A message processing method, communication device and system
By determining the target rate through sending and receiving control messages between communication devices, the problem of poor processing pressure and scalability caused by packet-by-packet modification of network nodes in the prior art is solved, and high-performance service message transmission and reasonable rate control are achieved.
Patent Information
- Application Number
- CN202510599757.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-09
- Publication Date
- 2026-11-10
AI Technical Summary
In the existing network, the service packet processing strategy requires each network node to modify it packet by packet, which increases the processing pressure, affects performance, has poor scalability, is difficult to deploy, and the transmission rate calculation is unreasonable.
By sending and receiving control messages between communication devices, the target rate is determined, avoiding network nodes from processing service messages packet by packet. By using link information to update the rate, a reasonable transmission rate can be determined.
It achieves high-performance service message processing, reduces the computing and storage burden on network nodes, and improves the reliability of network transmission and bandwidth utilization.
Smart Images

Figure CN123074986A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data communication technology, and in particular to a message processing method, communication device and system. Background Technology
[0002] With the advent of the computing power network era, emerging applications such as geographically distributed machine learning and distributed artificial intelligence (AI) large model training have also developed rapidly. These applications, which rely on the transmission of massive amounts of business packets, have also placed higher demands on network performance.
[0003] In current network service packet processing strategies, each network node from the sender to the receiver needs to include its collected status information or calculated control information in an extended header of the service packet before transmitting it to the receiver. The receiver, upon receiving the service packet with this extended header, then sends it back to the sender. The sender determines the transmission rate based on this extended header and controls the transmission of its service packets accordingly. The aforementioned status and control information can be understood as information that needs to be involved in determining the transmission rate. Therefore, the current strategy is flow-based, requiring network nodes to modify each service packet in the service flow packet on a packet-by-packet basis. This not only increases the processing load on network nodes and affects network performance, but also requires each network node to modify its network protocol stack, resulting in poor scalability and difficult deployment. Summary of the Invention
[0004] Based on this, this application provides a message processing method, communication device, and system that, without affecting the transmission of service messages in the network, efficiently determines the reasonable transmission rate of the service message to be sent in the transmission path by controlling the message, thus making it possible to achieve high-performance service message processing.
[0005] Firstly, this application provides a message processing method applied to a first communication device. The method may include, for example, the first communication device sending a first control message including first link information to a second communication device via a first path. The first link information is used to determine a target rate, which is the maximum allowed transmission rate of the service message to be sent in the first path. Then, the first communication device receives a second control message carrying the target rate sent by the second communication device. In this way, without affecting the transmission of service messages in the network, the first communication device obtains the target rate—that is, the maximum allowed transmission rate of service messages on the first path—by transmitting the first control message along the first path and from the received second control message. This decouples the determination of the target rate of the path from the transmission process of user data in the network, eliminating the need for each network node on the first path to modify the service messages and ensuring normal transmission of service messages in the network. Furthermore, it eliminates the need for each network node on the first path to process service messages packet by packet, making the processing granularity controllable and effectively reducing the consumption of computing and storage resources. Thus, high-performance service message processing becomes possible.
[0006] It should be noted that the target rate in this application is defined as "the maximum allowed transmission rate of the service message to be sent in the first path." Here, "allowed" means that in the process of balancing the highest possible bandwidth utilization and the lowest possible congestion probability, the maximum allowed transmission rate of the service message to be sent on the first path is allowed. It does not completely limit the actual transmission rate of forwarding the service message to be sent. In other words, after determining the target rate, the sending end can send the service message to be sent on the first path at the target rate, or at a transmission rate greater than the target rate, or at a transmission rate less than the target rate. However, sending based on the target rate is more reasonable, while sending based on a transmission rate greater than the target rate may cause congestion, resulting in the service message to be sent not being transmitted effectively, and sending based on a transmission rate less than the target rate may result in low bandwidth utilization.
[0007] In some possible implementations, the first communication device can be the sender of the first path. After the first communication device receives the second control message, the method may further include: the first communication device receiving a first service message, the transmission path of which is the first path; and the first communication device sending the first service message to the second communication device via the first path at a target rate. Thus, after determining the target rate based on the method provided in this application, transmitting the corresponding service message at that target rate achieves an effective balance between high bandwidth utilization and reliable network operation (which can be understood as minimizing congestion).
[0008] As an example, before sending the first control message, the method may further include: the first communication device constructing the first control message. The first control message can be any type of control message capable of transmission in a network and carrying first link information. The function of the first control message includes: by transmitting the first link information along the first path, enabling each network node along the first path to compare the first link information with its own maintained second link information, and update the first link information in the first control message based on a preset principle, thereby obtaining the link rate (i.e., the target rate) of the bottleneck link (also called the target link) of the first service flow transmitted through the first path. In this way, the sending end of the first path, by constructing the first control message including the first link information, prepares for determining a reasonable target rate in this application.
[0009] As an example, the target rate is determined based on the first link information and the second link information of each link on the first path. The target rate belongs to the second link information of the target link on the first path, and the target link is the link that limits the transmission rate of the service message to be sent on the first path. It can be seen that the target rate determined by each network node on the first path based on its own maintained second link information and the first link information in the first control message is a reasonable transmission rate on the first path that can balance bandwidth utilization and reliable transmission of service messages.
[0010] As an example, the first link information may include an initial overload rate and / or an initial rate, where the initial overload rate is 0 and the initial rate is related to the network interface card (NIC) bandwidth of the first communication device. For example, the initial rate can be set to be equal to the NIC bandwidth of the first communication device. The second link information for each link on the first path may include the link overload rate and / or the link rate of the corresponding link. The link overload rate indicates the proportion of overloaded data on the link to the total input data on the link, and the link rate is the available bandwidth for service packets destined for that link.
[0011] As another example, the first link information may also include an initial link identifier, which may be set to 0. Then, the second control message may further include: the link identifier of the target link, providing a data basis for the sender to update the state information of network nodes and link information based on the target link's link identifier.
[0012] As an example, the method may further include: a first communication device constructing a third control message, the third control message including the link identifier and target rate of the target link; the first communication device sending the third control message to a second communication device through a first path, the third control message being used by each third communication device on the first path to update the status information of the corresponding link, the status information indicating the traffic status of the corresponding link, and the status information being used to calculate the link overload rate and link rate of the corresponding link. The first communication device sending the third control message may be periodic, for example, the first communication device sending the third control message to the second communication device through the first path at a preset first time interval. In this way, each network node on the first path updates the traffic status of its own link based on the third control message, thereby calculating a reasonable link overload rate and link rate, providing a reliable data foundation for determining an accurate target rate.
[0013] As an example, the timing of the first communication device sending a first control message to the second communication device via a first path may include, but is not limited to: a second time interval elapsed since the transmission of the control message carrying first link information, and / or a third time interval elapsed since the reception of the control message carrying the target rate, and / or an update to the link overload rate and / or link rate in the first path, and / or a determination that there are service messages to be transmitted via the first path. Thus, by flexibly designing the timing of triggering the transmission of the first control message, the method provided in this application can be triggered and executed under various circumstances, thereby making it possible to determine a reasonable target rate.
[0014] In some possible implementations, the first communication device may be a network node in the first path. In such cases, the method may further include: the first communication device receiving a fourth control message from a fourth communication device, the fourth control message carrying third link information used to determine the target rate, and the fourth communication device belonging to the first path; thereby, the first communication device obtains the first control message based on the third link information. In this way, each network node in the first path can determine the first control message to be sent based on the third link information carried in the received fourth control message, and each network node participates in the process of determining the target rate, making it possible to determine a reasonable target rate.
[0015] As an example, after receiving the fourth control message but before obtaining the first control message, the method may further include: the first communication device determining a candidate rate based on third link information and second link information of the first link. The first link information includes the candidate rate, where the first link is the link from the first communication device to the fifth communication device on the first path, the candidate rate is the maximum allowed transmission rate of the service message to be sent on the second path, the second path is the portion of the first path from the head node to the fifth communication device, and the candidate rate is greater than or equal to the target rate of the target link of the first path, where the target link is the link that limits the transmission rate of the service message to be sent on the first path. In this example, the first communication device obtaining the first control message may specifically include: the first communication device processing the fourth control message based on the candidate rate to obtain the first control message. In one scenario, if the overload rate carried by the third link information is less than the overload rate in the second link information, or if the overload rate carried by the third link information is equal to the overload rate in the second link information and the rate in the third link information is greater than the rate in the second link information, then the first communication device updates the third link information in the fourth control message to the value of the second link information of the first link, and obtains the first control message. In this case, the first link information is equal to the second link information of the first link. In another scenario, if the overload rate carried by the third link information is greater than the overload rate in the second link information, or if the overload rate carried by the third link information is equal to the overload rate in the second link information and the rate in the third link information is less than or equal to the rate in the second link information, then the first communication device keeps the value of the third link information in the fourth control message unchanged and obtains the first control message. In this case, the first link information is equal to the third link information.
[0016] As an example, the first link information may include the overload rate and / or the rate of the second path, where the rate of the second path is the maximum allowed transmission rate of the service message to be sent on the second path; the second link information may include the link overload rate and / or the link rate of the first link, where the link overload rate of the first link is used to indicate the proportion of overloaded data on the first link to the amount of input data on the first link, and the link rate of the first link is the available bandwidth of the service message with the first link as the target link.
[0017] As another example, the first link information may also include the link identifier of the candidate link on the second path, and / or the second control message may also include the link identifier of the target link.
[0018] As an example, the method may further include: a first communication device receiving a fifth control message from a fourth communication device, the fifth control message including a link identifier and a target rate for the target link, the target link being a link that limits the transmission rate of the service message to be sent on the first path; the first communication device updating the status information of the first link based on the fifth control message, the status information indicating the traffic status of the first link. In this example, the method may further include: the first communication device updating the link overload rate and the link rate of the first link based on the status information. Thus, each network node on the first path can update the traffic status of its own link based on the fifth control message, thereby calculating a reasonable link overload rate and link rate, providing a reliable data basis for determining an accurate target rate.
[0019] In some possible implementations, the second control message can be understood as a response message to the first control message, which is a request message. The first control message can also be considered a signaling request packet (I), and the second control message is the signaling response packet corresponding to the signaling request packet (I). Correspondingly, the third or fifth control message can be a signaling request packet (II).
[0020] Secondly, this application also provides a message processing method applied to a second communication device. This method may include, for example, the second communication device receiving a first control message sent by a first communication device. The first control message includes first link information used to determine a target rate, which is the maximum allowed transmission rate of the service message to be sent in a first path. The second communication device obtains a second control message based on the first link information. The second control message includes a first rate used to determine the target rate. Thus, without affecting the transmission of service messages in the network, the second communication device obtains a second control message carrying the first rate based on the first link information in the received first control message, so as to determine the maximum allowed transmission rate of the service message on the first path based on the first rate. This decouples the process of determining the target rate from the transmission flow of user data in the network. Not only does it eliminate the need for each network node to modify the service message, thus not affecting the normal transmission of the service message in the network, but it also eliminates the need for each network node to process the service message packet by packet, making the processing granularity controllable and effectively reducing the consumption of computing and storage resources. This makes high-performance service message processing possible.
[0021] In some possible implementations, the second communication device can be a network node on the first path. In this case, both the first control message and the second control message can be understood as control messages from the sending end to the receiving end, used to determine the target rate.
