, the input end of the signal relay module is connected to the communication port of the input end of the first adapter card, and the output end of the signal relay module is adapted to be connected to the communication port of the first input end of the second adapter card, and is used to receive the communication signal in the first output signal through the communication port of the input end of the first adapter card, and generate a first communication signal based on the communication signal in the first output signal; A power supply separation module, the input end of the power supply separation module is connected to the power supply port of the input end of the first adapter card, and the output end of the power supply separation module is adapted to be connected to the power supply port of the first input end of the second adapter card, and is used to receive the power supply voltage in the first output signal through the power supply port of the input end of the first adapter card, and generate a first power supply voltage based on the power supply voltage in the first output signal.

3. The network card connection circuit according to claim 2, wherein, The signal relay module includes: A differential signal amplification unit, the input end of the differential signal amplification unit is connected to the communication port of the input end of the first adapter card, and is used to amplify the communication signal in the first output signal to generate the first communication signal.

4. The network card connection circuit according to claim 3, wherein The signal relay module further includes: An electrostatic protection unit, one end of the electrostatic protection unit is respectively connected to the communication port of the input end of the first adapter card and the input end of the differential signal amplification unit, and the other end of the electrostatic protection unit is grounded.

5. The network card connection circuit according to claim 3, wherein The signal relay module further includes: An impedance matching unit, one end of the impedance matching unit is connected to the output end of the differential signal amplification unit, and the other end of the impedance matching unit serves as the output end of the signal relay module.

6. The network card connection circuit according to any one of claims 2-5, characterized in that, The output end of the signal relay module is adapted to be connected to the communication port of the first input end of the second adapter card through a shielded cable.

7. The network card connection circuit according to claim 2, wherein The power supply separation module includes: a filtering unit and a noise isolation unit, and is used to perform filtering and noise isolation processing on the power supply voltage in the first output signal to generate a first power supply voltage.

8. The network card connection circuit according to claim 2 or 7, characterized in that, The power supply separation module further includes: A power supply protection unit, and is used to cut off the input of the power supply voltage in response to a network card migration request.

9. The network card connection circuit according to claim 2, wherein The second adapter card includes: A signal reshaping module, the input end of the signal reshaping module is connected to the communication port of the first input end of the second adapter card, and the output end of the signal reshaping module is adapted to be connected to the communication port of the output end of the second adapter card, for receiving the first communication signal and compensating for signal loss of the first communication signal to generate a second communication signal; A voltage conversion module, the input end of the voltage conversion module is connected to the power supply port of the first input end of the second adapter card, and the output end of the voltage conversion module is adapted to be connected to the power supply port of the output end of the second adapter card, for receiving the first supply voltage and converting the first supply voltage to generate a second supply voltage; Wherein, the second communication signal and the second supply voltage are integrated at the output end of the second adapter card to obtain the second output signal.

10. The network card connection circuit according to claim 9, characterized in that The signal reshaping module includes: A signal re-driving unit, for performing high-frequency gain compensation on the first communication signal based on an initial gain to generate the second communication signal.

11. The network card connection circuit according to claim 10, wherein The signal re-driving unit is further configured to adjust the initial gain based on the signal transmission error rate test result, and perform high-frequency gain compensation on the first communication signal according to the adjusted initial gain.

12. The network card connection circuit according to claim 9, wherein The voltage conversion module includes: A step-down unit, for reducing the first supply voltage to a target voltage to generate the second supply voltage.

13. The network card connection circuit according to claim 1, wherein The second adapter card further includes: A hot-swap button, for triggering a hot-swap request and sending it to the second processor, so that the second processor triggers a hot-swap control process based on the hot-swap request.

14. A server, characterized in that, Comprising: A plurality of processors, wherein the plurality of processors include a second processor and a first processor; The network card connection circuit according to any one of claims 1-13.

15. The server according to claim 14, wherein Determining the second processor based on the utilization rate of each processor in the plurality of processors.

Images