Signal transmission mode adjustment method, apparatus and system
By switching the signal transmission mode when a channel anomaly is detected in the AI large model computing data center, the signal reliability problem caused by the parallel transmission of multiple channels of optical modules is solved, and the reliability of signal transmission is improved while reducing the transmission rate.
Patent Information
- Application Number
- PCT/CN2025/080228
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-04-08
- Filing Date
- 2025-03-03
- Publication Date
- 2025-11-27
AI Technical Summary
In AI large-scale model computing data centers, the multi-channel parallel transmission of optical modules suffers from poor signal transmission reliability due to failure or abnormality of any channel, affecting the overall stability of optical signal transmission.
When a channel anomaly is detected, the signal transmission mode is switched to the second signal transmission mode to reduce the transmission rate and ensure normal signal transmission. This includes adjusting the channel working state through self-negotiation or coordination with the control server, switching to independent mode or reducing the modulation format, and other processing.
It effectively reduces the impact of channel anomalies on signal transmission, improves the reliability of signal transmission, and ensures normal signal transmission even with partial data rate loss.
Smart Images

Figure CN2025080228_27112025_PF_FP_ABST
Abstract
Description
Method, device and system for adjusting signal transmission mode
[0001] The present application claims priority from the Chinese patent application No. 202410420794.0 filed on April 8, 2024, and entitled "Method, device and system for adjusting signal transmission mode", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD
[0002] The present application relates to the field of communication technology, and in particular to a method, device and system for adjusting signal transmission mode. BACKGROUND
[0003] With the emergence and popularity of large model computing applications, a new type of data center for artificial intelligence (AI) large model computing has emerged. Unlike traditional general-purpose computing data centers, AI large model cluster computing requires a large number of computing cards, and the computing cards are interconnected through a large number of high-speed optical modules. The optical modules used in AI cluster computing data centers have extremely high speeds, and the optical modules are applied in a concentrated manner and in large quantities.
[0004] In existing data centers, the optical module receives multiple parallel data from the switch, modulates the multiple data to obtain multiple optical signals, and then sends the multiple optical signals out through multiple optical fibers to form optical interconnection of the data center. Since the optical module relies on multi-channel parallelism, any channel failure or abnormality will affect the normal operation of the optical module, resulting in poor reliability of optical signal transmission. SUMMARY
[0005] The embodiments of the present application provide a method, device and system for adjusting signal transmission mode, which can effectively deal with the situation of channel abnormality, reduce the influence of channel abnormality on signal transmission, and improve the reliability of signal transmission.
[0006] In a first aspect, the embodiments of the present application provide a method for adjusting a signal transmission mode. Specifically, in a normal working state, a first communication device uses a first signal transmission mode to perform signal transceiving with a second communication device. The first communication device and the second communication device are connected through K channels, K is an integer greater than 1, and the first signal transmission mode is that the first communication device performs transceiving of a first modulation format signal with the second communication device through the K channels. If the first communication device receives abnormal signals from m channels among the K channels, the first communication device switches the m channels to a second signal transmission mode to perform signal transceiving with the second communication device. 1≤m≤K, and the signal transmission rate between the first communication device and the second communication device after the m channels are switched to the second signal transmission mode is less than the signal transmission rate of the first signal transmission mode.
[0007] In this embodiment, if the first communication device receives abnormal signals from m channels among the K channels, the first communication device and the second communication device will switch the working state. Compared with the normal working state, both the communication devices at both ends are subjected to speed reduction processing, for example, closing abnormal channels or changing the modulation format of signals, so as to effectively cope with the situation that the channels are abnormal. Although part of the transmission rate is sacrificed, the normal transmission of signals is still ensured, the influence of channel abnormality on signal transmission is reduced, and the reliability of signal transmission is improved.
[0008] In some possible embodiments, if the first communication device receives abnormal signals from m channels, the first communication device switches the m channels to a second signal transmission mode to perform signal transceiving with the second communication device, including: the first communication device first switches the m channels to the second signal transmission mode to send signals to the second communication device, and then receives signals sent by the second communication device after the m channels are switched to the second signal transmission mode.
[0009] In the embodiment, the first communication device and the second communication device switch the working state through self-negotiation. If the first communication device finds that the signal received from the m channels of the K channels is abnormal, the first communication device first adjusts the working mode of the sending end, that is, switches to the second signal transmission mode to send the signal to the second communication device. Since the first communication device switches to the second signal transmission mode, the signal received by the second communication device based on the previous first signal transmission mode is also abnormal, and therefore the second communication device also synchronously changes the working mode of the sending end and switches to the second signal transmission mode to send the signal to the first communication device. Further, the first communication device detects that the sending end of the second communication device has also changed the working mode, and thus determines that both parties have switched to the new working mode. That is, the first communication device that detects the abnormal signal first adjusts the working mode of the sending end, which is equivalent to initiating the negotiation request for the working mode switching to the second communication device. After receiving the negotiation request from the first communication device, the second communication device also adjusts the working mode of the sending end, which is equivalent to feeding back to the first communication device that the negotiation request has been agreed. The first communication device based on the signal transmission mode switching of the second communication device can confirm that the second communication device has agreed to the negotiation request, which is equivalent to completing the negotiation process between the two parties. In this way, the self-negotiation mode does not need the participation of the control server, and is more practical.
[0010] In some possible embodiments, if the first communication device receives the signal sent by the second communication device by switching the m channels to the second signal transmission mode without abnormality, the method further includes that the first communication device sends a reporting message, and the reporting message is used to indicate that the first communication device performs signal transmission and reception with the second communication device based on the second signal transmission mode. That is, after the first communication device and the second communication device switch the working state, it is further detected whether the received signal is abnormal. If there is no abnormality, it means that the two parties can perform signal transmission based on the switched working state, which provides a good guarantee for the normal signal transmission.
[0011] In some possible embodiments, if the first communication device finds that the signal received from the m channels is abnormal, before the first communication device switches the m channels to the second signal transmission mode to perform signal transmission and reception with the second communication device, the method further includes that the first communication device sends a first message to the control server, and the first message is used to indicate that the first communication device finds that the signal received from the m channels is abnormal; and the first communication device receives a second message sent by the control server based on the first message, and the second message is used to indicate that the first communication device switches the m channels to the second signal transmission mode to perform signal transmission and reception with the second communication device.
[0012] In this embodiment, if the first communication device receives abnormal signals from the m channels, the first communication device reports the abnormal situation to the control server, and the control server makes a decision and issues instructions to the first communication device and the second communication device to switch the working states of the first communication device and the second communication device. Through this method of centralized control of the working states of the communication devices by the control server, the implementation mode of the present scheme is enriched.
[0013] In some possible implementation modes, before the first communication device switches the m channels to the second signal transmission mode and performs signal transceiving with the second communication device, the method further includes: the first communication device performs at least one of the following detections on the signals received from the K channels: the first communication device detects the modulation format of the signals; the first communication device detects the number of Loss of Signal (LoS) alarms; the first communication device detects the number of Loss of Frame (LoF) alarms; the first communication device detects the number of packet loss; and the first communication device detects the number of uncorrectable errors. This embodiment provides multiple implementation modes for abnormality detection, and tries to ensure the accuracy of abnormality detection as much as possible.
[0014] In some possible implementation modes, the abnormality of the signals received by the first communication device from one channel includes at least one of the following situations: the number of LoS alarms of the signals received by the first communication device from one channel is greater than or equal to a first threshold value; the number of LoF alarms of the signals received by the first communication device from one channel is greater than or equal to a second threshold value; the number of packet loss of the signals received by the first communication device from one channel is greater than or equal to a third threshold value; and the number of uncorrectable errors of the signals received by the first communication device from one channel is greater than or equal to a fourth threshold value.
[0015] In some possible implementation modes, the data transmitted through the K channels in the first signal transmission mode are in a coupling mode of mutual correlation, which can be referred to as a bonding mode. Before the first communication device switches the m channels to the second signal transmission mode and performs signal transceiving with the second communication device, if the first communication device receives abnormal signals from the m channels, the method further includes: the first communication device switches the data transmitted through the K channels to a mode of mutual independence, which is a breakout mode different from the bonding mode. Since the data transmitted through the channels in the breakout mode are mutually independent, the normal transmission of other channels can be avoided from being affected by the abnormal channel, and the signal transmission mode of the abnormal channel can be adjusted in a targeted manner, which is conducive to ensuring the normal transmission of signals.
[0016] In some possible implementation manners, the switching of the m channels into the second signal transmission mode by the first communication device for signal transceiving with the second communication device includes but is not limited to the following cases, which enrich the implementation manners of the scheme.