[0022] As an example, the method may further include: a second communication device determining a first rate based on first link information and second link information of the first link, wherein the first link is a link on a first path from the second communication device to the third communication device, the first rate is the maximum allowed transmission rate of the service message to be sent on the second path, the second path is the portion of the first path from the head node to the third communication device, and the first rate is greater than or equal to the target rate of a target link on the first path, the target link being a link that limits the transmission rate of the service message to be sent on the first path. In this example, the second communication device obtaining a second control message may, for example, include: the second communication device processing the first control message based on the first rate to obtain a second control message, the second control message including the first rate.
[0023] As an example, the method may further include: the second communication device sending a second control message to the third communication device via the first link. In this way, the third communication device can also, based on the second control message and the second link information it maintains belonging to the first path, further determine the maximum allowed transmission rate of the service message to be sent from the head node to the next hop of the third communication device on the first path, providing reliable data for finally determining a reasonable target rate for the first path.
[0024] As an example, the first link information includes the overload rate and / or the rate of the third path, which is the portion of the first path from the head node to the second communication device; the second link information includes the link overload rate and / or the link rate of the first link.
[0025] For example, when the first link information includes the overload rate and the rate of the third path, and the second link information includes the link overload rate and the link rate of the first link, the process by which the second communication device determines the first rate may include: if the link overload rate of the first link is greater than the overload rate of the third path, or if the link overload rate of the first link is equal to the overload rate of the third path and the link rate of the first link is less than the rate of the third path, then the link rate of the first link is determined to be the first rate; if the link overload rate of the first link is less than the overload rate of the third path, or if the link overload rate of the first link is equal to the overload rate of the third path and the link rate of the first link is greater than or equal to the rate of the third path, then the rate of the third path is determined to be the first rate.
[0026] As an example, if the first rate is the link rate of the first link, the second control message also includes the link overload rate of the first link. If the first rate is the rate of the third path, the second control message also includes the overload rate of the third path.
[0027] As an example, the method may further include: a second communication device receiving a third control message from a first communication device, the third control message including a link identifier and a target rate of a target link, the target link being a link that limits the transmission rate of the service message to be sent on a first path; the second communication device updating the status information of the first link based on the third control message, the status information being used to indicate the traffic status of the first link; and the second communication device sending the third control message to a third communication device. For example, updating the status information of the first link by the second communication device may include: adding the target rate carried in the third control message to the sum of the target rates in all third control messages carried on the first link to obtain the updated sum of the target rates in all third control messages carried on the first link; and / or, if the link identifier of the target link is used to indicate the first link, the second communication device increments the number of third control messages with the first link as the target link by one to obtain the updated number of third control messages with the first link as the target link; if the link identifier of the target link does not indicate the first link, the second communication device adds the target rate carried in the third control message to the sum of the target rates in the third control messages carried on the first link that are not with the first link as the target link to obtain the updated sum of the target rates in the third control messages carried on the first link that are not with the first link as the target link.
[0028] In this example, the method may further include: the second communication device updating the link overload rate and / or the link rate of the first link based on status information. For example, the process of the second communication device updating the link overload rate and / or the link rate of the first link may include: the second communication device obtaining the updated link overload rate of the first link based on the sum of the target rates in all third control messages carried on the updated first link and the link bandwidth of the first link; and / or, the second communication device obtaining the updated link rate of the first link based on the link bandwidth of the first link, the sum of the target rates in the third control messages carried on the updated first link that do not target the first link, and the updated number of third control messages that target the first link.
[0029] In some other possible implementations, the second communication device can be the receiving end on the first path. In this case, the first control message can be understood as a control message from the sending end to the receiving end for determining the target rate, and the second control message can be understood as a response message to the first control message. The first rate is the target rate.
[0030] As an example, the method may further include: a second communication device sending a second control message to the head node of the first path (i.e., the sender of the first path). In this way, the head node of the first path can obtain the target rate based on the second control message, making it possible for the head node of the first path to transmit service messages at a reasonable target rate, thereby improving the network's performance in transmitting service messages to some extent.
[0031] As an example, the first link information may include the overload rate and / or target rate of the target link, which is the link that limits the transmission rate of the service message to be sent on the first path.
[0032] As another example, the first link information may also include the link identifier of the target link, and the second control message may also include the link identifier of the target link.
[0033] As an example, the method may further include: a second communication device receiving a fourth control message via a first path, the fourth control message including the link identifier and target rate of the target link; and the second communication device discarding the fourth control message. This fourth control message can be understood as a control message used to update the status information, link overload rate, and link rate maintained on each network node along the first path.
[0034] Thirdly, this application provides a message processing apparatus applied to a first communication device, which may include at least a sending unit and a receiving unit. The sending unit is configured to send a first control message to a second communication device via a first path. The first control message includes first link information used to determine a target rate, which is the maximum allowed transmission rate of the service message to be sent in the first path. The receiving unit is configured to receive a second control message sent by the second communication device, the second control message carrying the target rate.
[0035] In some possible implementations, the apparatus further includes an acquisition unit. This acquisition unit is configured to acquire a first service message, the transmission path of which is a first path; the transmission unit is further configured to transmit the first service message to a second communication device via the first path at a target rate.
[0036] As an example, the device also includes a first building unit. This first building unit is used to build a first control message.
[0037] As an example, the target rate is the rate determined based on the first link information and the second link information of each link on the first path. The target rate belongs to the second link information of the target link on the first path, and the target link is the link that limits the transmission rate of the service message to be sent on the first path.
[0038] As an example, the first link information includes an initial overload rate and / or an initial rate, where the initial overload rate is 0 and the initial rate is related to the network card bandwidth of the first communication device; the second link information includes the link overload rate and / or the link rate of the corresponding link, where the link overload rate indicates the proportion of overloaded data on the link to the input data on the link, and the link rate is the available bandwidth of service packets with the link as the target link.
[0039] As an example, the first link information also includes an initial link identifier, and / or the second control message also includes the link identifier of the target link, with the initial link identifier being 0.
[0040] As an example, the device also includes a second construction unit. This second construction unit is used to construct a third control message, which includes the link identifier and target rate of the target link. The sending unit is further used to send the third control message to the second communication device via the first path. The third control message is used by each third communication device on the first path to update the status information of the corresponding link. The status information indicates the traffic situation of the corresponding link and is used to calculate the link overload rate and link rate of the corresponding link. For example, the sending unit is specifically used to send the third control message to the second communication device via the first path at preset first time intervals.
[0041] As an example, the sending unit is specifically configured to: in response to a second time interval being reached after the time interval for sending a control message carrying first link information, and / or in response to a third time interval being reached after the time interval for receiving a control message carrying a target rate, and / or in response to an update of the link overload rate and / or link rate of the link in the first path, and / or in response to determining that there is a service message to be transmitted through the first path, send a first control message to the second communication device through the first path.
[0042] In some other possible implementations, the device further includes a first processing unit. The receiving unit is further configured to receive a fourth control message from the fourth communication device, the fourth control message carrying third link information used to determine the target rate, and the fourth communication device belonging to the first path; then, the first processing unit is configured to obtain the first control message based on the third link information.
[0043] As an example, the device further includes a second processing unit. This second processing unit is configured to determine a candidate rate based on third link information and second link information of the first link. The first link information includes the candidate rate, where the first link is a link on a first path from the first communication device to the fifth communication device, the candidate rate is the maximum allowed transmission rate of the service message to be sent on the second path, the second path is the portion of the first path from the head node to the fifth communication device, and the candidate rate is greater than or equal to the target rate of the target link of the first path, where the target link is a link that limits the transmission rate of the service message to be sent on the first path.
[0044] As an example, the first link information includes the overload rate of the second path and / or the rate of the second path, where the rate of the second path is the maximum allowed transmission rate of the service message to be sent on the second path; the second link information includes the link overload rate of the first link and / or the link rate of the first link, where the link overload rate of the first link is used to indicate the proportion of overloaded data on the first link to the amount of input data on the first link, and the link rate of the first link is the available bandwidth of the service message with the first link as the target link.
[0045] As an example, the first link information may also include the link identifier of the candidate link on the second path, and / or the second control message may also include the link identifier of the target link.
[0046] As an example, the device also includes a third processing unit. The receiving unit is further configured to receive a fifth control message from the fourth communication device, the fifth control message including the link identifier and target rate of the target link, the target link being a link that limits the transmission rate of the service message to be sent on the first path; then, the third processing unit is configured to update the status information of the first link based on the fifth control message, the status information being used to indicate the traffic status of the first link.
[0047] As an example, the device also includes a fourth processing unit. This fourth processing unit is used to update the link overload rate and link speed of the first link based on status information.
[0048] In some possible implementations, the second control message is a response message to the first control message.
[0049] It should be noted that for the relevant description of the message processing device in the third aspect, please refer to the corresponding description in the first aspect.
[0050] Fourthly, this application also provides a message processing apparatus applied to a second communication device, which may include at least a receiving unit and a processing unit. The receiving unit is configured to receive a first control message sent by a first communication device, the first control message including first link information used to determine a target rate, the target rate being the maximum allowed transmission rate of the service message to be sent in a first path; the processing unit is configured to obtain a second control message based on the first link information, the second control message including a first rate used to determine the target rate.
[0051] In some possible implementations, the processing unit is further configured to determine a first rate based on the first link information and the second link information of the first link, wherein the first link is the link from the second communication device to the third communication device on the first path, the first rate is the maximum transmission rate allowed for the service message to be sent on the second path, the second path is the portion of the first path from the head node to the third communication device, and the first rate is greater than or equal to the target rate of the target link of the first path, and the target link is the link that limits the transmission rate of the service message to be sent on the first path.
[0052] As an example, the device also includes a transmitting unit. This transmitting unit is used to transmit a second control message to a third communication device via a first link.
[0053] As an example, the first link information includes the overload rate and / or the rate of the third path, which is the portion of the first path from the head node to the second communication device; the second link information includes the link overload rate and / or the link rate of the first link.
[0054] As an example, the first link information includes the overload rate and the speed of the third path, and the second link information includes the link overload rate and the link speed of the first link. The processing unit is specifically used to: if the link overload rate of the first link is greater than the overload rate of the third path, or if the link overload rate of the first link is equal to the overload rate of the third path and the link speed of the first link is less than the speed of the third path, then determine the link speed of the first link as the first speed; if the link overload rate of the first link is less than the overload rate of the third path, or if the link overload rate of the first link is equal to the overload rate of the third path and the link speed of the first link is greater than or equal to the speed of the third path, then determine the speed of the third path as the first speed.
[0055] As an example, if the first rate is the link rate of the first link, the second control message also includes the link overload rate of the first link; if the first rate is the rate of the third path, the second control message also includes the overload rate of the third path.
[0056] As an example, the device also includes a sending unit. A receiving unit is further configured to receive a third control message from the first communication device, the third control message including a link identifier and a target rate of the target link, the target link being a link that limits the transmission rate of the service message to be sent on the first path; a processing unit is further configured to update the status information of the first link based on the third control message, the status information indicating the traffic status of the first link; and a sending unit is further configured to send the third control message to the third communication device. For example, the processing unit is specifically configured to: add the target rate carried in the third control message to the sum of the target rates in all third control messages carried on the first link to obtain the updated sum of the target rates in all third control messages carried on the first link; and / or, the processing unit is specifically configured to: if the link identifier of the target link is used to indicate the first link, then increment the number of third control messages with the first link as the target link by one to obtain the updated number of third control messages with the first link as the target link; if the link identifier of the target link does not indicate the first link, then add the target rate carried in the third control message to the sum of the target rates in the third control messages carried on the first link that are not with the first link as the target link to obtain the updated sum of the target rates in the third control messages carried on the first link that are not with the first link as the target link.
