Network device and processing method thereof

Abstract

The application discloses a network device and a processing method thereof, and relates to the technical field of optical modules, wherein the network device comprises a switch chip and a multi-stage signal conditioning device; the switch chip is connected with a linear direct-drive pluggable optical module through the multi-stage signal conditioning device; the multi-stage signal conditioning device comprises a plurality of driving continuous-time linear equalizers, a plurality of switches and a controller; the plurality of driving continuous-time linear equalizers are connected in series; the output end of each driving continuous-time linear equalizer is connected with an output port through a switch; the input end of the frontmost driving continuous-time linear equalizer is connected with an input port and connected with an output port through a switch; and the controller is used for controlling the on-off state of the switches according to a control signal sent by the network device; while improving the compensation capability, low power consumption and low time delay are realized; and the application scene of the pluggable low-power-consumption optical module can be expanded to a higher-rate product on the basis of the original network device architecture.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical modules, and in particular to a network device and a processing method thereof. BACKGROUND

[0002] High-speed signals will be severely attenuated when transmitted over long distances, resulting in reduced signal bandwidth and degraded signal transmission quality. Currently, there are two commonly used methods. One is to retime the signal through DSP (Digital Signal Processing), and the other is to relay the signal through LPO (Linear drive Pluggable Optics). Among them, the DSP scheme is to apply digital signal processing technology, and its performance is the strongest. However, the power consumption of DSP is also very large. The higher the rate, the stronger the compensation capability, and the greater the power consumption. The LPO scheme is to apply analog signal processing technology, and its compensation capability is general, but its power consumption is lower, and it is more suitable for links with low compensation requirements.

[0003] In a higher rate link, the compensation capability of the LPO scheme is not enough, especially for long link scenarios. The signal attenuation caused by link loss cannot be compensated by the LPO module. However, the DSP scheme has high power consumption and high cost. Regardless of the cost, its delay is too high to meet the low delay requirement of AI (Artificial Intelligence) networking. SUMMARY

[0004] The present application provides a network device and a processing method thereof to improve the compensation capability while achieving low power consumption and low latency.

[0005] The present application provides a network device connected to a linear direct-drive pluggable optical module. The network device includes a switch chip and at least one multi-stage signal conditioning device. The first signal pin of the switch chip is connected to the linear direct-drive pluggable optical module through the multi-stage signal conditioning device. The multi-stage signal conditioning device includes N drive continuous-time linear equalizers, N+1 switches, a controller, an input port, and an output port. N is a positive integer greater than 1.

[0006] The N drive continuous-time linear equalizers are connected in series. The output of each drive continuous-time linear equalizer is connected to the output port through one switch. In the series direction of the N drive continuous-time linear equalizers, the input of the first drive continuous-time linear equalizer is connected to the input port and the output port through one switch.

[0007] The controller is configured to control the on-off state of the N+1 switches according to a control signal sent by the network device.

[0008] The application further provides a processing method of a network device, the network device being the network device of the first aspect, and the method comprising:

[0009] sending a first control signal to the controller to make the multi-stage signal conditioning device only connect the switch connected to the input port;

[0010] in the case that the multi-stage signal conditioning device is located at the receiving end, acquiring a data transmission quality parameter of a network port of the network device;

[0011] determining a compensation level according to the data transmission quality parameter, generating a second control signal according to the compensation level, and making the controller connect the corresponding switch according to the second control signal.

[0012] The application further provides an electronic device, comprising a memory configured to store a computer program and a processor configured to execute the computer program to implement the steps of the processing method of any of the network devices.

[0013] The application further provides a computer readable storage medium, and the computer readable storage medium stores a computer program, wherein the computer program is executed by a processor to implement the steps of the processing method of any of the network devices.

[0014] The application further provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the steps of the processing method of any of the network devices.