[0017] The first communication device closes the m channels, transmits the first modulation format signal to the second communication device through n channels other than the m channels in the K channels, receives the first modulation format signal transmitted by the second communication device through the n channels, and K = m + n.
[0018] Alternatively, the first communication device transmits the second modulation format signal to the second communication device through the K channels, and receives the second modulation format signal transmitted by the second communication device through the K channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal.
[0019] Alternatively, the first communication device transmits the second modulation format signal to the second communication device through the m channels, transmits the first modulation format signal to the second communication device through n channels other than the m channels in the K channels, receives the second modulation format signal transmitted by the second communication device through the m channels, and receives the first modulation format signal transmitted by the second communication device through the n channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal, and K = m + n.
[0020] Alternatively, the first communication device transmits the first modulation format signal to the second communication device through n channels other than the m channels in the K channels, closes p channels, transmits the second modulation format signal to the second communication device through q channels other than the p channels in the m channels, receives the first modulation format signal transmitted by the second communication device through the n channels, and receives the second modulation format signal transmitted by the second communication device through the q channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal, K = m + n, m = p + q, and p and q are both integers greater than or equal to 1.
[0021] In some possible implementation manners, the first modulation format signal is a four-level pulse amplitude modulation (Pulse Amplitude Modulation 4-leve, PAM4) signal, and the second modulation format signal is a two-level pulse amplitude modulation (Pulse Amplitude Modulation 2-leve, PAM2) signal or a non-return to zero modulation (Non-Return to Zero, NRZ) signal. It has good application value in the scene where the channel rate exceeds 50 gigabits per second (Gigabits Per Second, Gbps).
[0022] In some possible implementation, if the number of LoS alarms of the signals received by the first communication device from each of the m channels is greater than or equal to the first threshold, the switching of the m channels into the second signal transmission mode for signal transceiving with the second communication device by the first communication device includes: the first communication device closing the m channels, sending the first modulation format signals to the second communication device through the n channels other than the m channels in the K channels, receiving the first modulation format signals sent by the second communication device through the n channels, and K = m + n.
[0023] Alternatively, if the number of LoS alarms of the signals received by the first communication device from each of the p channels is greater than or equal to the first threshold, the switching of the m channels into the second signal transmission mode for signal transceiving with the second communication device by the first communication device includes: the first communication device sending the first modulation format signals to the second communication device through the n channels other than the m channels in the K channels, closing the p channels, sending the second modulation format signals to the second communication device through the q channels other than the p channels in the m channels, receiving the first modulation format signals sent by the second communication device through the n channels, and receiving the second modulation format signals sent by the second communication device through the q channels, K = m + n, m = p + q, and p and q are both integers greater than or equal to 1.
[0024] That is, in a possible scenario, if the abnormality of the received signals detected by the first communication device is LoS, changing the modulation format of the signals in this scenario has no meaning, and directly closing the related channels can be equivalent to formulating a reasonable response mode according to the specific situation of the reception abnormality.
[0025] In some possible implementation, if the number of uncorrectable errors of the signals received by the first communication device from each of the m channels is greater than or equal to the fourth threshold, the method further includes: the first communication device detecting the modulation format of the signals received from the m channels; if the first communication device receives the first modulation format signals from the m channels, the first communication device sends the second modulation format signals to the second communication device through the m channels, receives the second modulation format signals sent by the second communication device through the m channels, sends the first modulation format signals to the second communication device through the n channels other than the m channels in the K channels, and receives the first modulation format signals sent by the second communication device through the n channels, K = m + n. That is, in a possible scenario, if the abnormality of the received signals detected by the first communication device is uncorrectable error, the modulation format of the signals can be changed first instead of directly closing the channels, and the transmission rate is as much as possible not sacrificed.
[0026] In some possible implementation manners, if the number of uncorrectable errors of the second modulation format signal received by the first communication device from each of the m channels is greater than or equal to the fourth threshold value, the method further includes: the first communication device closing the m channels, sending the first modulation format signal to the second communication device through the n channels, and receiving the first modulation format signal sent by the second communication device through the n channels. That is, in a possible scenario, if the abnormal reception still occurs after the modulation format of the signal is changed, the channels can be closed, and the normal transmission of the signal can be preferentially ensured through further speed reduction processing.
[0027] In some possible implementation manners, if the number of uncorrectable errors of the signal received by the first communication device from each of the m channels is greater than or equal to the fourth threshold value, the method further includes: the first communication device detecting the modulation format of the signal received by the m channels; if the first communication device receives the second modulation format signal from the m channels, the first communication device sends the second modulation format signal to the second communication device through the m channels, and sends the first modulation format signal to the second communication device through the n channels other than the m channels in the K channels. That is, in a possible scenario, if the first communication device detects that the received signal is the second modulation format, it indicates that the second communication device changes the modulation format first, and the first communication device also needs to change the modulation format of the to-be-sent signal synchronously to ensure the smooth progress of the auto-negotiation.
[0028] In some possible implementation manners, if the first communication device receives the signal from the m channels in the K channels, the first communication device switches the m channels to the second signal transmission mode to perform signal transceiving with the second communication device, including: the first communication device switches the K channels to the second signal transmission mode to perform signal transceiving with the second communication device. After the K channels are switched to the second signal transmission mode, the signal transmission rate between the first communication device and the second communication device is less than the signal transmission rate of the first signal transmission mode. This embodiment is equivalent to switching the channels that have abnormal reception and the channels that have no abnormal reception to the second signal transmission mode, for example, switching the first communication device and the second communication device to perform transceiving of the second modulation format signal through the K channels, which enriches the implementation manners of the speed reduction processing of the present solution.
[0029] In a second aspect, an embodiment of the present application provides a communication device, which is referred to as a first communication device herein. The first communication device comprises a transceiver. The transceiver is configured to perform signal transceiving with a second communication device in a first signal transmission mode. The first communication device and the second communication device are connected by K channels, where K is an integer greater than 1. The first signal transmission mode is a mode in which the first communication device performs signal transceiving with the second communication device in a first modulation format via the K channels. If the first communication device receives signals from m channels among the K channels abnormally, the first communication device switches the m channels to a second signal transmission mode, where 1≤m≤K. The signal transmission rate between the first communication device and the second communication device in the second signal transmission mode is less than the signal transmission rate in the first signal transmission mode.
[0030] In some possible implementation manners, if the first communication device receives signals from the m channels abnormally, the transceiver is configured to first switch the m channels to the second signal transmission mode to send signals to the second communication device, and then receive signals sent by the second communication device by switching the m channels to the second signal transmission mode.
[0031] In some possible implementation manners, if the first communication device receives signals sent by the second communication device by switching the m channels to the second signal transmission mode without abnormality, the transceiver is further configured to send a reporting message. The reporting message is used to instruct the first communication device to perform signal transceiving with the second communication device in the second signal transmission mode.
[0032] In some possible implementation manners, if the first communication device receives signals from the m channels abnormally, before the first communication device performs signal transceiving with the second communication device by switching the m channels to the second signal transmission mode, the transceiver is further configured to send a first message to a control server. The first message is used to indicate that the first communication device receives signals from the m channels abnormally. The transceiver is further configured to receive a second message sent by the control server according to the first message. The second message is used to instruct the first communication device to perform signal transceiving with the second communication device by switching the m channels to the second signal transmission mode.
[0033] In some possible implementation manners, the first communication device further comprises a processing unit. Before the first communication device performs signal transceiving with the second communication device by switching the m channels to the second signal transmission mode, the processing unit is configured to perform at least one of the following detections on signals received from the K channels: detecting a modulation format used by the signals; detecting a number of LoS alarms; detecting a number of frame loss LoF alarms; detecting a number of packet losses; and detecting a number of uncorrectable errors.
[0034] In some possible implementation manners, the processing unit receiving the abnormal signal from one channel includes at least one of the following: the processing unit receiving a number of LoS alarms of the signal from one channel being greater than or equal to a first threshold; the processing unit receiving a number of LoF alarms of the signal from one channel being greater than or equal to a second threshold; the processing unit receiving a number of packet loss of the signal from one channel being greater than or equal to a third threshold; and the processing unit receiving a number of uncorrectable errors of the signal from one channel being greater than or equal to a fourth threshold.
[0035] In some possible implementation manners, the transceiving unit is specifically configured to: close the m channels, send the first modulation format signal to the second communication device through n channels of the K channels other than the m channels, receive the first modulation format signal sent by the second communication device through the n channels, and K = m + n.