[0057] As an example, the processing unit is also configured to update the link overload rate and / or link speed of the first link based on state information. For instance, the processing unit is specifically configured to: obtain the updated link overload rate of the first link based on the sum of the target rates in all third control messages carried on the updated first link and the link bandwidth of the first link; and / or, the processing unit is specifically configured to: obtain the updated link speed of the first link based on the link bandwidth of the first link, the sum of the target rates in the third control messages carried on the updated first link that do not target the first link, and the updated number of third control messages that target the first link.
[0058] In some other possible implementations, the second control message is a response message to the first control message, the first rate is the target rate, and the device further includes a sending unit. This sending unit is used to send the second control message to the head node of the first path.
[0059] As an example, the first link information includes the overload rate and / or target rate of the target link, which is the link that limits the transmission rate of the service message to be sent on the first path.
[0060] As an example, the first link information also includes the link identifier of the target link, and the second control message also includes the link identifier of the target link.
[0061] As an example, the receiving unit is also configured to receive a fourth control message through the first path, the fourth control message including the link identifier and target rate of the target link; the processing unit is also configured to discard the fourth control message.
[0062] It should be noted that for the relevant description of the message processing device in the fourth aspect, please refer to the corresponding description in the second aspect.
[0063] Fifthly, this application provides a communication device, which includes a processor and a memory; the processor is configured to execute instructions stored in the memory to cause the communication device to implement the method corresponding to the first aspect or the second aspect and their possible implementations.
[0064] In a sixth aspect, this application also provides a communication device, which includes an interface and a processor; the interface is used to receive instructions and transmit them to the processor; the processor is used to execute the method corresponding to the first aspect or the second aspect and their possible implementations.
[0065] In a seventh aspect, this application also provides a communication system, which includes a first communication device and a second communication device. The first communication device is used to execute the method corresponding to the first aspect and its possible implementations; the second communication device is used to execute the method corresponding to the second aspect and its possible implementations.
[0066] Eighthly, this application also provides a storage medium storing instructions that, when executed on a processor, implement the method corresponding to the first aspect or the second aspect and their possible implementations.
[0067] Ninthly, this application also provides a program product comprising a program that, when run on a processor, implements a method corresponding to the first aspect or the second aspect and its possible implementations.
[0068] In a tenth aspect, this application provides a chip including an interface circuit and a processing circuit, wherein the interface circuit is connected to the processing circuit. The interface circuit is used to perform receiving and transmitting operations in the methods corresponding to the first aspect or the second aspect and their possible implementations; the processing circuit is used to perform other operations in the methods corresponding to the first aspect or the second aspect and their possible implementations besides the receiving and transmitting operations.
[0069] In the eleventh aspect, this application provides a chip including a memory and a processor. The memory is used to store instructions, and the processor is used to call and execute the instructions from the memory to implement the methods corresponding to the first aspect, the second aspect, and their possible implementations. Attached Figure Description
[0070] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings.
[0071] Figure 1 This is a flowchart illustrating a current business message processing strategy.
[0072] Figure 2 This is a flowchart illustrating an example of a processing strategy in an embodiment of this application.
[0073] Figure 3 This is a flowchart illustrating a message processing method 100 in an embodiment of this application.
[0074] Figure 4 for Figure 3 A flowchart illustrating one possible scenario 1;
[0075] Figure 5 for Figure 3 A flowchart illustrating one possible scenario 2;
[0076] Figure 6 for Figure 3 A flowchart illustrating one possible scenario 3;
[0077] Figure 7 for Figure 3 A flowchart illustrating one possible scenario 4;
[0078] Figure 8 This is a flowchart illustrating an example of a message processing method in an embodiment of this application.
[0079] Figure 9 This is a flowchart illustrating another example of a message processing method in the embodiments of this application;
[0080] Figure 10 This is a schematic diagram of the structure of a first communication device 1000 in an embodiment of this application;
[0081] Figure 11 This is a schematic diagram of the structure of a second communication device 1100 in an embodiment of this application;
[0082] Figure 12 This is a schematic diagram of the structure of a communication system 1200 according to an embodiment of this application;
[0083] Figure 13 This is a schematic diagram of the structure of a communication device 1300 according to an embodiment of this application;
[0084] Figure 14This is a schematic diagram of the structure of a chip 1400 in an embodiment of this application. Detailed Implementation
[0085] Understandably, if the transmission rate of service packets in a network is not reasonable, it will affect the network's performance. For example, if the transmission rate is too high, it will cause network congestion; if the transmission rate is too low, it will waste network bandwidth. Therefore, transmitting service packets at a reasonable rate to achieve a balance between reliable network operation and high bandwidth utilization is a crucial aspect of network control. More and more applications require high-performance transmission of service packets. For instance, various emerging cloud-based internet applications have an increasing demand for efficient and effective transmission of massive amounts of data. Furthermore, in the era of computing power networks, emerging applications such as geographically distributed machine learning and distributed artificial intelligence (AI) large-scale model training are rapidly developing, relying on ultra-high-speed transmission capabilities for massive amounts of data and long-distance cross-regional network transmission capabilities, requiring support for ultra-high-speed, long-distance data transmission.
[0086] Currently, network processing strategies for service packets typically require each network node from the sender to the receiver to include its collected status information—used to determine the sender's transmission rate for subsequent service packets—in an extended header of the service packet and transmit it to the receiver. The receiver then sends this extended header back to the sender, which uses the status information in the extended header to determine the transmission rate and continues forwarding subsequent service packets at that rate. Alternatively, each network node from the sender to the receiver can include its calculated control information (which may include the transmission rate) used to determine the sender's transmission rate for subsequent service packets in an extended header of the service packet and transmit it to the receiver. The receiver then sends this extended header back to the sender, which uses the control information in the extended header to determine the transmission rate and continues forwarding subsequent service packets at that rate. It is evident that this processing strategy requires each network node to modify the service packets packet by packet, increasing the processing pressure on the network nodes and affecting the network's processing performance. On the one hand, when there are many service packets, it wastes network resources such as storage and processing. On the other hand, by using user service packets to control the transmission rate, it reduces the reliability of the service packet content itself in the network transmission, resulting in a poor user experience.
[0087] For example, such as Figure 1As shown, assuming that path 1 of the network includes a sender 10, network node 11, network node 12, and receiver 13 in sequence, according to the current network's service packet processing strategy, the process of processing a single packet can include: S11, the sender 10 obtains the service packet 1 that needs to be transmitted through path 1, adds a congestion control header to the service packet 1, and sends the service packet 1 to network device 11; S12, network device 11 calculates the suggested transmission rate 1, writes the transmission rate 1 into the congestion control header, and obtains the service packet 2; S13, network device 11 transmits the service packet... Service message 2 is sent to network device 12; S14, after calculating the suggested transmission rate 2, network device 12 writes the transmission rate 2 into the congestion control header of service message 2, thus obtaining service message 3; S15, network device 12 sends service message 3 to receiving end 13; S16, receiving end 13 sends the congestion control header in service message 3 to sending end 10; S17, sending end 10 obtains the transmission rate 3 based on the received congestion control header, where the transmission rate 3 can be transmission rate 2, transmission rate 1, or a transmission rate determined based on transmission rate 1 and transmission rate 2. Thus, for subsequent service messages transmitted through path 1, such as service message 4, this process can also include: S18, sending end 10 to receiving end 13 via path 1 at a transmission rate of 3. It is evident that this process requires each network node to calculate and modify service messages packet by packet, affecting the transmission of user traffic in the network.
[0088] Furthermore, current processing strategies suffer from poor scalability and deployment difficulties. For example, in-band network telemetry (INT) requires each network node to insert its collected status information into the INT header of the service packet as metadata. However, not all network nodes support INT, necessitating significant upgrades and modifications to each node. Similarly, the Rate Control Protocol (RCP) requires each network node to obtain information such as service traffic and queue length from the underlying hardware, calculate the transmission rate, and add it to the congestion control header extended in the service packet. This increases hardware dependence on network nodes, requiring adaptive modifications to each node for compatibility. In practice, modifications to each network node are necessary. For cases where congestion control headers are handled through the kernel protocol stack, each network node needs to modify its own kernel protocol stack. Alternatively, for cases where congestion control headers are handled by adding software applications, each network node needs to copy data packets from the underlying layer (such as the hardware network card) to the application layer via the kernel protocol stack. The application layer then processes the packets, and the processed packets are copied back from the application layer to the underlying layer via the kernel protocol stack. The underlying layer then sends the processed packets. This process is complex, time-consuming, and incurs significant overhead for the network nodes.
[0089] Furthermore, the current processing strategy is not reasonable in calculating the transmission rate. Taking RCP as an example, setting a large transmission rate for each service flow first will cause queuing; then, the transmission rate is reduced to eliminate congestion, obtaining an estimate of the number of service flows, and then the transmission rate is calculated based on the estimated number of service flows. However, this aggressive calculation method will cause a large burst of traffic in a short period of time when encountering a large surge in service traffic, resulting in high instantaneous congestion and packet loss.
[0090] Based on this, embodiments of this application provide a more reasonable message processing method, which may include, for example, a first communication device sending a first control message to a second communication device through a first path, the first control message including first link information for determining a target rate, the target rate being the maximum allowed transmission rate of the service message to be sent in the first path; after receiving the first control message, the second communication device obtains a second control message including a first rate based on the first link information, the first rate being used to determine the target rate; thereby, the first communication device obtains a third control message, the third control message carrying the target rate, the third control message being the same as or related to the second control message. In this way, without affecting the transmission of service messages in the network, the maximum allowed transmission rate of service messages on the first path is efficiently determined by transmitting control messages in the network. This decouples the message processing method provided in this application embodiment from the transmission process of user data in the network. Not only does it eliminate the need for each network node to modify the service messages and ensure the normal transmission of service messages in the network, but it also eliminates the need for each network node to process service messages packet by packet, making the processing granularity controllable and effectively reducing the occupation of computing and storage resources. Thus, it makes it possible to achieve high-performance service message processing.
[0091] Furthermore, the communication apparatus in the method provided in this application involves a sending end, intermediate network nodes, and a receiving end in the network. In addition to message forwarding, each device requires the sending end to have the function of sending a first control message and processing a third control message; the receiving end to have the function of processing a first control message and sending a third control message; and the network nodes to have at least the following functions: (1) reading and modifying the first control message, (2) storing and periodically updating local link information, and (3) forwarding the second control message. Therefore, the method provided in this application only requires adding corresponding control message construction and processing logic, without requiring significant modifications to the devices in the network, especially without relying on or modifying the hardware of each device. This improves the scalability of the method and reduces deployment difficulty.
[0092] Furthermore, the method provided in this application embodiment can reasonably determine the maximum allowable transmission rate (i.e., target rate) when the service message to be sent is transmitted along an entire path based on the link information transmitted in the control message. Since the target rate is determined based on the actual load situation in the network at the current or recent period, transmitting the service message in the network at the target rate can achieve a relative balance between reliable network operation and high bandwidth utilization, overcome the problem of network congestion caused by excessively high transmission rates of service messages, and also avoid the problem of network bandwidth waste caused by excessively low transmission rates of service messages.