[0015] The application adds a multi-stage signal conditioning device between a switch chip and a linear direct-drive pluggable optical module, adjusts the compensation level of the multi-stage signal conditioning device according to a data transmission quality parameter, and compensates the transmission signal according to the determined compensation level, thereby improving the compensation capability, and realizing low power consumption and low latency; the scheme can expand the application scenario of the pluggable low-power LPO optical module to a higher-rate product on the basis of the original overall architecture and printed circuit board design. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the application, the drawings needed in the embodiments will be briefly introduced. 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.

[0017] Figure 1A structural schematic diagram of a network device provided by an embodiment of the present application is provided.

[0018] Figure 2 A structural schematic diagram of a multi-stage signal conditioning device provided by an example of the present application is provided.

[0019] Figure 3 A structural schematic diagram of a network device provided by an example of the present application is provided.

[0020] Figure 4 A flowchart of a processing method of a network device provided by an embodiment of the present application is provided.

[0021] Figure 5 A structural schematic diagram of a processing device of a network device provided by an embodiment of the present application is provided. DETAILED DESCRIPTION

[0022] 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 some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, any other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0023] It should be noted that, in the description of the present application, the terms “comprise”, “contain” or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. The terms “first”, “second” and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0024] Term explanation:

[0025] DSP: Digital Signal Processing, is a technology for processing digital signals through mathematical algorithms. Unlike analog signals (continuous time, continuous amplitude), digital signals are discrete time, discrete amplitude, which need to be converted from analog signals to digital form through analog-to-digital conversion (ADC), and restored to analog signals through digital-to-analog conversion (DAC) after processing. In the field of high-speed signals, DSP can realize compensation, shaping and optimization of signals through digital signal algorithms.

[0026] Optical module: is an important part of optical fiber communication, and is an optoelectronic device that realizes photoelectric conversion and electro-optic conversion functions.

[0027] Serdes: serializer / deserializer, converts parallel data into high-speed serial signal transmission (serialization), and the receiving end restores it to parallel data (deserialization), used for inter-chip, inter-device long-distance communication (such as PCIe protocol, Ethernet protocol, optical fiber communication). The core solves the problem of signal integrity, and through clock recovery and equalization technology to resist interference and attenuation.

[0028] LPO: Linear-direct-drive Pluggable Optical Modules, linear direct-drive pluggable optical modules, there is no DSP in such modules, and it is the lowest power consumption in pluggable modules, but its signal compensation capability is also the worst;

[0029] NPO: Near Packaged Optics, compared with pluggable modules, NPO places the optical engine in the optical module on the PCB board, the periphery of the switch chip, so that the distance from the switch chip to the optical engine is greatly shortened, and the loss caused by the wiring is smaller;

[0030] CPO: Co-Packaged Optics, compared with NPO, CPO has a shorter distance between the optical engine and the switch chip, and they are directly packaged together, so the loss caused by the wiring is minimal, and the signal integrity performance is better.

[0031] In related technologies, high-speed signals will be severely attenuated when transmitted over long distances, resulting in reduced signal bandwidth and degraded signal transmission quality. There are two commonly used methods, one is to retime the signal through a DSP module, and the other is to relay the signal through an LPO module. Among them, the DSP scheme is to apply digital signal processing technology, and its performance is the strongest, but the power consumption of DSP is also very large, the higher the speed, the stronger the compensation capability, and the greater the power consumption. The LPO scheme is to apply analog signal processing technology, and its compensation capability is general, but its power consumption is lower, and it is more suitable for links with low compensation requirements.

[0032] Optical Module as an important part of fiber communication, is an optoelectronic device that realizes the function of photoelectric conversion and electro-optic conversion. Optical module can realize the interconnection of network card and switch and the interconnection of different level switches. However, with the increase of Serdes rate, the influence of signal loss, crosstalk and other factors is also increasing, which has a great influence on the stability of transmission. Therefore, pluggable optical module begins to use DSP scheme to improve signal quality in the era of 50G / lane, and DSP module is widely used in the era of 50G / lane and 100G / lane. However, DSP module also has its disadvantages, that is, high power consumption, large delay, high cost, regardless of cost, power consumption is the fundamental reason for its large-scale use, followed by high delay, which cannot meet the requirement of low delay for AI networking.