[0036] Or, send the second modulation format signal to the second communication device through the K channels, and receive the second modulation format signal sent by the second communication device through the K channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal.
[0037] Or, send the second modulation format signal to the second communication device through the m channels, send the first modulation format signal to the second communication device through n channels of the K channels other than the m channels, receive the second modulation format signal sent by the second communication device through the m channels, and receive the first modulation format signal sent by the second communication device through the n channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal, and K = m + n.
[0038] Or, send the first modulation format signal to the second communication device through n channels of the K channels other than the m channels, close the p channels, send the second modulation format signal to the second communication device through q channels of the m channels other than the p channels, receive the first modulation format signal sent by the second communication device through the n channels, and receive the second modulation format signal sent by the second communication device through the q channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal, K = m + n, m = p + q, and p and q are both integers greater than or equal to 1.
[0039] In some possible implementation manners, the data transmitted through the K channels based on the first signal transmission mode are in a coupling mode of mutual correlation, which can be referred to as a bonding mode. The first communication device further includes a processing unit. If the first communication device abnormally receives signals from m channels of the K channels, before the first communication device switches the m channels to the second signal transmission mode and performs signal transceiving with the second communication device, the processing unit is configured to switch the data transmitted through the K channels to a mode of mutual independence.
[0040] In some possible implementation manners, the first modulation format signal is a PAM4 signal, and the second modulation format signal is a PAM2 signal or an NRZ signal.
[0041] In some possible implementation manners, if the number of LoS alarms of the signals received by the first communication device from each of the m channels is greater than or equal to a first threshold, the processing unit is configured to close the m channels; and the transceiving unit is configured to send the first modulation format signal to the second communication device through n channels of the K channels except the m channels, receive the first modulation format signal sent by the second communication device through the n channels, and K = m + n. Alternatively, if the number of LoS alarms of the signals received by the first communication device from each of the p channels is greater than or equal to the first threshold, the processing unit is configured to close the p channels; and the transceiving unit is configured to send the second modulation format signal to the second communication device through q channels of the m channels except the p channels, receive the first modulation format signal sent by the second communication device through the n channels, and receive the second modulation format signal sent by the second communication device through the q channels, K = m + n, m = p + q, and p and q are both integers greater than or equal to 1.
[0042] In some possible implementation manners, if the number of uncorrectable errors of the signals received by the first communication device from each of the m channels is greater than or equal to a fourth threshold, the processing unit is configured to detect the modulation format of the signals received from the m channels; if the first communication device receives the first modulation format signal from the m channels, the transceiving unit is configured to send the second modulation format signal to the second communication device through the m channels, receive the second modulation format signal sent by the second communication device through the m channels, send the first modulation format signal to the second communication device through n channels of the K channels except the m channels, and receive the first modulation format signal sent by the second communication device through the n channels, and K = m + n.
[0043] In some possible implementation manners, if the number of uncorrectable errors of the second modulation format signal received by the first communication device from each of the m channels is greater than or equal to the fourth threshold value, the processing unit is configured to close the m channels; and the transceiver is configured to send the first modulation format signal to the second communication device through the n channels, and receive the first modulation format signal sent by the second communication device through the n channels.
[0044] In some possible implementation manners, if the number of uncorrectable errors of the signal received by the first communication device from each of the m channels is greater than or equal to the fourth threshold value, the processing unit is configured to detect the modulation format of the signal received from the m channels; and if the first communication device receives the second modulation format signal from the m channels, the transceiver is configured to send the second modulation format signal to the second communication device through the m channels, and send the first modulation format signal to the second communication device through the n channels other than the m channels in the K channels.
[0045] In some possible implementation manners, if the first communication device receives an abnormal signal from the m channels in the K channels, the transceiver is specifically configured to: switch the K channels to a second signal transmission mode to perform signal transceiving with the second communication device. After the K channels are switched to the second signal transmission mode, the signal transmission rate between the first communication device and the second communication device is less than the signal transmission rate of the first signal transmission mode.
[0046] In a third aspect, an embodiment of the present application provides a signal transmission mode adjustment method. Specifically, in a normal working state, a first communication device performs signal transceiving with a second communication device in a first signal transmission mode. The first communication device and the second communication device include K channels, K is an integer greater than 1, and the first signal transmission mode is that the first communication device performs transceiving of a first modulation format signal with the second communication device through the K channels. If the first communication device receives an abnormal signal from m channels in the K channels, a control server receives a first message from the first communication device, the first message being used to indicate that the first communication device receives an abnormal signal from the m channels in the K channels, and 1≤m≤K. The control server sends a second message to the first communication device according to the first message, and sends a third message to the second communication device according to the first message. The second message is used to instruct the first communication device to switch the m channels to a second signal transmission mode to perform signal transceiving with the second communication device. The third message is used to instruct the second communication device to switch the m channels to the second signal transmission mode to perform signal transceiving with the first communication device. 1≤m≤K, and after the m channels are switched to the second signal transmission mode, the signal transmission rate between the first communication device and the second communication device is less than the signal transmission rate of the first signal transmission mode.
[0047] In a fourth aspect, an embodiment of the present application provides a control server. The control server comprises a transceiver. The transceiver is configured to receive a first message from a first communication device, the first message being used to indicate that the first communication device receives abnormal signals from m channels of K channels, 1≤m≤K; send a second message to the first communication device according to the first message, and send a third message to a second communication device according to the first message. The second message is used to instruct the first communication device to switch the m channels to a second signal transmission mode for signal transmission with the second communication device. The third message is used to instruct the second communication device to switch the m channels to the second signal transmission mode for signal transmission with the first communication device. Wherein, 1≤m≤K, and a signal transmission rate between the first communication device and the second communication device after the m channels are switched to the second signal transmission mode is less than a signal transmission rate of the first signal transmission mode.
[0048] In a fifth aspect, an embodiment of the present application provides a chip. The chip comprises a processor. The processor is configured to execute the method according to any one of the embodiments of the first aspect and the third aspect.
[0049] In a sixth aspect, an embodiment of the present application provides a communication device. The communication device comprises a processor and an interface. The interface is configured to send a received signal to the processor and send a signal from the processor. The processor is configured to execute the method according to any one of the embodiments of the first aspect.
[0050] In a seventh aspect, an embodiment of the present application provides an optical module. The optical module comprises a processor and an interface. The interface is configured to send a received signal to the processor and send a signal from the processor. The processor is configured to execute the method according to any one of the embodiments of the first aspect.
[0051] In an eighth aspect, an embodiment of the present application provides a control server. The control server comprises a processor and an interface. The interface is configured to send a received signal to the processor and send a signal from the processor. The processor is configured to execute the method according to any one of the embodiments of the second aspect.
[0052] In a ninth aspect, an embodiment of the present application provides a communication system. The communication system comprises a plurality of communication devices. The communication devices communicate with each other. Each of the communication devices is configured to execute the method according to any one of the embodiments of the first aspect.
[0053] In some possible implementation manners, the communication system further comprises a control server. The control server is configured to execute the method according to any one of the embodiments of the second aspect.
[0054] In a tenth aspect, the present application provides a computer readable storage medium. The computer readable storage medium stores instructions. When the instructions are executed by a computer, the method according to any one of the embodiments of the first aspect and the third aspect is implemented.
[0055] In a eleventh aspect, the present application provides a computer program product, which comprises program instructions for implementing the method of any of the embodiments of the first aspect and the third aspect when the computer program product is executed. BRIEF DESCRIPTION OF DRAWINGS
[0056] Fig. 1 is a schematic diagram of a cluster network architecture applied in the embodiments of the present application;
[0057] Fig. 2 is a schematic diagram of an embodiment of optical module for optical signal transmission;
[0058] Fig. 3 is a schematic diagram of an embodiment of the adjustment method of signal transmission mode in the embodiments of the present application;
[0059] Fig. 4 is a schematic diagram of a first scenario of switching signal transmission mode in the embodiments of the present application;
[0060] Fig. 5 is a schematic diagram of a second scenario of switching signal transmission mode in the embodiments of the present application;
[0061] Fig. 6 is a schematic diagram of a third scenario of switching signal transmission mode in the embodiments of the present application;
[0062] Fig. 7 is a schematic diagram of a fourth scenario of switching signal transmission mode in the embodiments of the present application;
[0063] Fig. 8 is a schematic diagram of a scenario of self-negotiation between communication devices in the embodiments of the present application;
[0064] Fig. 9 is a schematic diagram of a possible structure of communication device or optical module in the embodiments of the present application;
[0065] Fig. 10 is a schematic diagram of another possible structure of communication device or optical module in the embodiments of the present application. DETAILED DESCRIPTION
[0066] The embodiments of the present application provide an adjustment method, device and system of signal transmission mode, which can effectively deal with the situation of abnormal channel, reduce the influence of abnormal channel on signal transmission, and improve the reliability of signal transmission.