[0093] For example, such as Figure 2 As shown, it is still assumed that path 1 of the network includes a sender 10, a network node 11, a network node 12, and a receiver 13 in sequence. The link from the sender 10 to the network node 11 on path 1 is called link 1, the link from the network node 11 to the network node 12 on path 1 is called link 2, and the link from the network node 12 to the receiver 13 on path 1 is called link 3. The network node 11 maintains the link information 2 of link 2, which may include, but is not limited to: the link overload rate o2 of link 2, the link speed g2 of link 2, and the link identifier ID2 of link 2. The network node 12 maintains the link information 3 of link 3, which may include, but is not limited to: the link overload rate o3 of link 3, the link speed g3 of link 3, and the link identifier ID3 of link 3.
[0094] According to the message processing method provided in the embodiments of this application, before transmitting the service message of service flow 1 through path 1, the process of performing one processing step may include: S21, the sending end 10 constructs a control message 1, which includes link information 1. The link information 1 may include an initial overload rate o1 and / or an initial rate g1. The link information 1 may also include an initial link identifier ID1. The link information 1 may be set as: ID1 = 0, o1 = 0, g1 = the network card bandwidth Cr of the sending end 10. S22, the sending end 10 sends the control message 1 to the network node 11 through link 1. S23, the network node 11 processes the control message 1 based on the comparison between the link information 1 and the link information 2 it maintains, to obtain a control message 2. The control message 2 may include link information 1'. For example, if network node 11 determines that o2 > o1, or determines that o1 = o2 and g2 < g1, then link 2 is determined to be a candidate bottleneck link for service flow 1 on path 1. Therefore, the link information 1 in control message 1 is updated to the corresponding value of link information 2. Specifically, ID1 = ID2, o1 = o2, and g1 = g2 are set to obtain control message 2 (i.e., the updated control message 1). In this case, link information 1' is link information 2. Alternatively, if network node 11 determines that o2 < o1, or o1 = o2 and g2 ≥ g1, then link 2 is considered unlikely to be a bottleneck link for service flow 1 on path 1. Therefore, the value of link information 1 in control message 1 is maintained to obtain control message 2 (which can be understood as the same as control message 1, but named control message 2 for consistency in subsequent descriptions). In this case, link information 1' is link information 1. S24, network node 11 sends control message 2 to network node 12 through link 2. S25, network node 12 processes control message 2 based on a comparison of link information 1' and its own maintained link information 3 to obtain control message 3, which may include link information 1'. Taking link information 1' as an example, where ID1 = ID2, o1 = o2, g1 = g2, for instance, if network node 12 determines o3 > o2, or determines o2 = o3 and g3 < g2, then it determines that compared to link 2, link 3 is a candidate bottleneck link for service flow 1 on path 1. Therefore, it updates the link information 1' in control message 2 to the corresponding value of link information 3. Specifically, it sets... ID1 = ID3, o1 = o3, g1 = g3, to obtain control message 3 (i.e., the updated control message 2). In this case, link information 1” is link information 3. For example, if network node 12 determines that o3 < o2, or o2 = o3 and g3 ≥ g2, then it is considered that compared to link 2, link 3 cannot be the bottleneck link of service flow 1 on path 1. Therefore, the value of link information 1' in control message 2 is maintained to obtain control message 3 (which can be understood as the same as control message 2, and named control message 3 for consistency in subsequent descriptions). In this case, link information 1” is link information 1'.S26, network node 12 sends control message 3 to receiver 13 via link 3. S27, receiver 13 obtains link information 1” from control message 3 and constructs control message 4. Control message 4 can be understood as a response message to control message 1. Control message 4 can carry all or part of link information 1”. Taking link information 1” as ID1=ID3, o1=o3, g1=g3 as an example, control message 4 must at least carry g1=g3, and may also carry ID1=ID3 and / or o1=o3. S28, receiver 13 sends control message 4 to sender 10. S29, sender 10 uses g3 carried in control message 4 as the target rate, controlling the service messages in service flow 1 to be transmitted on path 1 at this target rate.
[0095] Following S29, on one hand, the sending end 10 can set the transmission rate x corresponding to service flow 1. The value of x is g1 in control message 4, that is, the target rate corresponding to service flow 1 determined in the last execution of this process. The sending end 10 can also set IDt corresponding to service flow 1. IDt is the link identifier of the bottleneck link corresponding to service flow 1 determined in the last execution of this process. Taking link 3 as the bottleneck link of service flow 1 as an example, x = g3 and IDt = ID3 are set. Thus, the sending end 10 can send the service message of service flow 1 through path 1 with x. On the other hand, in order to enable the method provided in this application embodiment to determine a relatively reasonable target rate (also known as the transmission rate) under changes in network load, the sending end 10 can also construct and send a control message 5 to the receiving end 13 through path 1. The control message 5 may include IDt and x corresponding to service flow 1, so that network nodes 11 and 12 can dynamically adjust the link information they maintain based on changes in network load, ensuring that the link information they maintain reflects the latest load of the corresponding link, so that the next execution of S21 to S29 can determine a more reasonable target rate that conforms to the latest load in the network. It should be noted that the transmission of service packets of service flow 1 and the updating of the link information maintained by the network nodes are two decoupled processes and do not affect each other. In addition, the timing of determining the target rate and the timing of the network nodes updating the link information they maintain also do not affect each other.
[0096] like Figure 2As shown, after S29, the process of the network node updating the maintained link information may include, for example, the following steps: S31, the sending end 10 constructs control message 5, which may include at least IDt and x. S32, the sending end 10 sends control message 5 to network node 11 via link 1. S33, network node 11 updates link information 2 based on IDt and x. S34, network node 11 sends control message 5 to network node 12 via link 2. S35, network node 12 updates link information 3 based on IDt and x. S36, network node 12 sends control message 5 to receiving end 13 via link 3. S37, receiving end 13 discards control message 5. It should be noted that the update process of S31 to S37 can be triggered periodically. For example, the sending end 10 stores the time interval 1 for sending control message 5. When the time interval 1 is reached after the last control message 5 was sent, the update process is triggered. Alternatively, for the first transmission of control message 5, the update process can be triggered when the time interval 2 (which can be the same as or different from time interval 1) is reached after the time interval 2 is reached after the time interval 1 is sent or after the time interval 2 is reached after the time interval 2 is reached after the time interval 1 is sent. It should be noted that when the first transmission of control message 5 is when the time interval 2 is reached after the time interval 1 is sent, the value of time interval 2 can be flexibly set and there is no limit to its size. In this case, the function of sending control message 5 can include reserving bandwidth for service flow 1 to be sent through path 1. When the first transmission of control message 5 is when the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 2 is reached after the time interval 1 is reached, time interval 2 needs to be a small value (e.g., 1 millisecond).
[0097] Regarding the specific implementation of network nodes updating the link information they maintain based on control message 5, the implementation methods of S33 and S35 are similar. Taking S33 as an example, S33 may include: S331, network node 11 updates Y, such as Y = Y + x; S332, network node 11 determines whether IDt is the same as ID2. If they are the same, then link 2 is considered to be the target link of the corresponding service flow, and N is updated, such as N = N + 1. If they are not the same, then link 2 is considered not to be the target link of the corresponding service flow, and U is updated, such as U = U + x. Optionally, S33 may also include the process of updating the link overload rate and updating the link rate. For example, S33 may also include: S333, network device 11 calculates the updated o2 based on Cl and the updated Y; S334, network device 11 calculates the updated g2 based on Cl, the updated N, and the updated U. The explanations of Y, U, and N, as well as the specific implementation methods of S333 and S334, can be found in the relevant descriptions in the corresponding glossary below.
[0098] It should be noted that the execution order of S333 and S334 is not limited; they can be executed sequentially or simultaneously. The following description uses simultaneous execution as an example. S333 and S334 can be triggered periodically or by events (e.g., S333 and S334 are triggered after S332 is executed; or S333 is triggered after Y is updated, and S334 is triggered after U and N are updated). Taking periodic triggering as an example, network node 11 stores the time interval 3 for updating o2 and g2 (which can be the same as or different from time intervals 1 and 2). When time interval 3 has elapsed since the last update of o2 and g2, S333 and S334 are triggered. It should be noted that after updating o2 and g2, network node 11 can clear its local traffic statistics to facilitate the next stage of network load statistics and link information updates.
[0099] In this way, network nodes can continuously update the link information they maintain based on changes in network load, making it possible to determine a relatively reasonable target rate in the next execution of S21 to S29.
[0100] To make the technical solutions provided in the embodiments of this application clearer, the terms involved in the embodiments of this application will be explained below.
[0101] Bottleneck links and bottleneck flows are relative concepts; the explanation of one relies on the other. Specifically, the link providing the least available bandwidth to a service flow is defined as its bottleneck link; a link whose available bandwidth limits the increase in the transmission rate of a service flow is defined as its bottleneck flow. Conversely, all links traversed by a service flow, except for its bottleneck links, are considered non-bottleneck links; a link whose available bandwidth does not limit the increase in the transmission rate of a service flow is defined as its non-bottleneck flow. For example, service flow 1 is transmitted through path 1, which includes links 1, 2, and 3. The available bandwidths of links 1, 2, and 3 are 10 gigabits per second (Gbps), 8 Gbps, and 12 Gbps, respectively. Link 2, with the lowest available bandwidth, is the bottleneck link for service flow 1, while links 1 and 3 are non-bottleneck links. The maximum transmission rate that service flow 1 can achieve on path 1 is 8 Gbps. Link 2 limits the increase in the transmission rate of service flow 1. Therefore, service flow 1 is a bottleneck flow on link 2 and a non-bottleneck flow on links 1 and 3. Assuming service flow 2 is transmitted through path 2, which includes links 1, 2, and 4, with available bandwidths of 10 Gbps, 8 Gbps, and 5 Gbps, link 4 limits the increase in the transmission rate of service flow 2, while link 2 does not. Therefore, service flow 2 is a non-bottleneck flow on link 2.
[0102] Link information maintained on a network node can be understood as information reflecting the load status of that link. Link information may include, but is not limited to, link overload rate, link speed, and traffic statistics. Traffic statistics may include, for example, Y, N, and U. Here, Y can be understood as the sum of x values in all control messages 5 carried on the corresponding link; N can be understood as the number of control messages 5 carried on the corresponding link with that link as the destination link; and U can be understood as the sum of x values in control messages 5 carried on the corresponding link with that link as a non-destination link.
[0103] Link overload rate can be understood as the proportion of traffic exceeding the link bandwidth passing through a given link, reflecting the link's load status. Traffic passing through a link can be represented by Y. As an example, the link overload rate can be calculated as (Y-Cl) / Y, where Cl is the link bandwidth. Another example is... Calculate, where C l γ represents the link bandwidth of the corresponding link.l Δ represents the target bandwidth utilization rate of the corresponding link. l The time interval for updating the link overload rate and link speed (which can correspond to time interval 3 above) is ∈, and the time interval for sending control message 5 (which can correspond to time interval 1 above).
[0104] Link rate, in this context, can be understood as the suggested transmission rate of a bottleneck flow through a link, depending on the link bandwidth, as well as N and U. As an example, the link rate can be calculated using (Cl-U) / N, where Cl is the link bandwidth. As another example, the link rate can be calculated using... Calculate, where C l γ represents the link bandwidth of the corresponding link. l Δ represents the target bandwidth utilization rate of the corresponding link. l The time interval for updating link overload rate and link speed (corresponding to time interval 3 above) is ∈, which is the time interval for sending control message 5 (corresponding to time interval 1 above). Link speed can be achieved through metrics such as link fair rate or window size. Fair rate can refer to the traffic per second that the bottleneck flow is expected to transmit on a link, while window size can refer to the traffic that the bottleneck flow is expected to transmit on a link in one round-trip time (RTT).