[0033] In the current AI networking, the power consumption of a 400G optical module is as high as about 9W, and the power consumption of the current DSP module in the whole network accounts for more than 20%. Not only the power consumption of the module itself increases, but also the device end needs to configure a higher power fan, or increase the cost to use liquid cooling scheme for heat dissipation.

[0034] And in the era of single-wave 100G, LPO scheme began to appear, the power consumption of LPO module is reduced by 50-70% (400G about 2.5-4W) compared with DSP module, the reduction is very considerable, not only that, LPO module also greatly reduces the network delay, which is very important for AI computing, currently LPO module has begun to be applied in 100G / lane 51.2T switch in batches. However, the increase of error rate caused by the increase of Serdes rate still forces people to use CDR (clock recovery, clock data recovery) and DSP scheme, until because of the problem of power consumption, in the era of higher speed 100G / lane, LPO has been paid attention again, then whether LPO can continue to 200G / lane and higher speed models, at present, there are still many problems to be solved. The main problem is that the compensation ability of LPO module is not enough in the higher speed link, especially for long link scene, the signal attenuation caused by link loss is not compensated by LPO module.

[0035] In order to shorten the link distance and reduce the loss, without returning to the high power consumption state of the DSP module era, the 200G / lane era begins to appear NPO and CPO two forms of products. NPO is near package optical, compared with pluggable module, NPO places the optical engine in the optical module on the PCB (Printed Circuit Board), the periphery of the switch chip, so that the distance from the switch chip to the optical engine is greatly shortened, and the loss caused by the wiring is smaller. CPO is co-packaged optics, compared with NPO, CPO has a closer distance between the optical engine and the switch chip, and is directly packaged together, with minimal loss caused by internal wiring of the chip package, and better signal integrity performance. Whether it is NPO or CPO is the continuation of LPO form, but the form of pluggable module has become the form of on-board near package or co-package, at this time the network interconnection no longer needs pluggable optical module. In the 400G / lane era of higher speed, the performance advantage of NPO or CPO will be very obvious, but at present, the large-scale application of NPO and CPO faces severe challenges.

[0036] The biggest challenge faced by NPO and CPO is that the ecological environment is not mature enough, and the follow-up operation and maintenance cost is too high. First of all, in terms of operation and maintenance, LPO is still a form of pluggable optical module, and the operation and maintenance cost is very low. If one module fails, it can be replaced directly, which will not affect the normal use of other modules, nor will it cause large-scale network interruption. However, the operation and maintenance of NPO and CPO has a big problem. First of all, the integration of the optical engine (OE) of NPO and CPO is very high, and the main speed of the pluggable optical module is currently 400G / 800G. A single OE generally integrates multiple links, and once an OE fails, it will affect multiple links. In addition, the main scheme of NPO and CPO at present is external laser, that is, a laser provides light source for multiple OEs, so the impact of laser abnormality is greater. In addition, whether it is NPO or CPO, once the optical path is abnormal, it can only be stopped and opened, and even returned to the factory for repair, so it will cause large-scale network interruption, which is fatal to AI intelligent network. Another problem is that the industry ecology of NPO and CPO is very different from the existing pluggable optical module industry. At present, the NPO and CPO industry is dominated by a few chip manufacturers, and the ecology of this industry is very closed. For our company, a device manufacturer, the purchase is no longer just their chips, but a complete solution, even a complete machine, and there is little room for play. While the original optical module industry chain has hundreds or even thousands of companies in the entire industry, from the most basic components, glue, and materials to modules, software, and modules, there is a perfect industry chain. The entire industry chain system is open, and the division of labor and cooperation is good. If the chip manufacturers continue to build a closed ecological environment and do not allow the original optical module industry chain manufacturers to participate in the construction of this ecological environment, the cost of NPO or CPO may be very high, and it is difficult to be widely used.