[0067] It should be noted that the terms "first", "second", and the like, if any, in the description and claims of this application and in the above description of the drawings merely specify the names of similar objects, and do not necessarily require or imply any specific order or sequence of these similar objects. It should be understood that the use of such terms herein is merely to distinguish similar objects from one another, and does not necessarily require or imply any specific order or sequence of these similar objects. It should be understood that the data thus used can be interchanged, where appropriate, so that the embodiments described herein can be carried out in sequences other than those illustrated or described herein. Furthermore, the terms "comprise" and "have", as well as any variations thereof, are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or apparatus that includes a list of steps or units need not be limited to those steps or units that are clearly listed, but can include other steps or units that are not clearly listed or inherent to such processes, methods, products, or apparatuses.
[0068] FIG. 1 is a schematic diagram of a cluster network architecture applied in embodiments of the present application. As shown in FIG. 1, it is a cluster network architecture based on optical signal transmission, which can include a control server, an interconnection device, a computing device management switch, etc. The control server is used to control the allocation, scheduling and management of computing tasks of the entire cluster network. The interconnection device is used to realize the interconnection and intercommunication between computing devices, mainly used to realize the exchange of parameter plane data between computing devices. The management switch is used to realize the transmission of management and control plane data between devices in the cluster network. The computing device is used to execute computing tasks.
[0069] Specifically, the control server mainly includes a global management module, a computing task global allocation module, and an input / output module (I / O module) embedded with a data link monitoring and control module. In addition to including a standard I / O module, the interconnected device also includes a management module and a switching module embedded with a data link monitoring and control module. It should be understood that the data link monitoring and control module can also not be integrated with the switching module of the interconnected device, but embedded into the I / O module of the interconnected device, and the specific implementation can be flexibly processed as needed. In addition to including standard computing chips and I / O modules, the computing device also includes a data link monitoring and control module, a management module, and a computing task allocation module. The computing chips include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), a tensor processing unit (TPU), etc. It should be understood that the data link monitoring and control module can also be integrated with the I / O module of the computing device. Among them, the data link monitoring and control module in each device is used to monitor the link and data transceiving state of the I / O port in real time, such as monitoring the loss of signal (LoS) alarm, the loss of frame (LoF) alarm, the received signal strength indicator (RSSI), the bit error ratio (BER), etc.
[0070] It should be noted that the cluster network architecture shown in FIG. 1 can be specifically applied to a new type of data center for artificial intelligence (AI) large model computing. Unlike traditional general-purpose computing data centers, AI large model cluster computing requires a large number of computing cards, and the computing chips are interconnected through a large number of high-speed optical modules. The optical module used for optical interconnection of the AI cluster computing data center has an extremely high speed, and the optical module is applied in a concentrated manner and in a large quantity. After the channel rate of the optical module currently used in the mainstream data center exceeds 50 gigabits per second (Gbps), in order to reduce the bandwidth requirement for the device, pulse amplitude modulation 4-level (PAM4) is usually used. Although PAM4 modulation can reduce the bandwidth specification requirement by half, due to the small level interval of PAM4 modulation, there are "three eyes" in the eye diagram after modulation, and the modulation is more sensitive to multiple path interference (MPI) caused by reflection in the optical fiber link. On the other hand, during the engineering deployment stage of the optical fiber infrastructure of the data center, due to the complex and large number of optical fiber connection wiring, problems such as end face contamination, scratches, and reflections caused by non-standard construction often occur. Under the combined action of the above factors, the AI cluster computing data center has a high frequency of interruption due to optical interconnection problems, and the computing chips used in the AI cluster computing are mutually coupled in the computing task, and are closely linked. Once a computing error or interruption occurs in a certain link, the entire computing task will be affected.
[0071] It should be understood that the above-mentioned devices in the cluster network architecture shown in Figure 1 are all configured with optical modules, and the interaction between the devices is realized by transmitting optical signals through the optical modules. Figure 2 is a schematic diagram of an embodiment of optical signal transmission of an optical module. As shown in Figure 2, taking the optical signal transmission between two 400G SR4 optical modules as an example, optical module 1 receives multiple data streams from switch chip 1. After digital signal processing (DSP), the multiple data streams are modulated by a driving array and a light source array to obtain multiple optical signals, which are coupled into multiple optical fibers and transmitted to optical module 2. The photo detector (PD) array of optical module 2 converts the received optical signals into electrical signals and outputs the electrical signals to a trans-impedance amplifier (TIA) array. After DSP, the multiple data streams are sent to switch chip 2. Similarly, the way optical module 2 transmits optical signals to optical module 1 is also similar. As can be seen, there are four channels between the two optical modules shown in Figure 2. The optical module sends four optical signals to the opposite end through the four channels, and receives four optical signals from the opposite end through the four channels, that is, each channel supports the transmission of optical signals in two directions. Since the optical modules rely on multi-channel parallel transmission, the data transmitted in multiple channels are associated, and the failure or abnormality of any channel will cause the optical module to be unable to completely recover the original data according to the received optical signals, especially in the above-mentioned AI cluster computing data center scenario. The problem occurs more frequently.
[0072] To this end, an embodiment of the present application provides a signal transmission mode adjustment method. When any one of the multiple parallel transmission optical modules fails or abnormally, the signal transmission mode of the optical module can be adjusted for speed reduction processing, thereby effectively dealing with the abnormality of the channel. Although part of the transmission rate is sacrificed, the normal transmission of the signal is still guaranteed, the influence of the channel abnormality on the signal transmission is reduced, and the reliability of the signal transmission is improved.
[0073] Figure 3 is a schematic diagram of an embodiment of a signal transmission mode adjustment method in an embodiment of the present application. As shown in Figure 3, the embodiment includes the following steps. It should be understood that the communication device in this embodiment can be an optical module, or the communication device can be a device including an optical module and other modules, for example, the interconnection device and the computing device in the scenario shown in Figure 1. It should also be understood that this embodiment is introduced by taking the first communication device detecting the abnormality of the received signal first as an example. If the second communication device detects the abnormality of the received signal first, a similar implementation is also used, which will not be introduced one by one here.
[0074] 101、The first communication device adopts a first signal transmission mode to perform signal transmission and reception with the second communication device.
[0075] Specifically, the first communication device and the second communication device have K channels, K is an integer greater than 1. Wherein each channel here refers to a bidirectional communication channel containing receiving and sending, that is, each channel contains a sending channel and a receiving channel. For example, the channel here can be a physical channel such as an optical fiber. For another example, the channel here can also be a wavelength channel, that is, K different wavelength signals are transmitted in K channels respectively, and the K different wavelength signals can be coupled into the same optical fiber for transmission by wavelength division multiplexing. The first signal transmission mode refers to that the first communication device transmits and receives the first modulation format signal with the second communication device through the K channels, that is, the first modulation format signal is transmitted in each channel, the data sending end is modulated by the first modulation format, and the data receiving end is demodulated by the first modulation format. The application does not limit the specific type of the first modulation format signal, for example, the first modulation format signal can be a PAM4 signal. And in the first signal transmission mode, the data transmitted in the K channels are associated, that is, the data to be transmitted is first subjected to joint forward error correction (FEC) encoding, and each symbol in the FEC codeword after encoding is distributed to K channels for transmission by round robin, for example, each symbol includes 10 bits. In this mode, the data of the K channels is coupled to each other, so the receiving side of the opposite end needs to receive the data of the K channels and then perform joint FEC decoding. This scenario can be called a bonding mode or a mutually associated coupling mode. As an example, 400G data streams are transmitted through K=4 channels, and each symbol in the FEC codeword after KP4 FEC encoding is distributed to 4 channels by round robin, and the data receiving end needs to combine the data received in the 4 channels to perform FEC joint decoding to completely recover the original data sent by the sending end. Another mode different from the above bonding mode is called a breakout mode or a mutually independent mode, and the data stream transmitted in each channel is independently FEC encoded, that is, the data transmitted in each channel is independent and has no association. Even if some channels are abnormal, it will not affect the data receiving end to receive the data stream and perform FEC decoding from other normal channels. For example, the bonding mode of 4 channels transmitting 400G data streams is switched to the breakout mode of 4 channels each transmitting 100G data streams.
[0076] 102、The first communication device detects the signals received by the K channels.