[0105] A communication device can be understood as any type of device or functional module or chip within a device that has message forwarding capabilities. Depending on the deployment location, communication devices play different roles in the methods provided in this application embodiment. These roles correspond to the service flow and the path through which the service flow is transmitted. For a specific path of the service flow, the role of each communication device remains fixed. However, for different transmission paths of different service flows or different transmission ports of the same service flow, the roles of each communication device may differ. The roles of the communication devices involved in this application embodiment can include: receiver, transmitter, and network node. Taking a communication device as an example, the transmitter and receiver can be end-side devices such as servers and hosts, or network devices such as switches, routers, and firewalls; the network node can be a network device such as a switch, router, or firewall.
[0106] It should be noted that the embodiments of this application only address one path for transmitting a single service flow, determining the maximum allowed transmission rate (i.e., target rate) of the service packets transmitted on that path. For scenarios such as load balancing and multi-path transmission, a service flow may traverse multiple paths in the network. For different paths, the method provided in the embodiments of this application can be executed separately to obtain the target rate for the corresponding path, thereby controlling the service packets transmitted through that path. The execution of the method provided in the embodiments of this application for each path does not affect each other.
[0107] To provide a clearer description of the embodiments of this application, the methods provided in the embodiments of this application will be described below with reference to the accompanying drawings.
[0108] Figure 3 This is a flowchart illustrating a message processing method 100 provided in an embodiment of this application. Method 100 targets a first path for transmitting a first service flow. The embodiment of this application describes the interaction between a first communication device and a second communication device on the first path. On the first path, the second communication device can be the next n-hop node of the first communication device, where n is an integer greater than or equal to 1. It should be noted that the steps performed by the first communication device in method 100 can be implemented as separate embodiments, and the steps performed by the second communication device can be implemented as separate embodiments.
[0109] As an example, the first communication device can be understood as the head node (i.e., the sender) of the first path, and the second communication device can be understood as the tail node (i.e., the receiver) of the first path. See [link to documentation] for details. Figure 4 The corresponding description, corresponding to Figure 2 In the example shown, the first communication device can correspond to the transmitter 10, and the second communication device can correspond to the receiver 13. As another example, the first communication device can be understood as a network node in the middle of the first path, and the second communication device can be understood as the tail node of the first path. See [link to relevant documentation] for details. Figure 5 The corresponding description, corresponding to Figure 2 In the example shown, the first communication device can correspond to network node 11 or network node 12, and the second communication device can correspond to receiver 13; as another example, the first communication device can be understood as the head node of the first path, and the second communication device can be understood as a network node in the middle of the first path. For details, please refer to [link to relevant documentation]. Figure 6 The corresponding description, corresponding to Figure 2 In the example shown, the first communication device can correspond to the transmitter 10, and the second communication device can correspond to network node 11 or network node 12; as another example, the first communication device can be understood as network node 1 in the first path, and the second communication device can be understood as network node 2 in the first path. See [link to relevant documentation] for details. Figure 7The corresponding description, corresponding to Figure 2 In the example shown, the first communication device can correspond to network node 11, and the second communication device can correspond to network node 12.
[0110] like Figure 3 As shown, the method 100 may include, for example, the following steps S101 to S104:
[0111] S101, the first communication device sends a first control message to the second communication device through a first path. The first control message includes first link information, which is used to determine the target rate. The target rate is the maximum transmission rate allowed for the service message to be sent in the first path.
[0112] S102, the second communication device receives the first control message sent by the first communication device.
[0113] The first control message can be any type of control message that can be transmitted in the network and carries first link information. The function of the first control message includes: by carrying the first link information and transmitting it on the first path, each network node on the first path compares the first link information with the second link information it maintains, and updates the first link information in the first control message based on a preset principle, so as to obtain the link rate (i.e. the target rate) of the bottleneck link of the first service flow transmitted through the first path.
[0114] The first control message can also be called a signaling request packet (I), and can be implemented, for example, through an Internet Control Message Protocol (ICMP) message. The first link information can be carried in the header or payload of the ICMP message. This application does not specifically limit the specific format of the first control message or the way it carries the first link information.
[0115] The target rate is determined based on the first link information and the second link information of each link on the first path. The target rate belongs to the second link information of the target link (i.e., the bottleneck link) on the first path. The target link is the link that limits the transmission rate of the service message to be sent on the first path. The second link information may include the link overload rate and / or the link rate of the corresponding link. The link overload rate indicates the proportion of overloaded data on the corresponding link to the total input data on that link. The link rate is the available bandwidth of the service message with the corresponding link as the target link.
[0116] In the case where the first communication device is the head node of the first path, such as Figure 4 or Figure 6As shown, before S101, the method 100 may further include: S105, the first communication device constructs a first control message. It is understood that the first link information may carry information related to the bottleneck link suggested when transmitting the first service flow on the first path. The first link information may include: an initial overload rate and / or an initial rate, where the initial overload rate can be understood as the link overload rate of the bottleneck link suggested when transmitting the first service flow on the first path; the initial rate can be understood as the link rate of the bottleneck link suggested when transmitting the first service flow on the first path. Optionally, the first link information may also include an initial link identifier, which can be understood as the link identifier of the bottleneck link suggested when transmitting the first service flow on the first path. Wherein, the initial overload rate can be set to 0, the initial rate is related to the network interface card bandwidth of the first communication device, such as the initial rate can be set to be equal to the network interface card bandwidth of the first communication device, and the initial link identifier can be set to 0.
[0117] In some possible implementations, after S105, the triggering conditions for executing S101 may include, but are not limited to: Condition 1, a second time interval has elapsed since the last transmission of the first control message carrying the first link information; Condition 2, a third time interval has elapsed since the last reception of the third control message carrying the target rate; Condition 3, the link overload rate and / or link rate of the links in the first path are updated; Condition 4, it is determined that there is a service message to be transmitted through the first path. Execution of S101 is triggered when at least one of conditions 1 to 4 is satisfied. As an example, S101 may include: in response to the second time interval elapsed since the transmission of the control message carrying the first link information, and / or in response to the third time interval elapsed since the last reception of the control message carrying the target rate, and / or in response to the update of the link overload rate and / or link rate of the links in the first path, and / or in response to the determination that there is a service message to be transmitted through the first path, the first communication device sends the first control message to the second communication device through the first path.
[0118] It should be noted that, in this case, the process by which each network node on the first path processes and modifies the first control message can be found in [reference needed]. Figure 5 , Figure 6 or Figure 7 The corresponding description.
[0119] In the case where the first communication device is a network node in the middle of the first path, such as Figure 5 or Figure 7As shown, before S101, the method 100 may further include: S106, where the first communication device receives a fourth control message from a fourth communication device (which may be the head node of the first path or a network node in the middle of the first path), the fourth control message carrying third link information used to determine the target rate, and the fourth communication device belonging to the first path; S107, where the first communication device obtains the first control message based on the third link information. Assuming the first path includes a first link from the first communication device to a fifth communication device (the fifth communication device may be the next-hop node of the first communication device on the first path, a network node, or a receiving end), then between S106 and S107, the method 100 may, for example, further include: S108, where the first communication device determines a candidate rate based on the third link information and the second link information of the first link it maintains, the candidate rate being the maximum allowed transmission rate of the service message to be sent on the second path, the second path being the portion of the first path from the head node to the fifth communication device, the candidate rate being greater than or equal to the target rate of the target link of the first path, and the target link being the link that limits the transmission rate of the service message to be sent on the first path. Therefore, S107 may include, for example, the first communication device processing the fourth control message based on the candidate rate to obtain the first control message. The first link information includes the candidate rate.
[0120] In this case, the first link information may include the overload rate of the second path and / or the rate of the second path (i.e., the candidate rate), where the rate of the second path is the maximum allowed transmission rate of the service message to be sent on the second path. The second link information of the first link may include: the link overload rate of the first link and / or the link rate of the first link, wherein the link overload rate of the first link is used to indicate the proportion of overloaded data on the first link to the input data of the first link, and the link rate of the first link is the available bandwidth of the service message with the first link as the target link. If the fourth communication device is a node in the first path, then before S106, the method 100 may further include: the fourth communication device constructing a fourth control message, the fourth control message and the third link information being described in the above-mentioned "case where the first communication device is the head node of the first path" regarding the first control message and the first link information; if the fourth communication device is a network node in the middle of the first path, then the third link information may include the overload rate and / or the rate of the third path, wherein the rate of the third path is the maximum allowed transmission rate of the service message to be sent on the third path, and the third path is the part of the first path from the head node to the first communication device. Optionally, the first link information may further include the link identifier of the link corresponding to the candidate rate (which can be understood as a candidate bottleneck link), and the third link information may further include the link identifier of the candidate bottleneck link of the third path.
[0121] If the third link information includes overload rate o6, speed g6, and link identifier ID6, and the second link information of the first link includes link overload rate o7, link speed g7, and link identifier ID7, then S108 may include, for example: S1081, the first communication device determines the outgoing link as the first link based on the destination address of the fourth control message; S1082, the first communication device determines whether o7 is greater than o6. If yes, then execute S1084; if no, then execute S1083; S1083, the first communication device determines whether o7 is equal to o6 and g7 is less than g6. If yes, then execute S1084; otherwise, execute S1085; S1084, the first communication device sets the overload rate in the fourth control message to o7, the speed to g7, and the link identifier to ID7; S1085, the first communication device keeps the overload rate, speed, and link identifier in the fourth control message unchanged. It should be noted that after executing S1084, the first link information in the first control message obtained after S107 can be the same as the second link information of the first link; after executing S1085, the first link information in the first control message obtained after S107 can be the same as the third link information.
[0122] It should be noted that, in this case, the process of the sender constructing and sending the fourth control message on the first path can be found in [reference needed]. Figure 4 and Figure 6 The corresponding description.
[0123] In both scenarios described above, if the second communication device is an intermediate network node on the first path, the first communication device can send a first control message to the second communication device via at least one link on the first path. After receiving the first control message, the second communication device can process and modify the first control message. If the second communication device is the tail node of the first path, the first communication device can send a first control message to the second communication device via the first path. After receiving the first control message, the second communication device can construct a third control message and send it to the head node of the first path.
[0124] S103, the second communication device obtains a second control message based on the first link information. The second control message includes a first rate, which is used to determine the target rate.
[0125] S104, the head node of the first path receives a third control message, which carries the target rate.
[0126] The third control message can be understood as a response message to the first control message. If the first control message is called a signaling request packet (I), then the third control message can be called a signaling response packet corresponding to the signaling request packet (I).
[0127] The target rate included in the third control message is the link rate of the target link, which is the bottleneck link of the first path. Therefore, the third control message may also include the link overload rate of the target link and / or the link identifier of the target link.
[0128] It should be noted that the transmission path of the third control message can be the reverse path of the first path or other paths. Taking the first communication device as the head node of the first path as an example, S104 may include: the first communication device receiving the third control message. If the first communication device is not the head node of the first path, but the transmission path of the third control message can be the reverse path of the first path, then S104 may include: after receiving the third control message, the first communication device forwards the third control message to the head node of the first path. It should be noted that... Figure 3 S104 in the example is the first communication device receiving a third control message.