[0037] In the face of the above situation, some views are to continue to use the pluggable DSP optical module scheme, and to use more effective cooling methods such as cold plate or immersion liquid cooling scheme to dissipate heat from the module, and to continue to use the DSP optical module scheme, which has the smallest challenge to the entire ecosystem, but the R&D and maintenance difficulty of the equipment manufacturer and the end user also increases accordingly; the disadvantages include: high power consumption and the need to use a new liquid cooling design, the power consumption of a 200G / lane 1.6T optical module is about 30W, and the power consumption of a next-generation 400G / lane 3.2T optical module is estimated to be 50-60W, and the current mainstream optical module DSP has adopted a 3nm process technology from TSMC, and it is difficult to reduce power consumption through process optimization; the current 1.6T switch has a plan to use a cold plate liquid cooling scheme to dissipate heat from the 1.6T optical module, and the traditional air cooling design is already difficult to meet, but the scheme of laying water pipes on the optical cage for heat dissipation has always been a problem in the industry, first of all, the optical cage is small in size and relatively irregular, and there is no flat surface, unlike the cold plate liquid cooling design on the PCB, which is more feasible. To 400G / lane 3.2T optical module, cold plate is also difficult to dissipate heat effectively, and can only use immersion liquid cooling, which is a great challenge to the entire industry. Some views are to abandon the pluggable optical module interconnection scheme, build a new NPO or CPO ecosystem, and use a linear optical scheme to package the optical engine close to the switch chip to shorten the link wiring and ensure better link signal quality; however, the NPO and CPO schemes have the following problems: inconvenient maintenance, single optical engine damage or laser damage will cause multiple link shutdowns, and once a problem occurs, it cannot be replaced by plugging, but only by powering off the entire machine for maintenance, or even by returning to the factory for repair; closed ecosystem, imperfect industrial chain, only a few chip manufacturers are currently involved, and the cost is high.

[0038] Under this background, the entire industry is thinking about countermeasures, whether it is equipment manufacturers, optical module manufacturers, or upstream analog chip manufacturers. This application is from the perspective of equipment manufacturers, in order to continue the use of LPO modules at higher speeds, the scheme will not bring a significant increase in power consumption, and can meet the low latency requirements of AI networking; continue to use the original machine architecture, the entire ecosystem is more mature, the R&D cycle of the entire machine is shorter, and the risk is minimal; there is no need to develop a more difficult cold plate scheme for a special optical cage, and there is no need to worry about the maintenance problems brought by the immersion liquid cooling scheme, nor to rebuild the NPO or CPO ecosystem chain.

[0039] In order for those skilled in the art to better understand the present application, the following will further describe the present application in detail in combination with the drawings and specific embodiments.

[0040] Embodiments of the present application provide a network device, Figure 1 is a structural schematic diagram of a network device provided by embodiments of the present application, as Figure 1 shown, the network device connects a linear direct-drive pluggable optical module, and the network device comprises a switch chip (Switch) and at least one multi-stage signal conditioning device, and a first signal pin of the switch chip is connected to the linear direct-drive pluggable optical module through the multi-stage signal conditioning device.

[0041] In some embodiments, the multi-stage signal conditioning device is a controllable multi-stage amplification on-board Re-Driver chip, and through signal compensation of the multi-stage amplification on-board Re-Driver on a signal link, it can be realized that in the existing device framework system, the low-power pluggable LPO optical module scheme (and the heat dissipation of the LPO module continues to use air cooling) can continue to be used, and the demand for higher-rate signal transmission can also be met.