[0077] The first communication device detects the data receiving condition of each receiving port in real time. The detected parameters include, but are not limited to, the modulation format of the signal, LoS alarm, LoF alarm, RSSI, packet loss, uncorrectable error, etc. As an example, if the first communication device is an optical module, the corresponding modules for detecting the above parameters can be integrated in the optical module, for example, the optical module is used to detect the modulation format of the signal, LoS alarm, RSSI and uncorrectable error. As another example, if the first communication device is a device including an optical module and other modules, the above various parameters can be detected by the other modules in the first communication device except the optical module, for example, the data link monitoring and control module in the interconnection device and the computing device in the above Fig. 1 is used to detect LoF alarm, packet loss and uncorrectable error.
[0078] 103. The first communication device determines whether the signal received from the K channels is abnormal.
[0079] Specifically, taking one of the channels as an example, the abnormality of the signal received from the channel by the first communication device includes, but is not limited to, at least one of the following situations. 1. The number of LoS alarms of the signal received from the channel is greater than or equal to a first threshold value. 2. The number of LoF alarms of the signal received from the channel is greater than or equal to a second threshold value. 3. The number of packet losses of the signal received from the channel is greater than or equal to a third threshold value. 4. The number of uncorrectable errors of the signal received from the channel is greater than or equal to a fourth threshold value. It should be understood that the application does not limit the specific values of the above-mentioned first threshold value, second threshold value, third threshold value and fourth threshold value, for example, each of the above-mentioned threshold values can take the value of 3.
[0080] It should be noted that for the above case 1, the signal loss (LoS) alarm can be caused by fiber breakage, loose or virtual connection of the joint, failure of the opposite end laser, etc. If the number of continuous LoS alarms is greater than or equal to the first threshold, it means that there is no false detection, and it is determined that the received signal is abnormal. For the above case 4, if the number of error bits or bytes exceeds the error correction capability of the FEC during FEC decoding, it means that an uncorrectable error has occurred. The size of the data subjected to FEC decoding and the error correction capability of the FEC depend on the code type of the FEC encoding. If the number of continuous uncorrectable errors is greater than or equal to the fourth threshold, it can be determined that the received signal is abnormal. For the above case 2, after the data stream is recovered by FEC decoding, the boundaries of the data frames can be determined by frame synchronization. The data frame loss (LoF) can be caused by LoS or uncorrectable error. If the number of continuous LoF alarms is greater than or equal to the second threshold, it can be determined that the received signal is abnormal. For the above case 3, a data packet usually includes multiple data frames. By detection, it can be determined whether there is a loss of data packets, i.e., packet loss. The packet loss can be caused by LoS or uncorrectable error. If the number of continuous packet losses is greater than or equal to the third threshold, it can be determined that the received signal is abnormal.
[0081] 104. The first communication device adopts a second signal transmission mode to perform signal transceiving with the second communication device.
[0082] In this embodiment, if the first communication device receives an abnormal signal from at least one of the K channels, the first communication device makes an adjustment and adopts a second signal transmission mode to perform signal transceiving with the second communication device. It should be noted that before switching to the second signal transmission mode, the first communication device will first switch the data transmission of the K channels from the bonding mode to the breakout mode. Since the data transmitted by each channel is independent in the breakout mode, the normal transmission of other channels can be avoided from being affected by the abnormal channel, and the signal transmission mode of the abnormal channel can be adjusted accordingly. Furthermore, after switching to the second signal transmission mode, if the data of each channel can be normally transmitted based on the second signal transmission mode and no longer abnormal, the data transmission of the K channels can adopt the bonding mode or the breakout mode, which is not limited here. For example, based on the second signal transmission mode, the data transmission between channels is correlated. For another example, based on the second signal transmission mode, the data transmission between channels is independent. For another example, based on the second signal transmission mode, the data transmission between a part of channels is correlated, and the data transmission between another part of channels is independent.
[0083] It should be understood that after switching from the first signal transmission mode to the second signal transmission mode, the signal transmission rate between the first communication device and the second communication device is reduced, which is equivalent to a signal transmission mode with a speed reduction process, and part of the transmission rate is sacrificed to ensure normal signal transmission. Among them, this speed reduction process can have many different implementations, and the following will take K=4 as an example to introduce respectively.
[0084] FIG. 4 is a schematic diagram of a first switching signal transmission mode scenario in the embodiment of the application. As shown in FIG. 4, in the first signal transmission mode, the four channels between the first communication device and the second communication device transmit the first modulation format signal, and the signal transmission of the four channels adopts the bonding mode. The first communication device receives abnormal signals from channel 1. After switching to the second signal transmission mode, the first communication device and the second communication device transmit the second modulation format signal to each other through channel 1, and the remaining three channels still transmit the first modulation format signal. As an example, the first modulation format signal is a PAM4 signal, and the second modulation format signal is a two-level pulse amplitude modulation (PAM2) signal or a non-return to zero (NRZ) signal. In a scenario where the channel quality is not good, using a PAM2 signal or an NRZ signal reduces the decision error compared to using a PAM4 signal, which is beneficial to reduce the bit error rate. It is equivalent to that the transmission rate of channel 1 is halved, and the transmission rates of other channels remain unchanged. For example, the transmission rate is 400G in the first signal transmission mode, and the transmission rate is 350G in the second signal transmission mode.
[0085] FIG. 5 is a schematic diagram of a second switching signal transmission mode scenario in the embodiment of the application. As shown in FIG. 5, in the first signal transmission mode, the four channels between the first communication device and the second communication device transmit the first modulation format signal, and the signal transmission of the four channels adopts the bonding mode. The first communication device receives abnormal signals from channel 1. After switching to the second signal transmission mode, the first communication device and the second communication device close the signal transmission of channel 1, and the remaining three channels still transmit the first modulation format signal. For example, the transmission rate is 400G in the first signal transmission mode, and the transmission rate is 300G in the second signal transmission mode.
[0086] Fig. 6 is a schematic diagram of a third scenario of switching the signal transmission mode according to an embodiment of the present application. As shown in Fig. 6, in the first signal transmission mode, the four channels between the first communication device and the second communication device all transmit the first modulation format signal, and the signal transmission of the four channels adopts the bonding mode. The first communication device receives abnormal signals from channel 1. After switching to the second signal transmission mode, the four channels between the first communication device and the second communication device all transmit the second modulation format signal. For example, the first modulation format signal is a PAM4 signal, and the second modulation format signal is a PAM2 signal or an NRZ signal, which is equivalent to halving the transmission rate of each channel. If the transmission rate in the first signal transmission mode is 400G, then the transmission rate in the second signal transmission mode is 200G.
[0087] Fig. 7 is a schematic diagram of a fourth scenario of switching the signal transmission mode according to an embodiment of the present application. As shown in Fig. 7, in the first signal transmission mode, the four channels between the first communication device and the second communication device all transmit the first modulation format signal, and the signal transmission of the four channels adopts the bonding mode. The first communication device receives abnormal signals from channel 1 and channel 2. After switching to the second signal transmission mode, the first communication device and the second communication device close the signal transmission of channel 1, and transmit the second modulation format signal to each other through channel 2, and the remaining two channels still transmit the first modulation format signal. For example, the first modulation format signal is a PAM4 signal, and the second modulation format signal is a PAM2 signal or an NRZ signal, which is equivalent to closing channel 1, halving the rate of channel 2, and keeping the transmission rate of the other channels unchanged. If the transmission rate in the first signal transmission mode is 400G, then the transmission rate in the second signal transmission mode is 250G. It should be noted that, for the multi-channel abnormal scenario shown in Fig. 7, in addition to the implementation of the second signal transmission mode provided herein, the implementation of Figs. 4 to 6 can also be flexibly transformed and combined, which will not be described herein again.
[0088] It should be understood that the present application does not limit the specific types of the first modulation format and the second modulation format, as long as the rate can be reduced by changing the modulation format. For example, it can also be from a PAM8 signal to a PAM4 signal, and it can also be from a PAM2 signal or an NRZ signal to a modulation format signal with a lower transmission rate, and the like.
[0089] It should be noted that switching between the first communication device and the second communication device to the second signal transmission mode can be applied in a plurality of different scenarios. For example, one possible scenario is that the first communication device and the second communication device switch to the second signal transmission mode through self-negotiation. For another example, another possible scenario is that the first communication device or the second communication device reports the reception anomaly to the control server, and the control server controls the first communication device and the second communication device to switch to the second signal transmission mode. The two possible scenarios will be introduced below.