[0129] In the case where the second communication device is a network node in the middle of the first path, assuming the first path includes a second link from the second communication device to the sixth communication device, then, between S102 and S103, method 100 may further include: S109, the second communication device determines a first rate based on the first link information and the second link information of the second link, the first rate being the maximum allowed transmission rate of the service message to be sent on the fourth path, the fourth path being the portion of the first path from the head node to the sixth communication device, the first rate being greater than or equal to the target rate of the target link (i.e., the bottleneck link) of the first path, the target link being the link that limits the transmission rate of the service message to be sent on the first path. Then, S103 may, for example, include: the second communication device processing a first control message based on the first rate to obtain a second control message, the second control message including the first rate.
[0130] The first link information may include the overload rate and / or the rate of the third path, where the third path is the portion of the first path from the head node to the second communication device; the second link information may include the link overload rate and / or the link rate of the second link. Optionally, the first link information may further include the link identifier of a candidate bottleneck link on the third path, and the second link information may further include the link identifier of the second link.
[0131] As an example, if the first link information includes the overload rate and the rate of the third path, and the second link information includes the link overload rate and the link rate of the second link, S109 may include, for example: S1091, if the link overload rate of the second link is greater than the overload rate of the third path, or if the link overload rate of the second link is equal to the overload rate of the third path and the link rate of the second link is less than the rate of the third path, then the link rate of the second link is determined to be the first rate; S1092, if the link overload rate of the second link is less than the overload rate of the third path, or if the link overload rate of the second link is equal to the overload rate of the third path and the link rate of the second link is greater than or equal to the rate of the third path, then the rate of the third path is determined to be the first rate. After S1091, the first rate is the link rate of the second link, and the second control message may also include the link overload rate of the second link and / or the link identifier of the second link; after S1092, the first rate is the rate of the third path, and the second control message may also include the overload rate of the third path and / or the link identifier of the link to which the first rate belongs.
[0132] If the first link information includes overload rate o7, speed g7, and link identifier ID7, and the second link information includes link overload rate o8, link speed g8, and link identifier ID8, then S109 may include, for example: S41, the second communication device determines the outgoing link as the second link based on the destination address of the first control message; S42, the second communication device determines whether o8 is greater than o7. If yes, then execute S44; if no, then execute S43; S43, the second communication device determines whether o8 is equal to o7 and g8 is less than g7. If yes, then execute S44; otherwise, execute S45; S44, the second communication device sets the overload rate in the first control message to o8, the speed to g8, and the link identifier to ID8; S45, the second communication device maintains the overload rate, speed, and link identifier in the first control message. It should be noted that S44 can correspond to S1091 above. After executing S44, the link information carried in the second control message obtained through S103 can be the same as the second link information of the second link. S45 can correspond to S1092 above. After executing S45, the link information carried in the second control message obtained through S103 is still the first link information.
[0133] In this case, after S103, method 100 may further include: S110, the second communication device sends a second control message to the sixth communication device via the second link. If the sixth communication device is the tail node of the first path, the steps performed by the sixth communication device can be found in [reference needed]. Figure 4 or Figure 5The description of the steps performed by the second communication device is as follows: the third control message may be constructed by the sixth communication device. S104 may include, for example, the first communication device receiving the third control message sent by the sixth communication device. If the sixth communication device is a network node in the first path, the steps performed by the sixth communication device can be found in [reference needed]. Figure 5 China's first communication device Figure 6 Second communication device or Figure 7 The first or second communication device in the process is described in relation to the steps performed. The third control message is constructed by the tail node of the first path. S104 may include, for example, the first communication device receiving the third control message sent by the tail node of the first path.
[0134] When the second communication device is the tail node of the first path, the third control message is the same as the second control message. The first rate is the target rate of the first path. After the second communication device obtains the second control message, it can execute S111: The second communication device sends the second control message to the head node of the first path.
[0135] In this case, S103 may include, for example, the second communication device constructing a response message to the first control message—a second control message—and carrying first link information in the second control message. The first link information may include, in addition to the target rate, the overload rate of the target link and / or the link identifier of the target link, where the target link is the link that limits the transmission rate of the service message to be sent on the first path. If the first link information includes the overload rate of the target link, the second control message may also include the overload rate of the target link; if the first link information includes the link identifier of the target link, the second control message may also include the link identifier of the target link.
[0136] In some possible implementations, after S104, for a service message with a transmission path of the first path, the first communication device can send the service message on the first path at a target rate. As an example, the method 100 may further include: the first communication device obtaining a first service message, the transmission path of which is the first path; then, the first communication device sending the first service message to the second communication device via the first path at a target rate. Thus, based on the target rate determined by the interaction of control messages in this embodiment, the corresponding service message is transmitted, achieving an effective balance between high bandwidth utilization and reliable network operation (which can be understood as minimizing congestion).
[0137] In order to enable the message processing method provided in this application embodiment to adapt to the dynamic changes in network load, this application embodiment also provides that by transmitting a fourth control message on the first path, the second link information maintained by each network node on the first path is updated, so that each time the above S101 to S104 are executed, each network node participates in determining the target rate based on the updated link information that best reflects the link load of the corresponding link, making it possible to determine a reasonable target rate that conforms to the current network load.
[0138] The fourth control message can be referred to as a signaling request packet (II), in contrast to the first control message, which is called a signaling request packet (I).
[0139] When the first communication device is the head node of the first path, the method 100 may further include: S51, the first communication device constructs a fourth control message, which includes the link identifier and target rate of the target link; S52, the first communication device sends the fourth control message to the second communication device through the first path, and the fourth control message is used by each network node on the first path to update the status information of the corresponding link. The status information is used to indicate the traffic status of the corresponding link, and the status information is used to calculate the link overload rate and link rate of the corresponding link. The link status information can be understood as a part of the link information, such as including Y, N, and U mentioned above, which will not be elaborated further here.
[0140] As an example, S52 could be a step triggered by the first communication device at a preset first time interval. The first time interval could be a parameter configured on the first communication device, the value of which could be flexibly set.
[0141] When the first communication device is a network node in the middle of the first path, the method 100 may further include: S61, the head node of the first path (such as the fourth communication device) constructs a fourth control message, which includes the link identifier and target rate of the target link; S62, the head node of the first path sends the fourth control message through the first path; S63, the first communication device receives the fourth control message; S64, the first communication device updates the status information of the first link based on the fourth control message, which is used to indicate the traffic status of the first link. Optionally, after S63, the method 100 may further include: the first communication device continues to forward the fourth control message through the first path so that subsequent network nodes can also update the information they maintain.
[0142] For example, S64 may include: S641, the first communication device adds the target rate carried in the fourth control message to the sum of the target rates in all fourth control messages carried on the first link to obtain the updated sum of the target rates in all fourth control messages carried on the first link, i.e., Y is updated; and / or, S642, if the link identifier of the target link is used to indicate the first link, the number of fourth control messages with the first link as the target link is incremented by one to obtain the updated number of fourth control messages with the first link as the target link, i.e., N is updated; S643, if the link identifier of the target link does not indicate the first link, the target rate carried in the fourth control message is added to the sum of the target rates in the fourth control messages carried on the first link that are not with the first link as the target link to obtain the updated sum of the target rates in the fourth control messages carried on the first link that are not with the first link as the target link, i.e., U is updated.
[0143] Assuming that in S64, the first communication device updates Y and U, or updates Y and N, then the method 100 may further include: S65, the first communication device updates the link overload rate and / or the link rate of the first link based on the updated state information. As an example, S65 may include: the first communication device obtaining the updated link overload rate of the first link based on the sum of the target rates (i.e., Y) of all fourth control messages carried on the updated first link, and the link bandwidth (i.e., Cl) of the first link; and / or, the first communication device obtaining the updated link rate of the first link based on the link bandwidth (i.e., Cl), the sum of the target rates (i.e., U) of the fourth control messages carried on the updated first link that do not target the first link, and the updated number of fourth control messages targeting the first link (i.e., N). For a detailed update process, please refer to [link to relevant documentation]. Figure 2 Related explanations.
[0144] In the case where the second communication device is a network node in the middle of the first path, the process of updating the status information of the second link and updating the corresponding link overload rate and link speed based on the updated status information can be found in the relevant operations performed by the first communication device in the case where the first communication device is a network node in the middle of the first path. It will not be repeated here.
[0145] When the second communication device is the tail node, the method 100 may further include: the second communication device receiving a fourth control message through the first path, the fourth control message including the link identifier of the target link and the target rate. Optionally, the second communication device may discard the fourth control message.
[0146] It should be noted that the fourth control message may not carry the link identifier of the target link. In this case, each network node of the first path may only update the Y in the status information of the corresponding link, and update the link overload rate based on the updated Y. In the process of determining the target rate, only the overload rate carried in the first control message and the latest link overload rate maintained by the network node may be compared. That is, the target link is determined only based on the overload rate, and the link rate of the target link is determined as the target rate. The link rate of the target link may be obtained in other ways, and the specific method of obtaining it is not limited in this embodiment.
[0147] like Figure 4 As shown, the first communication device is the head node of the first path, and the second communication device is the tail node of the first path. The method 100 may sequentially include: S105, S101, S102, S103, S111 and S104.
[0148] like Figure 5 As shown, the first communication device is a network node in the middle of the first path, the second communication device is the tail node of the first path, and the head node of the first path is the fourth communication device. The first and second communication devices also include a fifth communication device. Therefore, the method 100 can sequentially include: S106, S108, S107, S101, S102, S103, S111, and S104. It should be noted that the steps performed by the fifth communication device can be referred to S108, S107, and S101 performed by the first communication device.
[0149] like Figure 6 As shown, the first communication device is the head node of the first path, the second communication device is a network node in the middle of the first path, and the tail node of the first path is the sixth communication device. Therefore, the method 100 can sequentially include: S105, S101, S102, S109, S103, S110, and S104. Between S110 and S104, the method may further include: the sixth communication device obtaining a third control message based on the second control message and sending the third control message to the first communication device.
[0150] like Figure 7 As shown, the first communication device is network node 1 in the middle of the first path, the second communication device is network node 2 in the middle of the first path, the head node of the first path is the fourth communication device, and the tail node of the first path is the sixth communication device. Therefore, the method 100 can sequentially include: S106, S108, S107, S101, S102, S109, S103, S110, and S104. Between S110 and S104, the method can further include: the sixth communication device obtaining a third control message based on the second control message and sending the third control message to the fourth communication device.
[0151] As can be seen, through method 100, without affecting the transmission of service packets in the network, the maximum allowed transmission rate of service packets on the first path is efficiently determined by transmitting control messages in the network. On the one hand, this decouples method 100 from the user data transmission process in the network. Not only does it eliminate the need for each network node to modify the service packets, thus ensuring normal transmission of service packets in the network, but it also eliminates the need for each network node to process service packets packet by packet, making the processing granularity controllable and effectively reducing the consumption of computing and storage resources. This makes it possible to achieve high-performance service packet processing. Moreover, each communication device in method 100 only needs to add corresponding control message construction and processing logic, without requiring significant modifications to the devices in the network, especially without relying on or modifying the hardware of each device. This improves the scalability of method 100 and reduces the difficulty of network deployment. Furthermore, in method 100, each network node can continuously update the link information it maintains based on control messages, making it possible to determine the target rate based on the actual load situation in the network at the current or recent time. Transmitting service packets in the network at the newly determined target rate can achieve a relative balance between reliable network operation and high bandwidth utilization, overcome the problem of network congestion caused by excessively high transmission rates of service packets, and also avoid the problem of wasting network bandwidth caused by excessively low transmission rates of service packets.