[0042] In some embodiments, the multi-stage signal conditioning device comprises N driver continuous-time linear equalizers (Driver CTLEs), N+1 switches, a controller, an input port (INPUT) and an output port (OUTPUT), N is a positive integer greater than 1; the N driver continuous-time linear equalizers are connected in series, the output end of each driver continuous-time linear equalizer is connected to the output port through a switch, along the series direction of the N driver continuous-time linear equalizers, the input end of the driver continuous-time linear equalizer at the front end is connected to the input port and connected to the output port through a switch, and the controller is used to control the on-off state of the N+1 switches according to a control signal sent by the network device.

[0043] As an example, as Figure 2As shown, the multi-stage signal conditioning device includes 3-stage step control, including 4 switches. When the switch SW0 is connected, the multi-stage signal conditioning device is equivalent to a high-speed wire, the chip is equivalent to being bypassed, at this time only the controller is working, the multi-stage drive continuous-time linear equalizer is in the power-off state, and the power consumption is the lowest, in the order of milliwatt, suitable for short link or LPO internal with sufficient compensation capacity scene; when the switch SW1 is connected, only the first stage of the multi-stage signal conditioning device drive continuous-time linear equalizer works, and the other two stages are in the power-off state, at this time the multi-stage signal conditioning device will perform primary compensation on the high-speed link, and the multi-stage signal conditioning device does not need to work at the highest power consumption, which can effectively reduce the power consumption; when the switch SW2 is connected, the first two stages of the multi-stage signal conditioning device drive continuous-time linear equalizer are in the working state, and the third stage of the multi-stage signal conditioning device drive continuous-time linear equalizer is powered off, at this time the multi-stage signal conditioning device will perform intermediate compensation on the high-speed link, and the multi-stage signal conditioning device still does not need to work at the highest power consumption; when the switch SW3 is connected, all the multi-stage signal conditioning device drive continuous-time linear equalizers are in the working state, at this time the multi-stage signal conditioning device will perform maximum compensation on the high-speed link, and the multi-stage signal conditioning device works at the highest power consumption, but the compensation capacity at this time is the strongest.

[0044] The external interface of the multi-stage signal conditioning device includes a low-speed signal channel connected to the controller in addition to the high-speed link and the power supply. The low-speed signal channel can be I2C or GPIO, etc. The network device controls the controller through the low-speed signal, thereby controlling the on-off of the multi-stage CTLE switch.

[0045] It should be noted that the multi-stage signal conditioning device is essentially a driver with multi-stage CTLE. In order to ensure the linearity of the high-speed signal, the compensation capacity of each CTLE is controlled at a small level, which can ensure the linearity of the high-speed signal as much as possible. For the multi-stage signal conditioning device, the control stage number is not limited. If the control stage number is N, then N+1 switches are included inside, which are controlled by the internal controller. The switch can send control signals or control instructions to the controller through I2C or other GPIO interface, thereby controlling the on-off of the switch.

[0046] The application embodiment innovatively proposes a multi-stage controllable driving scheme of an LPO optical module under a high-speed link, and by adopting a multi-stage signal conditioning device, i.e., a driver with multi-stage CTLE, the power consumption can be saved to the maximum extent while ensuring the quality of the transmission signal. The scheme increases a controllable multi-stage on-board multi-stage signal conditioning device between a switch chip and a pluggable LPO module, and the signal is amplified and shaped. The LPO module does not need to have strong compensation capability, so the power consumption of the LPO module is not high, and the LPO module does not need to be liquid-cooled. The task of signal relay amplification is transferred to the controllable multi-stage on-board multi-stage signal conditioning device chip. Compared with the digital signal compensation mechanism of the DSP, the power consumption of the analog signal compensation mechanism of the multi-stage signal conditioning device is smaller. According to the calculation, only about 1.5W-2W is needed for a single 1.6T link. In the short link scene without too much compensation, the power consumption is even in the milliwatt level. Compared with the use of the DSP module, the scheme still has a very large power consumption reduction capability (the power consumption of a single 1.6T DSP module is 25W-30W, and the power consumption of a single 1.6T LPO module is 10W-12W).