[0090] In the first possible scenario, in a normal working state, the first communication device and the second communication device adopt the first signal transmission mode for signal transceiving. If the first communication device finds that the reception signal from the m channels of the K channels is abnormal, the first communication device first adjusts the working mode of the sending end, that is, switches to the second signal transmission mode to send signals to the second communication device. Since the first communication device switches to the second signal transmission mode, the second communication device based on the previous first signal transmission mode for receiving signals will also be abnormal, so the second communication device will also change the working mode of the sending end synchronously, that is, switches to the second signal transmission mode to send signals to the first communication device. Further, the first communication device detects that the sending end of the second communication device has also changed the working mode, thereby determining that both sides have switched to the new working mode. That is, the first communication device that detects the reception signal anomaly first will first adjust the working mode of the sending end, which is equivalent to initiating a negotiation request for the working mode switching to the second communication device. The second communication device receives the negotiation request from the first communication device and also adjusts the working mode of the sending end accordingly, which is equivalent to feeding back to the first communication device that the negotiation request has been agreed. The first communication device based on the signal transmission mode switching of the second communication device can confirm that the second communication device has agreed to the negotiation request, which is equivalent to completing the negotiation process between the two sides.
[0091] Figure 8 is a schematic diagram of a scenario of self-negotiation between communication devices in the embodiments of the present application. As shown in Figure 8, in any of the first communication device and the second communication device, a control module is configured to control a plurality of modules according to information fed back by the modules. Specifically, the control module determines whether to use the first signal transmission mode or the second signal transmission mode for signal transmission. For example, the control module determines that the channels use the bonding mode, and then controls the FEC encoding module to use joint FEC encoding for data streams to be transmitted by the channels and to use joint FEC decoding for data streams received by the channels. For another example, the control module determines that the channels use the breakout mode, and then controls the FEC encoding module to use independent FEC encoding for data streams to be transmitted by the channels and to use independent FEC decoding for data streams received by the channels. For another example, the control module can control the modulation selection module to switch the signal modulation format, so that the first modulation format or the second modulation format can be used to modulate data to be transmitted by the channels, and the control module can also control the demodulation selection module to switch the signal demodulation mode, so that the demodulation mode corresponding to the first modulation format or the second modulation format can be used to demodulate signals received by the channels. For another example, the control module can control whether each optical transmitting unit and each optical receiving unit works, so as to adjust the opening or closing of the channels.
[0092] The FEC encoding module is configured to perform FEC encoding on input data. The FEC decoding module is configured to perform FEC decoding on input data, and is further configured to perform statistics on uncorrectable errors and feed back to the control module. The optical signal detection module is configured to detect received signals and feed back to the control module, for example, to detect RSSI or LoS, etc. The modulation format detection module is configured to detect modulation formats of received signals and feed back to the control module, for example, to detect that the received signal is a PAM4 signal or an NRZ signal, etc. The reporting module is configured to report information to a control server. The modulation selection module is configured to select a corresponding modulation format to modulate data to be transmitted according to an instruction of the control module. The demodulation selection module is configured to select a corresponding demodulation manner to demodulate received signals according to an instruction of the control module. The digital-to-analog conversion module is configured to convert digital signals into analog signals. The analog-to-digital conversion module is configured to convert analog signals into digital signals. The optical transmitting unit is configured to convert electrical signals into optical signals and transmit the optical signals. The optical receiving unit is configured to receive optical signals and convert the optical signals into electrical signals. As an example, all units or modules in the communication device in FIG. 8 can be integrated in an optical module. As another example, part of the units or modules in the communication device in FIG. 8 are integrated in an optical module, and another part of the units or modules can be integrated in other modules outside the optical module, for example, the FEC encoding module and the FEC decoding module in FIG. 8 can be integrated in the data link monitoring and control module shown in FIG. 1, and the control module in FIG. 8 can be integrated in the management module shown in FIG. 1.
[0093] As an example, as shown in FIG. 8, based on the first signal transmission mode, the control module of the first communication device controls the partial modules, so that the four optical transmitting units of the first communication device respectively transmit four paths of the first modulation format signals to the second communication device through the four channels, and the control module of the second communication device controls the modules, so that the four optical transmitting units of the second communication device respectively transmit four paths of the first modulation format signals to the first communication device through the four channels, that is, the first modulation format signals are transmitted and received between the first communication device and the second communication device through the four channels. Then, the control module of the first communication device determines that the signal received from the channel 4 is abnormal according to the feedback of the partial modules, and decides to send signals to the second communication device by using the second signal transmission mode, and controls the corresponding modules to change the working mode. The specific implementation mode of the second signal transmission mode can adopt any one of the modes introduced in FIGS. 4 to 7. Here, taking the control of the control module of the first communication device to control the modulation selection module and the demodulation selection module corresponding to the channel 4 to switch to the second modulation format as an example, then the optical transmitting unit 1 to the optical transmitting unit 3 of the first communication device still transmit the first modulation format signals, and the optical transmitting unit 4 of the first communication device transmits the second modulation format signals. The demodulation selection module corresponding to the channel 4 in the second communication device cannot normally demodulate the second modulation format signals from the channel 4 according to the previous demodulation mode corresponding to the first modulation format, the modulation format detection module of the second communication device detects that the signal received by the channel 4 adopts the second modulation format, and the control module of the second communication device also controls the modulation selection module and the demodulation selection module corresponding to the channel 4 to switch to the second modulation format, then the optical transmitting unit 1 to the optical transmitting unit 3 of the second communication device still transmit the first modulation format signals, and the optical transmitting unit 4 of the second communication device transmits the second modulation format signals. Further, the modulation format detection module of the first communication device detects that the signal received by the channel 4 adopts the second modulation format, so that the control module of the first communication device can determine that both sides have switched to the second signal transmission mode for signal transmission and reception, indicating that the both sides have completed negotiation, and the control module of the first communication device controls the reporting module to report the result of the negotiation of the both sides to the control server.
[0094] In the second possible scenario, in the normal working state, the first communication device and the second communication device adopt the first signal transmission mode for signal transmission and reception. If the first communication device finds that the signals received from m channels of K channels are abnormal, the first communication device reports the abnormal information to the control server, and waits for the instruction of the control server. The specific steps can be referred to the introduction of steps 105 and 106.
[0095] 105. The first communication device sends a first message to the control server.
[0096] The first communication device sends a first message to the control server, which informs the control server that the first communication device detects that the signals received from the m channels are abnormal. In addition, the first message can also carry specific detection parameters reflecting the abnormality to help the control server make decision analysis. As an example, in the scenario shown in FIG. 1, the first communication device and the second communication device are interconnected devices, and the first message sent by the management module in the first communication device is transmitted to the global management module of the control server through the management switch.
[0097] 106. The control server sends a second message to the first communication device and the second communication device.
[0098] After receiving the first message reported by the first communication device, the control server determines through analysis that the first communication device and the second communication device are to switch to the second signal transmission mode for signal transmission and reception. The specific implementation of the second signal transmission mode can adopt any of the modes introduced in FIGS. 4-7, which will not be repeated here. Further, the control server sends a second message to the first communication device and the second communication device, which informs the first communication device and the second communication device to adjust the working mode of their respective sending and receiving ends, so as to realize the signal transmission and reception of the first communication device and the second communication device in the second signal transmission mode.
[0099] 107. The first communication device determines whether the signals received from the K channels are abnormal.
[0100] In some possible scenarios, after the first communication device and the second communication device switch to the second signal transmission mode for signal transmission and reception, the first communication device further determines whether the signals received from the K channels are abnormal through real-time detection. The specific mode is similar to the above step 103, which will not be repeated here.
[0101] 108. The first communication device sends a confirmation message to the control server.
[0102] If the first communication device determines through detection that the signals received from each channel are no longer abnormal, it means that the first communication device and the second communication device can normally transmit signals based on the second signal transmission mode. Then, the first communication device sends a confirmation message to the control server, which confirms to the control server that the first communication device and the second communication device successfully complete the switching from the first signal transmission mode to the second signal transmission mode. As an example, in the scenario shown in FIG. 1, the global management module of the control server receives the confirmation message sent by the first communication device and synchronizes the confirmation message of the signal transmission mode switching between the first communication device and the second communication device to the computing task global allocation module. The computing task global allocation module adjusts the global allocation of the computing task according to the confirmation message and sends it to each device for execution.
[0103] Some possible embodiments are introduced below in combination with specific scenarios in which the received signals from the channels appear abnormal. It should be understood that the following embodiments are all introduced by taking the scenario in which the first communication device and the second communication device perform self-negotiation as an example. Based on the above introduction, the first communication device can also report to the control server and work according to the instructions issued by the control server. This mode is not introduced here in detail.