[0152] To make method 100 clearer, the following will combine... Figure 8 and Figure 9 Two examples of method 100 are presented. Figure 8 The corresponding method 100 is the process of determining the target rate for the first path traversed by the first service flow. Figure 9 This corresponds to the process in method 100 where each network node on the first path updates the link information.
[0153] Assume that on the first path, the sender sets the following parameters: (1) ∈: the time interval for sending signaling request packets (II); (2) C r : Network card bandwidth. The sending end maintains the following parameters: (1) ID T : The identifier ID of the current bottleneck link; (2)x r Expected transmission rate. Each network node sets the following parameters: (1) ID l : The unique identifier ID of the link (non-zero); (2)C l : Link bandwidth; (3) γ l Target bandwidth utilization of the link: (3)Δ l: The time interval for updating link information (taking the update of link overload rate and link speed as examples below); (4)∈: The time interval for the sender to send signaling request packets (II). Each network node maintains the following parameters for each link segment: (1)Y l : Field x in all signaling packets (i.e., signaling request packets (II)) passing through this link r The sum; (2) N l : The number of signaling packets of the bottleneck flow passing through this link; (3) U l : Field x in the signaling packets of non-bottleneck flows passing through this link r The sum; (4)o l Link overload rate (5)g l Link rate (e.g., the fair rate of the link).
[0154] Assume that the signaling request packet (I) and the signaling response packet carry the following fields: (1) ID s : The suggested bottleneck link identifier, (2)o s : Recommended bottleneck link overload rate; (3)g s : Recommended bottleneck link rate. The signaling request packet (II) carries the following fields: (1) ID T : The link identifier of the bottleneck link currently recorded by the sending end, (2)x r : The expected transmission rate of the sending end.
[0155] like Figure 8 As shown, this example may include, for example:
[0156] For the sending end of the first path, execute the following steps S71 to S76:
[0157] S71, construct a signaling request packet (I).
[0158] S72, a signaling request packet (I) is sent through the first path.
[0159] S73, reset timer a, wait for timeout, and return to execute S71 after timeout. It should be noted that the timeout period of timer a can correspond to the second time interval mentioned above.
[0160] S74, Receive signaling response packet.
[0161] S75 reads the contents of the signaling response packet to obtain the ID. s and g s .
[0162] S76, using ID s and g s Update IDs separately T and x r .
[0163] For the network nodes of the first path, execute the following steps S81 to S86:
[0164] S81, Receive signaling request packet (I).
[0165] S82, read the contents of the signaling request packet (I) to obtain the destination address and ID. s o s and g s .
[0166] S83, retrieve the outbound link and obtain the outbound link ID. l o l and g l .
[0167] S84, determine whether o is satisfied. l Greater than o s Alternatively, determine whether o is satisfied. l equals o s And g l Less than g s If so, execute S85 to S86; otherwise, execute S86.
[0168] S85, ID l o l and g l Write signaling request packet (I).
[0169] S86, forward signaling request packet (I) from the egress link.
[0170] For the receiving end of the first path, execute the following S91 to S94:
[0171] S91, Receive signaling request packet (I).
[0172] S92, read the contents of the signaling request packet (I) to obtain the ID. s o s and g s .
[0173] S93, construct a signaling response packet and write the ID. s o s and g s .
[0174] S94, send a signaling response packet to the sender.
[0175] In this way, without affecting user data in the network, the goal of determining a reasonable target rate for the first path through which the first service flow is transmitted is achieved, thus preparing for the reasonable transmission of user data transmitted through the first path in the network.
[0176] like Figure 9 As shown, this instance may include, for example:
[0177] For the sending end of the first path, execute the following Sa1 to Sa3:
[0178] Sa1, construct signaling request packet (II).
[0179] Sa2 sends a signaling request packet (II) through the first path.
[0180] Sa3 resets timer b to ∈, waits for a timeout, and then returns to execute Sa1. It should be noted that ∈ can correspond to either the first time interval or time interval 1 mentioned above.
[0181] For the network nodes of the first path, execute the following Sb1~Sb11:
[0182] Sb1, Receive signaling request packet (II).
[0183] Sb2, read the contents of the signaling request packet (II) to obtain the destination address and ID. T and x r .
[0184] Sb3, retrieve the outbound link and obtain the Y value of the outbound link. l N l and U l .
[0185] Sb4, update Y l =Y l +x r .
[0186] Sb5, determine ID T Is it related to ID? l If they are the same, then execute Sb6 to Sb7; otherwise, execute Sb8 and Sb7 in sequence.
[0187] Sb6, Update N l =N l +1.
[0188] Sb7, forwards signaling request packet (II) from the egress link.
[0189] Sb8, Update U l =U l +x r .
[0190] Sb9, reset timer c to Δ l Waiting for a timeout, after which execution will return to Sb10. It should be noted that Δ l It can be flexibly configured.
[0191] Sb10, Update o l and g l .
[0192] Sb11, Y l N l and U l Clear to zero and return to Sb9.
[0193] It should be noted that processes Sb1~Sb8 and Sb9~Sb11 are two decoupled processes. For example, these two processes can start simultaneously for the first time, that is, when the network node receives the signaling request packet (II), the timer c is reset to Δ. l .
[0194] For the receiving end of the first path, execute the following Sc1 to Sc2:
[0195] Sc1, Receive signaling request packet (II).
[0196] Sc2, Discard signaling request packet (II).
[0197] In this way, without affecting user data in the network, the link information of each network node on the first path is continuously updated, providing a reliable data foundation for determining a reasonable target rate for the first path through which the first service flow passes.
[0198] Accordingly, embodiments of this application also provide a first communication device 1000, such as... Figure 10 As shown, the device 1000 may include, for example, a transmitting unit 1001 and a receiving unit 1002.
[0199] The sending unit 1001 is used to send a first control message to the second communication device through a first path. The first control message includes first link information, which is used to determine the target rate. The target rate is the maximum allowed transmission rate of the service message to be sent in the first path. The function of the sending unit 1001 can be corresponding to... Figures 3-7 The relevant description of S101.
[0200] The receiving unit 1002 is used to receive a second control message sent by the second communication device, the second control message carrying the target rate. The function of the receiving unit 1002 can correspond to... Figures 3-7 The relevant description of S104.
[0201] In some possible implementations, the device 1000 further includes an acquisition unit. This acquisition unit is configured to acquire a first service message, the transmission path of which is a first path; the sending unit 1001 is further configured to send the first service message to a second communication device via the first path at a target rate.
[0202] As an example, the device 1000 also includes a first construction unit. This first construction unit is used to construct a first control message. The function of the first construction unit can correspond to... Figure 4 or Figure 6 The relevant description of S105.
[0203] As an example, the target rate is the rate determined based on the first link information and the second link information of each link on the first path. The target rate belongs to the second link information of the target link on the first path, and the target link is the link that limits the transmission rate of the service message to be sent on the first path.
[0204] As an example, the first link information includes an initial overload rate and / or an initial rate, where the initial overload rate is 0 and the initial rate is related to the network card bandwidth of the first communication device; the second link information includes the link overload rate and / or the link rate of the corresponding link, where the link overload rate indicates the proportion of overloaded data on the link to the input data on the link, and the link rate is the available bandwidth of service packets with the link as the target link.
[0205] As an example, the first link information also includes an initial link identifier, and / or the second control message also includes the link identifier of the target link, with the initial link identifier being 0.
[0206] As an example, the device 1000 also includes a second construction unit. This second construction unit is used to construct a third control message, which includes the link identifier and target rate of the target link. Then, the sending unit 1001 is further used to send the third control message to the second communication device via the first path. The third control message is used by each third communication device on the first path to update the status information of the corresponding link. The status information is used to indicate the traffic status of the corresponding link and to calculate the link overload rate and link rate of the corresponding link. For example, the sending unit 1001 is specifically used to send the third control message to the second communication device via the first path at preset first time intervals.
[0207] As an example, the sending unit 1001 is specifically configured to: in response to a second time interval being reached after the time for sending a control message carrying first link information, and / or in response to a third time interval being reached after the time for receiving a control message carrying a target rate, and / or in response to an update of the link overload rate and / or link rate of the link in the first path, and / or in response to determining that there is a service message to be transmitted through the first path, send a first control message to the second communication device through the first path.
[0208] In some other possible implementations, the device 1000 further includes a first processing unit. The receiving unit 1002 is further configured to receive a fourth control message from the fourth communication device. The fourth control message carries third link information used to determine the target rate, and the fourth communication device belongs to the first path. The first processing unit is then configured to obtain the first control message based on the third link information. Figure 5 or Figure 7 The relevant description of S106 in the middle. The function of the first processing unit can correspond to Figure 5 or Figure 7 The relevant description of S107.
[0209] As an example, the device 1000 also includes a second processing unit. This second processing unit is used to determine a candidate rate based on third link information and second link information of the first link. The first link information includes the candidate rate, where the first link is the link on the first path from the first communication device to the fifth communication device, the candidate rate is the maximum allowed transmission rate of the service message to be sent on the second path, the second path is the portion of the first path from the head node to the fifth communication device, and the candidate rate is greater than or equal to the target rate of the target link on the first path, where the target link is the link that limits the transmission rate of the service message to be sent on the first path. The function of the second processing unit can be corresponding to... Figure 5 or Figure 7 The relevant description of S108.
[0210] As an example, the first link information includes the overload rate of the second path and / or the rate of the second path, where the rate of the second path is the maximum allowed transmission rate of the service message to be sent on the second path; the second link information includes the link overload rate of the first link and / or the link rate of the first link, where the link overload rate of the first link is used to indicate the proportion of overloaded data on the first link to the amount of input data on the first link, and the link rate of the first link is the available bandwidth of the service message with the first link as the target link.
[0211] As an example, the first link information may also include the link identifier of the candidate link on the second path, and / or the second control message may also include the link identifier of the target link.
[0212] As an example, the device 1000 also includes a third processing unit. The receiving unit 1002 is further configured to receive a fifth control message from the fourth communication device, the fifth control message including a link identifier and a target rate of the target link, the target link being a link that limits the transmission rate of the service message to be sent on the first path; then, the third processing unit is configured to update the status information of the first link based on the fifth control message, the status information being used to indicate the traffic status of the first link.
[0213] As an example, the device 1000 also includes a fourth processing unit. This fourth processing unit is used to update the link overload rate and link speed of the first link based on status information.
[0214] In some possible implementations, the second control message is a response message to the first control message.
[0215] It should be noted that for relevant instructions regarding device 1000, please refer to [link / reference needed]. Figures 3-7 Description of the steps performed by the first communication device.
[0216] Accordingly, this application also provides a second communication device 1100, such as Figure 11 As shown, the device 1100 may include, for example, a receiving unit 1101 and a processing unit 1102.
[0217] The receiving unit 1101 is used to receive a first control message sent by the first communication device. The first control message includes first link information, which is used to determine the target rate. The target rate is the maximum allowed transmission rate of the service message to be sent in the first path. The function of the receiving unit 1101 can be corresponding to... Figures 3-7 The relevant description of S102.
[0218] Processing unit 1102 is configured to obtain a second control message based on the first link information. The second control message includes a first rate, which is used to determine the target rate. The function of processing unit 1102 can be corresponding to... Figures 3-7 The relevant description of S103.