[0047] In order to better protect the port of the network device from adapting the LPO module, in some embodiments, the network device includes a printed circuit board, as shown in Figure 3 The network device includes a plurality of multi-stage signal conditioning devices (Re-Driver chips), and the multi-stage signal conditioning devices correspond to a plurality of optical ports of the network device one by one. The wiring distance of each multi-stage signal conditioning device and the corresponding optical port on the printed circuit board is the same.

[0048] That is, by controlling the wiring distance between each multi-stage signal conditioning device and the optical port to be the same, the port of the network device is adapted to the LPO module.

[0049] In some embodiments, the controller can be a microcontroller such as a single-chip microcontroller. The controller includes a register, and different values are written to the register according to different control signals. According to the values written in the register, the corresponding switch in the N+1 switches is connected. That is, different switches are connected by simply writing the register. The scheme has a simple structure and is easy to implement.

[0050] It should be noted that the inventive concept of the scheme is not only applicable to the LPO optical module, but also applicable to the ACC active copper cable. It is not only applicable to the existing network protocol (Ethernet, Infiniband network), but also applicable to other communication protocols, such as PCIE, SATA, and other interfaces of the network device.

[0051] On the basis of the above-mentioned embodiments, the application further provides a processing method of a network device, as shown in Figure 4As shown, the processing method of the network device comprises the following steps:

[0052] In step 401, a first control signal is sent to the controller to make the multi-stage signal conditioning device only connect the switch connected with the input port.

[0053] It can be understood that the scheme needs to bypass the multi-stage signal conditioning device first, and then obtain the data transmission quality parameter such as the bit error rate; and then determine whether the compensation effect of the multi-stage signal conditioning device is needed according to the obtained data transmission quality parameter.

[0054] In step 402, in the case that the multi-stage signal conditioning device is located at the receiving end, the data transmission quality parameter of the network port of the network device is obtained, and the compensation level is determined according to the data transmission quality parameter.

[0055] In some embodiments, the data transmission quality parameter is the bit error rate, that is, whether compensation is needed and the compensation level are determined according to the bit error rate of the signal.

[0056] In some embodiments, the above-mentioned determining the compensation level according to the data transmission quality parameter comprises: determining the compensation level according to the data transmission quality parameter and the corresponding relationship between the data transmission quality parameter range and the compensation level.

[0057] That is, the corresponding relationship between the data transmission quality parameter range and the compensation level is set in advance, and after the data transmission quality parameter is obtained, it is determined which data transmission quality parameter range the data transmission quality parameter falls into, and the corresponding compensation level is determined according to the range.

[0058] In some embodiments, the above-mentioned determining the compensation level according to the data transmission quality parameter comprises: determining the compensation level according to the data transmission quality parameter and the wiring distance between the multi-stage signal conditioning device and the switch chip.

[0059] That is, the data transmission quality and the wiring distance can be considered at the same time, and the compensation level is determined according to the two factors, such as setting different weights for weighted summation to determine the compensation level.

[0060] In some embodiments, before obtaining the data transmission quality parameter of the network port of the network device, it comprises: performing parameter configuration on the linear direct drive pluggable optical module through the link training mechanism.

[0061] Therefore, for LPO modules of different manufacturers and different compensation capabilities, the network device end can realize the configuration of the optimal parameters through the link training mechanism, and through the parameter configuration, the optimal compensation capability of the LPO module is achieved; and on this basis, whether the multi-stage signal conditioning device is needed and the required compensation level are determined according to the data transmission quality parameter such as the bit error rate, so as to save the power consumption to the maximum.

[0062] In the case where the multi-stage signal conditioning device is located at the sending end, the data transmission quality parameter fed back by the opposite end is acquired, and the compensation level is determined according to the data transmission quality parameter; or the compensation level is determined based on the wiring distance between the multi-stage signal conditioning device and the switch chip.