[0104] Embodiment 1: The abnormality of the received signals is LoS.
[0105] Taking the scenario shown in FIG. 8 as an example, the first communication device performs RSSI detection on the signals received from the four channels and finds that the number of LoS alarms of the signals received from channel 4 is greater than or equal to the first threshold, thereby judging that channel 4 may be caused by signal loss problems such as fiber breakage, loose or virtual connection of the joint, failure of the laser at the opposite end, etc. Based on the scenario of signal loss abnormality, the modulation format of the modulated signal cannot be recovered for normal transmission, and therefore, the first communication device suspends the work of the optical transmitting unit 4, that is, closes channel 4 and no longer sends signals to the second communication device through channel 4. Correspondingly, the second communication device also finds that the number of LoS alarms of the signals received from channel 4 is greater than or equal to the first threshold, and the second communication device also closes channel 4 and no longer sends signals to the first communication device through channel 4.
[0106] Embodiment 2: The abnormality of the received signals is uncorrectable error code.
[0107] Taking the scenario shown in FIG. 8 as an example, the first communication device performs detection on the signals received from the four channels and finds that the number of uncorrectable error codes of the signals received from channel 4 is greater than or equal to the fourth threshold, and finds that the RSSI of the signals received from the four channels is normal, thereby judging that the received optical power of channel 4 is normal and the uncorrectable error code may be caused by signal reflection or other reasons. In addition, the first communication device also needs to detect and record the change of the signal modulation format of each channel before and after the abnormality in real time, for judging whether the first communication device or the second communication device appears abnormal first; for example, if the first communication device detects the first modulation format signal before and after the reception of the abnormality, it means that the first communication device appears abnormal first (the second communication device at the opposite end has not detected the abnormality). If the first communication device detects the first modulation format signal before the reception of the abnormality and detects the second modulation format signal after the reception of the abnormality, it means that the second communication device at the opposite end appears the reception abnormality first, which triggers the second communication device to actively change and adjust the modulation format of the signal transmitted by channel 4, thereby causing the uncorrectable error code of the signals received by the first communication device from channel 4.
[0108] The following is an example of the first communication device first appearing to receive an exception: The first communication device detects the first modulation format signal before and after the reception exception occurs. According to the first communication device, it is judged that the first communication device first appears to receive an exception. Therefore, the first communication device adjusts the signal transmitted by channel 4 to the second modulation format, and the signals transmitted by the remaining three channels remain in the first modulation format. Correspondingly, since the second communication device still demodulates the signal from channel 4 of the first communication device in the first modulation format, the second communication device channel 4 will appear to receive an exception. The second communication device determines that the receiving end has appeared to receive an exception and actively adjusts the modulation format of channel 4 by checking the modulation format change before and after the reception exception (from the first modulation format to the second modulation format). Therefore, the second communication device also adjusts the transmission and reception of channel 4 to the second modulation format, and the signals transmitted by the remaining three channels remain in the first modulation format. After the first communication device adjusts channel 4 to the second modulation format, it will continue to detect the received signal of channel 4. If it is detected within a specified time that the signal of channel 4 from the second communication device has been adjusted to the second modulation format and no longer appears to receive an exception, it means that both sides can transmit and receive signals in this way. If the signal of channel 4 from the second communication device has not been adjusted to the second modulation format within a specified time or has been adjusted to the second modulation format but still appears to receive an exception, i.e., the number of uncorrectable errors is greater than or equal to the fourth threshold value, it means that the previous adjustment has not been effective. Therefore, the first communication device will further adjust the signal transmission mode of channel 4, and the second communication device will synchronously adjust the signal transmission mode of channel 4 according to the adjustment mode of the first communication device until the first communication device detects that the received signal no longer appears to receive an exception. For example, the first communication device adjusts the transmission and reception signals of channel 4 to the third modulation format. Correspondingly, the second communication device also adjusts the transmission and reception signals of channel 4 to the third modulation format, which is equivalent to reducing the signal transmission rate by further changing the modulation format. For another example, the first communication device closes channel 4 and no longer transmits and receives signals with the second communication device through channel 4. Correspondingly, the second communication device also closes channel 4 and no longer transmits and receives signals with the first communication device through channel 4.
[0109] The following describes the possible structure of a communication device or optical module for performing the adjustment method of the above signal transmission mode.
[0110] FIG. 9 is a schematic diagram of a possible structure of a communication device or an optical module according to an embodiment of the present application. As shown in FIG. 9, the communication device or the optical module includes a processing unit 201 and a transceiver unit 202. The processing unit 201 is configured to perform the steps 102, 103 and 107 in the embodiment of FIG. 3, and the transceiver unit 202 is configured to perform the steps 101, 104, 105, 106 and 108 in the embodiment of FIG. 3. It should be understood that the communication device or the optical module according to the embodiments of the present application can also be implemented in other manners. For example, the division of the units in the communication device or the optical module is merely a logical function division, and other division manners can be adopted during actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some units can be separated. In addition, the various embodiments of the present application can be implemented in a form of a processing unit integrating each function unit or each function unit separated, or two or more function units integrated in a processing unit. The above-mentioned integrated unit can be implemented in a form of hardware or software function unit.
[0111] FIG. 10 is a schematic diagram of another possible structure of a communication device or an optical module according to an embodiment of the present application. As shown in FIG. 10, the communication device or the optical module includes a processor 301 and an interface circuit 302. It should be understood that the interface circuit 302 can be a transceiver or an input / output interface, and the interface circuit 302 is configured to receive a signal from another device outside the communication device or the optical module and transmit the signal to the processor 301, or transmit a signal from the processor 301 to another device outside the communication device or the optical module. Specifically, the processor 301 is configured to perform the steps 102, 103 and 107 in the embodiment of FIG. 3, and the interface circuit 302 is configured to perform the steps 101, 104, 105, 106 and 108 in the embodiment of FIG. 3. Optionally, the communication device or the optical module can further include a memory 303, where the memory 303 is configured to store program instructions and data.
[0112] The embodiments of the present application further provide a chip. The chip includes one or more interface circuits, and further integrates a processing circuit configured to implement the functions of the processor 301. As an example, the chip integrates a memory. As another example, when the chip does not integrate the memory, the chip can be connected with an external memory through an interface. The chip can complete the method steps of any one or more of the preceding embodiments. Alternatively, the chip implements the actions performed by the data processing apparatus in the embodiments of the present application according to program codes stored in the memory.
[0113] The embodiments of the present application further provide a computer-readable storage medium including a program or instructions, which, when executed on a computer, cause the method performed by the processor 301 in the method embodiments described above.
[0114] It should be understood that the processor mentioned in the embodiments of the present application can be implemented by hardware or software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc. When implemented by software, the processor can be a general-purpose processor which is implemented by reading software code stored in a memory. The memory can exist independently of the processor, or the memory can be integrated with the processor.
[0115] As an example, the processor in the embodiments of the present application can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor, and can also be a processing circuit implementing specific functions.
[0116] The memory in the embodiments of the present application can be a random access memory (RAM), a flash memory, a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), a register, a hard disk, a mobile hard disk, a CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor, so that the processor can read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a network device or a terminal device. Of course, the processor and the storage medium can also exist as discrete components in the network device or the terminal device.
[0117] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof.
[0118] When implemented by using hardware, the data processing method provided by the embodiments of the present application can be implemented without reading software codes or instructions, for example, by using a CPU, a DSP, an ASIC, an FPGA, other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.
[0119] When implemented by using software, the data processing method provided by the embodiments of the present application can be implemented in the form of a computer program product in whole or in part. The computer program product includes one or more computer programs or instructions. When the computer programs or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are executed in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a terminal device, or other programmable devices. The computer programs or instructions can be stored in a computer-readable storage medium or transmitted by the computer-readable storage medium. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server integrated with one or more available media. The available medium can be a magnetic medium, for example, a floppy disk, a hard disk, a magnetic tape; an optical medium, for example, a digital versatile disc (DVD); or a semiconductor medium, for example, a solid state disk (SSD).
[0120] Finally, it should be noted that: the above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. A method for adjusting a signal transmission mode, characterized in that, The method comprises: A first communication device transmits and receives signals with a second communication device in a first signal transmission mode, wherein K channels are included between the first communication device and the second communication device, K is an integer greater than 1, and the first signal transmission mode is that the first communication device transmits and receives first modulation format signals with the second communication device through the K channels. If the first communication device receives signals from m channels of the K channels abnormally, the first communication device switches the m channels to a second signal transmission mode to transmit and receive signals with the second communication device, wherein 1≤m≤K, and the signal transmission rate between the first communication device and the second communication device after the m channels are switched to the second signal transmission mode is less than the signal transmission rate of the first signal transmission mode.
2. The method of claim 1, wherein, If the first communication device receives signals from the m channels abnormally, the first communication device switches the m channels to a second signal transmission mode to transmit and receive signals with the second communication device, which comprises: The first communication device first switches the m channels to the second signal transmission mode to send signals to the second communication device, and then receives signals sent by the second communication device by switching the m channels to the second signal transmission mode.
3. The method of claim 2, wherein, If the first communication device receives signals sent by the second communication device by switching the m channels to the second signal transmission mode without abnormality, the method further comprises: The first communication device sends an acknowledgement message, which is used to indicate that the first communication device transmits and receives signals with the second communication device in the second signal transmission mode.
4. The method of claim 1, wherein, Before the first communication device switches the m channels to the second signal transmission mode to transmit and receive signals with the second communication device if the first communication device receives signals from the m channels abnormally, the method further comprises: The first communication device sends a first message to a control server, which is used to indicate that the first communication device receives signals from the m channels abnormally; The first communication device receives a second message sent by the control server according to the first message, which is used to indicate that the first communication device switches the m channels to the second signal transmission mode to transmit and receive signals with the second communication device.
5. The method according to any one of claims 1 to 4, characterized in that, Before the first communication device switches the m channels to the second signal transmission mode to transmit and receive signals with the second communication device, the method further comprises: The first communication device detects at least one of the following signals received from the K channels: The first communication device detects a modulation format used by the signals; The first communication device detects the number of loss of signal (LoS) alarms; The first communication device detects the number of loss of frame (LoF) alarms; The first communication device detects the number of packet loss; The first communication device detects the number of uncorrectable errors.
6. The method according to any one of claims 1 to 5, characterized in that, The first communication device receiving abnormal signals from one channel includes at least one of the following: the first communication device receiving LoS alarms from one channel for a number of times greater than or equal to a first threshold; the first communication device receiving LoF alarms from one channel for a number of times greater than or equal to a second threshold; the first communication device receiving packet loss from one channel for a number of times greater than or equal to a third threshold; and the first communication device receiving uncorrectable errors from one channel for a number of times greater than or equal to a fourth threshold.
7. The method according to any one of claims 1 to 6, characterized in that, The data transmitted by the K channels based on the first signal transmission mode is a coupling mode in which the data is correlated with each other. If the first communication device receives abnormal signals from m channels of the K channels, before the first communication device switches the m channels to a second signal transmission mode and performs signal transceiving with the second communication device, the method further includes: the first communication device switching the data transmitted by the K channels to an independent mode.
8. The method according to any one of claims 1 to 7, characterized in that, The first communication device switches the m channels to a second signal transmission mode and performs signal transceiving with the second communication device includes: The first communication device closes the m channels, transmits the first modulation format signal to the second communication device through n channels of the K channels other than the m channels, receives the first modulation format signal transmitted by the second communication device through the n channels, and K = m + n. Alternatively, The first communication device transmits a second modulation format signal to the second communication device through the K channels and receives the second modulation format signal transmitted by the second communication device through the K channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal. Alternatively, The first communication device transmits a second modulation format signal to the second communication device through the m channels, transmits the first modulation format signal to the second communication device through n channels of the K channels other than the m channels, receives the second modulation format signal transmitted by the second communication device through the m channels, and receives the first modulation format signal transmitted by the second communication device through the n channels, wherein the transmission rate based on the first modulation format signal is greater than the transmission rate based on the second modulation format signal, and K = m + n. Alternatively, The first communication device transmits the first modulation format signal to the second communication device through n channels of the K channels except the m channels, closes p channels of the m channels, transmits a second modulation format signal to the second communication device through q channels of the m channels except the p channels, receives the first modulation format signal transmitted by the second communication device through the n channels, and receives the second modulation format signal transmitted by the second communication device through the q channels, wherein the first modulation format signal transmission rate is greater than the second modulation format signal transmission rate, K = m + n, m = p + q, and p and q are both integers greater than or equal to 1.
9. The method of claim 8, wherein, The first modulation format signal is a four-level pulse amplitude modulation (PAM4) signal, and the second modulation format signal is a two-level pulse amplitude modulation (PAM2) signal or a non-return-to-zero (NRZ) signal.
10. The method according to any one of claims 1 to 7, characterized in that, If the number of LoS alarms of the signal received by the first communication device from each of the m channels is greater than or equal to a first threshold value, the first communication device switches the m channels to a second signal transmission mode for signal transceiving with the second communication device, including: The first communication device closes the m channels, transmits the first modulation format signal to the second communication device through n channels of the K channels except the m channels, receives the first modulation format signal transmitted by the second communication device through the n channels, K = m + n; Or, If the number of LoS alarms of the signal received by the first communication device from each of the p channels is greater than or equal to a first threshold value, the first communication device switches the m channels to a second signal transmission mode for signal transceiving with the second communication device, including: The first communication device transmits the first modulation format signal to the second communication device through n channels of the K channels except the m channels, closes p channels of the m channels, transmits a second modulation format signal to the second communication device through q channels of the m channels except the p channels, receives the first modulation format signal transmitted by the second communication device through the n channels, and receives the second modulation format signal transmitted by the second communication device through the q channels, K = m + n, m = p + q, and p and q are both integers greater than or equal to 1.
11. The method according to any one of claims 1 to 7, characterized in that, If the number of uncorrectable errors of the signal received by the first communication device from each of the m channels is greater than or equal to a fourth threshold value, the method further includes: The first communication device detects the modulation format of the signal received from the m channels; If the first communication device receives the first modulation format signal from the m channels, the first communication device transmits a second modulation format signal to the second communication device through the m channels, receives the second modulation format signal transmitted by the second communication device through the m channels, transmits the first modulation format signal to the second communication device through n channels of the K channels other than the m channels, and receives the first modulation format signal transmitted by the second communication device through the n channels, where K = m + n.
12. The method of claim 11, wherein, If the number of uncorrectable errors of the second modulation format signal received by the first communication device from each of the m channels is greater than or equal to a fourth threshold value, the method further comprises: The first communication device closes the m channels, transmits the first modulation format signal to the second communication device through the n channels, and receives the first modulation format signal transmitted by the second communication device through the n channels.
13. The method according to any one of claims 1 to 7, characterized in that, If the number of uncorrectable errors of the signal received by the first communication device from each of the m channels is greater than or equal to a fourth threshold value, the method further comprises: The first communication device detects the modulation format of the signal received by the m channels. If the first communication device receives the second modulation format signal from the m channels, the first communication device transmits the second modulation format signal to the second communication device through the m channels and transmits the first modulation format signal to the second communication device through n channels of the K channels other than the m channels.
14. The method according to any one of claims 1 to 7, characterized in that, If the first communication device receives an abnormal signal from m channels of the K channels, the first communication device switches the m channels to a second signal transmission mode for signal transceiving with the second communication device, which comprises: The first communication device switches the K channels to a second signal transmission mode for signal transceiving with the second communication device, wherein the signal transmission rate between the first communication device and the second communication device after the K channels are switched to the second signal transmission mode is less than the signal transmission rate of the first signal transmission mode.
15. A communication device, characterized by The communication device comprises a transceiving unit, which is configured to: adopt a first signal transmission mode for signal transceiving with a second communication device, wherein the first communication device and the second communication device comprise K channels, the K is an integer greater than 1, and the first signal transmission mode is a first modulation format signal transceiving between the first communication device and the second communication device through the K channels. If the first communication device receives an abnormal signal from m channels of the K channels, the first communication device switches the m channels to a second signal transmission mode, wherein 1 ≤ m ≤ K, and the signal transmission rate between the first communication device and the second communication device after the m channels are switched to the second signal transmission mode is less than the signal transmission rate of the first signal transmission mode.
16. A chip, characterized by The chip comprises a processor configured to execute the method of any one of claims 1 to 14. The chip comprises a processor configured to execute the method of any one of claims 1 to 14.
17. A communication device, characterized by The communication device comprises a processor for performing the method according to any one of claims 1 to 14 and an interface for sending received signals to the processor and sending signals from the processor.
18. A communication system, characterized by The communication system comprises a plurality of communication devices according to claim 17, which communicate with each other.
19. An optical module characterized by comprising: The optical module comprises a processor for performing the method according to any one of claims 1 to 14 and an interface for sending received signals to the processor and sending signals from the processor.