[0063] In the case where the multi-stage signal conditioning device is located at the sending end, the data transmission quality parameter cannot be directly acquired, and thus the data transmission quality parameter, such as the bit error rate, fed back by the opposite end, i.e., the corresponding receiving end, is required to determine the compensation level; in the case where the data transmission quality parameter of the opposite end cannot be acquired, the compensation level can be determined based on the wiring distance between the multi-stage signal conditioning device and the switch chip, and the longer the wiring distance, the higher the compensation level.

[0064] It can be understood that, due to the limitation of the chip size and structure layout, the wiring distances between the switch chip and the multi-stage signal conditioning devices are different, and the compensation amount of the multi-stage signal conditioning device is configured according to the length of the wiring, and the longer the wiring, the greater the compensation amount, so as to guarantee the signal consistency out of the multi-stage signal conditioning device, and thus better port consistency can be guaranteed.

[0065] In step 404, a second control signal is generated according to the compensation level, and the controller is connected according to the second control signal.

[0066] It can be understood that, the compensation level is controlled by controlling the connection of different switches, i.e., by sending a control signal to the controller to control the connection of the corresponding switch.

[0067] The processing method of the network device of the embodiment can increase the multi-stage signal conditioning device between the switch chip and the linear direct-drive pluggable optical module, adjust the compensation level of the multi-stage signal conditioning device according to the data transmission quality parameter, and compensate the transmission signal according to the determined compensation level, so as to improve the compensation capability while realizing low power consumption and low latency; the scheme can expand the application scenario of the pluggable low-power LPO optical module to 200G / lane or even higher rate products on the basis of the original overall architecture and printed circuit board design; compared with the DSP scheme, a large-power fan or a higher-cost liquid cooling scheme does not need to be designed, and the original overall architecture can still be used, and compared with the NPO or CPO scheme, the cost is lower, the ecology is perfect, and the maintenance cost is lower.

[0068] 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 realized by means of software and the necessary general hardware platform, and of course it can also be realized by hardware, but in many cases the former is a better embodiment.

[0069] To implement the processing method of the network device provided by the embodiments of the present disclosure, the embodiments of the present disclosure further provide a processing apparatus of a network device. The apparatus is configured in a processing system of the network device, as shown in the figure, and the apparatus comprises: Figure 5

[0070] The sending module 510 is configured to send a first control signal to the controller, so that the multi-stage signal conditioning device only connects the switch connected with the input port.

[0071] The data processing module 520 is configured to, in the case that the multi-stage signal conditioning device is located at the receiving end, acquire a data transmission quality parameter of the network port of the network device, and determine the compensation level according to the data transmission quality parameter.

[0072] The signal generation module 530 is configured to generate a second control signal according to the compensation level, so that the controller connects the corresponding switch according to the second control signal.

[0073] In some implementations, the data processing module 520 is further configured to:

[0074] In the case that the multi-stage signal conditioning device is located at the sending end, acquire a data transmission quality parameter fed back by the opposite end, and determine the compensation level according to the data transmission quality parameter; or, determine the compensation level based on the wiring distance between the multi-stage signal conditioning device and the switch chip.

[0075] In some implementations, when determining the compensation level according to the data transmission quality parameter, the data processing module 520 is specifically configured to:

[0076] determine the compensation level according to the corresponding relationship between the data transmission quality parameter range and the compensation level and the data transmission quality parameter.

[0077] In some implementations, when determining the compensation level according to the data transmission quality parameter, the data processing module 520 is specifically configured to:

[0078] determine the compensation level according to the data transmission quality parameter and the wiring distance between the multi-stage signal conditioning device and the switch chip.

[0079] In some implementations, before acquiring the data transmission quality parameter of the network port of the network device, the data processing module 520 is further configured to:

[0080] perform parameter configuration on the linear direct-drive pluggable optical module through a link training mechanism.

[0081] In some implementations, the data transmission quality parameter is a bit error rate.

[0082] ​The description of the features in the embodiments of the processing apparatus of the network device can refer to the related description of the embodiments of the processing method of the network device, which will not be repeated here.

[0083] The embodiments of the present application also provide an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor is configured to execute the computer program to perform the steps in any of the embodiments of the processing method of the network device.

[0084] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to perform the steps in any of the embodiments of the processing method of the network device when executed.

[0085] In an example embodiment, the computer readable storage medium described above can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0086] The embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps in any of the embodiments of the processing method of the network device.

[0087] The embodiments of the present application also provide another computer program product, which comprises a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in any of the embodiments of the processing method of the network device.

[0088] The skilled person can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0089] The above describes in detail the link fault root cause positioning provided by the present application. The principles and implementation manners of the present application are described by using specific examples, and the above description of the embodiments is only used to help understand the method of the present application and its core idea. It should be pointed out that, for those skilled in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. A network device, characterized in that, The network device connects to a linear direct-drive pluggable optical module. The network device includes a switch chip and at least one multi-level signal conditioning device. The first signal pin of the switch chip is connected to the linear direct-drive pluggable optical module through the multi-level signal conditioning device. The multi-level signal conditioning device includes N driving continuous-time linear equalizers, N+1 switches, a controller, an input port, and an output port, where N is a positive integer greater than 1. The N continuous-time linear equalizers are connected in series. The output of each continuous-time linear equalizer is connected to the output port through a switch. Along the series connection direction of the N continuous-time linear equalizers, the input of the first continuous-time linear equalizer is connected to the input port and also connected to the output port through a switch. The controller is used to control the on / off state of the N+1 switches according to the control signals sent by the network device; The network device includes a printed circuit board and a plurality of the multi-level signal conditioning devices, each of which corresponds one-to-one with a plurality of optical ports of the network device.

2. The network device according to claim 1, characterized in that, The trace distances of each of the multi-level signal conditioning devices and the corresponding optical ports on the printed circuit board are the same.

3. The network device according to claim 1, characterized in that, The controller includes a register. The controller writes different values ​​into the register according to different control signals, and controls the corresponding switches among the N+1 switches to be turned on according to the values ​​written in the register.

4. A processing method for a network device, characterized in that, The network device is the network device as described in claim 1 or claim 2, and the method includes: Send a first control signal to the controller, causing the multi-level signal conditioning device to connect only the switch connected to the input port; When the multi-level signal conditioning device is located at the receiving end, the data transmission quality parameters of the network port of the network device are obtained, and the compensation level is determined based on the data transmission quality parameters. Based on the compensation level, a second control signal is generated, causing the controller to activate the corresponding switch according to the second control signal.

5. The method according to claim 4, characterized in that, The method further includes: When the multi-level signal conditioning device is located at the transmitting end, the data transmission quality parameters fed back from the other end are obtained, and the compensation level is determined based on the data transmission quality parameters; or, the compensation level is determined based on the trace distance between the multi-level signal conditioning device and the switch chip.

6. The method according to claim 4, characterized in that, Determining the compensation level based on the data transmission quality parameters includes: The compensation level is determined based on the correspondence between the data transmission quality parameter range and the compensation level, as well as the data transmission quality parameters.

7. The method according to claim 4, characterized in that, Determining the compensation level based on the data transmission quality parameters includes: The compensation level is determined based on the data transmission quality parameters and the trace distance between the multi-level signal conditioning device and the switch chip.

8. The method according to claim 4, characterized in that, Before obtaining the data transmission quality parameters of the network port of the network device, the following steps are included: The parameters of the linear direct-drive pluggable optical module are configured through a link training mechanism.

9. The method according to claim 4, characterized in that, The data transmission quality parameter is the bit error rate.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the method as described in any one of claims 4 to 9.

Images