[0219] In some possible implementations, processing unit 1102 is further configured to determine a first rate based on first link information and second link information of the first link. The first link is the link on the first path from the second communication device to the third communication device. The first rate is the maximum allowed transmission rate of the service message to be sent on the second path. The second path is the portion of the first path from the head node to the third communication device. The first rate is greater than or equal to the target rate of the target link on the first path. The target link is the link that limits the transmission rate of the service message to be sent on the first path. The function of processing unit 1102 can correspond to... Figure 6 or Figure 7 The relevant description of S109.
[0220] As an example, the device 1100 also includes a transmitting unit. This transmitting unit is used to transmit a second control message to a third communication device via a first link. The function of the transmitting unit can correspond to... Figure 6 or Figure 7 The relevant description of S110.
[0221] As an example, the first link information includes the overload rate and / or the rate of the third path, which is the portion of the first path from the head node to the second communication device; the second link information includes the link overload rate and / or the link rate of the first link.
[0222] As an example, the first link information includes the overload rate and the speed of the third path, and the second link information includes the link overload rate and the link speed of the first link. The processing unit 1102 is specifically used to: if the link overload rate of the first link is greater than the overload rate of the third path, or if the link overload rate of the first link is equal to the overload rate of the third path and the link speed of the first link is less than the speed of the third path, then determine the link speed of the first link as the first speed; if the link overload rate of the first link is less than the overload rate of the third path, or if the link overload rate of the first link is equal to the overload rate of the third path and the link speed of the first link is greater than or equal to the speed of the third path, then determine the speed of the third path as the first speed.
[0223] As an example, if the first rate is the link rate of the first link, the second control message also includes the link overload rate of the first link; if the first rate is the rate of the third path, the second control message also includes the overload rate of the third path.
[0224] As an example, the device 1100 also includes a transmitting unit. The receiving unit 1101 is further configured to receive a third control message from the first communication device, the third control message including a link identifier and a target rate of the target link, the target link being a link that limits the transmission rate of the service message to be transmitted on the first path; the processing unit 1102 is further configured to update the status information of the first link based on the third control message, the status information indicating the traffic status of the first link; and the transmitting unit is further configured to transmit the third control message to the third communication device. For example, processing unit 1102 is specifically configured to: add the target rate carried in the third control message to the sum of the target rates in all third control messages carried on the first link to obtain the updated sum of the target rates in all third control messages carried on the first link; and / or, processing unit 1102 is specifically configured to: if the link identifier of the target link is used to indicate the first link, then increment the number of third control messages with the first link as the target link by one to obtain the updated number of third control messages with the first link as the target link; if the link identifier of the target link does not indicate the first link, then add the target rate carried in the third control message to the sum of the target rates in the third control messages carried on the first link that are not with the first link as the target link to obtain the updated sum of the target rates in the third control messages carried on the first link that are not with the first link as the target link.
[0225] As an example, processing unit 1102 is also configured to update the link overload rate and / or link speed of the first link based on status information. For example, processing unit 1102 is specifically configured to: obtain the updated link overload rate of the first link based on the sum of the target rates in all third control messages carried on the updated first link and the link bandwidth of the first link; and / or, processing unit 1102 is specifically configured to: obtain the updated link speed of the first link based on the link bandwidth of the first link, the sum of the target rates in the third control messages carried on the updated first link that do not target the first link, and the updated number of third control messages that target the first link.
[0226] In some other possible implementations, the second control message is a response message to the first control message, the first rate is the target rate, and the device 1100 further includes a sending unit. This sending unit is used to send the second control message to the head node of the first path. The function of the sending unit can correspond to... Figure 4 or Figure 5 The relevant description of S111.
[0227] As an example, the first link information includes the overload rate and / or target rate of the target link, which is the link that limits the transmission rate of the service message to be sent on the first path.
[0228] As an example, the first link information also includes the link identifier of the target link, and the second control message also includes the link identifier of the target link.
[0229] As an example, the receiving unit 1101 is also configured to receive a fourth control message through the first path, the fourth control message including the link identifier and target rate of the target link; the processing unit is also configured to discard the fourth control message.
[0230] It should be noted that for relevant instructions regarding device 1100, please refer to [link / reference needed]. Figures 3-7 Description of the steps performed by the second communication device.
[0231] Furthermore, embodiments of this application also provide a communication system 1200, such as... Figure 12 As shown. The communication system 1200 may include a first communication device 1201 and a second communication device 1202. Corresponding to method 100, the first communication device 1201 is used to implement the steps performed by the first communication device in method 100; the second communication device 1202 is used to implement the steps performed by the second communication device in method 100.
[0232] Accordingly, embodiments of this application also provide a communication device 1300, such as... Figure 13As shown. The communication device 1300 includes a processor 1301 and a memory 1302; the processor 1301 is used to execute instructions stored in the memory 1302 so that the communication device 1300 implements the method provided in the embodiments of this application.
[0233] Accordingly, this application also provides a communication device, which includes an interface and a processor; the interface is used to receive instructions and transmit them to the processor; the processor is used to implement the method provided in this application.
[0234] Accordingly, this application provides a chip 1400, such as... Figure 14 As shown. The chip 1400 includes an interface circuit 1401 and a processing circuit 1402, with the interface circuit 1401 connected to the processing circuit 1402. The interface circuit 1401 is used to perform the receiving and sending operations in the method provided in the embodiments of this application; the processing circuit 1402 is used to perform other operations in the method provided in the embodiments of this application besides the receiving and sending operations.
[0235] It should be noted that the modules or units with corresponding functions in the above communication device can be understood as interchangeable. For example, the transmitting unit 1001, the receiving unit 1002, and the receiving unit 1101 can correspond to the interface circuit 1401; the processing unit 1102 can correspond to the processor 1301 or the processing circuit 1402.
[0236] Furthermore, this application embodiment also provides a storage medium storing program code or instructions, which, when run on a processor, cause the processor to execute the method under any of the above embodiments.
[0237] Furthermore, this application also provides a program product that, when run on a processor, causes the processor to execute the method under any of the aforementioned implementations.
[0238] It should be understood that "determining B based on A" mentioned in the embodiments of this application does not mean determining B solely based on A, but also determining B based on A and / or other information.
[0239] It should be understood that the network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.
[0240] In this application, ordinal numbers such as “1”, “2”, “3”, “first”, “second”, and “third” are used to distinguish multiple objects, not to limit the order of multiple objects.
[0241] The reference to "A and / or B" in this application should be understood to include the following situations: including only A, including only B, or including both A and B.
[0242] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the methods of the above embodiments can be implemented by means of software plus a general-purpose hardware platform. Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as a read-only memory (ROM) / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or some parts of the embodiments of this application.
[0243] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system and device embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The device and system embodiments described above are merely illustrative. Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0244] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from this application, and these improvements and modifications should also be considered within the scope of protection of this application.
Claims
1. A message processing method, characterized in that, Applied to a first communication device, the method includes: A first control message is sent to a second communication device through a first path. The first control message includes first link information. The first link information is used to determine a target rate. The target rate is the maximum transmission rate allowed for the service message to be sent in the first path. The second control message sent by the second communication device is received, and the second control message carries the target rate.
2. The method according to claim 1, characterized in that, The method further includes: Obtain the first service message, the transmission path of the first service message is the first path; The first service message is sent to the second communication device via the first path at the target rate.
3. The method according to claim 2, characterized in that, The target rate is a rate determined based on the first link information and the second link information of each link on the first path. The target rate belongs to the second link information of the target link on the first path. The target link is a link that limits the transmission rate of the service message to be sent on the first path. The method further includes: Construct the first control message.
4. The method according to claim 3, characterized in that, The first link information includes an initial overload rate and / or an initial rate, wherein the initial rate is related to the network interface card bandwidth of the first communication device; The second link information includes the link overload rate and / or the link rate of the corresponding link. The link overload rate is used to indicate the proportion of overloaded data on the link to the input data of the link. The link rate is the available bandwidth of service packets with the link as the target link.
5. The method according to claim 4, characterized in that, The first link information also includes an initial link identifier, and / or the second control message also includes the link identifier of the target link.
6. The method according to claim 5, characterized in that, The method further includes: Construct a third control message, the third control message including the link identifier of the target link and the target rate; The third control message is sent to the second communication device through the first path. The third control message is used by each third communication device on the first path to update the status information of the corresponding link. The status information is used to indicate the traffic status of the corresponding link and to calculate the link overload rate and link speed of the corresponding link.
7. The method according to claim 6, characterized in that, Sending the third control message to the second communication device via the first path includes: The third control message is sent to the second communication device via the first path at a preset first time interval.
8. The method according to any one of claims 2-7, characterized in that, The step of sending the first control message to the second communication device via the first path includes: In response to a second time interval being reached after the time elapsed since sending the control message carrying the first link information, and / or in response to a third time interval being reached after the time elapsed since receiving the control message carrying the target rate, and / or in response to an update of the link overload rate and / or link rate of the link in the first path, and / or in response to determining that there is a service message to be transmitted through the first path, the first control message is sent to the second communication device through the first path.
9. The method according to claim 1, characterized in that, The method further includes: A fourth control message is received from a fourth communication device, the fourth control message carrying third link information, the third link information being used to determine the target rate, and the fourth communication device belonging to the first path; The first control message is obtained based on the third link information.
10. The method according to claim 9, characterized in that, The method further includes: Based on the third link information and the second link information of the first link, a candidate rate is determined. The first link information includes the candidate rate. The first link is the link on the first path from the first communication device to the fifth communication device. The candidate rate is the maximum allowed transmission rate of the service message to be sent on the second path. The second path is the portion of the first path from the head node to the fifth communication device. The candidate rate is greater than or equal to the target rate of the target link of the first path. The target link is the link that limits the transmission rate of the service message to be sent on the first path.
11. The method according to claim 10, characterized in that, The first link information includes the overload rate of the second path and / or the rate of the second path, where the rate of the second path is the maximum allowed transmission rate of the service message to be sent on the second path. The second link information includes the link overload rate of the first link and / or the link rate of the first link. The link overload rate of the first link is used to indicate the proportion of overloaded data on the first link to the input data of the first link. The link rate of the first link is the available bandwidth of service packets with the first link as the target link.
12. The method according to claim 11, characterized in that, The first link information also includes the link identifier of the candidate link on the second path, and / or the second control message also includes the link identifier of the target link.
13. The method according to any one of claims 9-12, characterized in that, The method further includes: The fourth communication device receives a fifth control message, which includes a link identifier of the target link and the target rate. The target link is a link that limits the transmission rate of the service message to be sent on the first path. Based on the fifth control message, update the status information of the first link, the status information being used to indicate the traffic status of the first link; Based on the status information, update the link overload rate and link speed of the first link.
14. The method according to any one of claims 1-13, characterized in that, The second control message is a response message to the first control message.
15. A message processing method, characterized in that, Applied to a second communication device, the method includes: The first control message sent by the first communication device is received. The first control message includes first link information. The first link information is used to determine the target rate. The target rate is the maximum allowed transmission rate of the service message to be sent in the first path. A second control message is obtained based on the first link information. The second control message includes a first rate, which is used to determine the target rate.
16. The method according to claim 15, characterized in that, The method further includes: Based on the first link information and the second link information of the first link, the first rate is determined. The first link is the link from the second communication device to the third communication device on the first path. The first rate is the maximum transmission rate allowed for the service message to be sent on the second path. The second path is the part of the first path from the head node to the third communication device. The first rate is greater than or equal to the target rate of the target link of the first path. The target link is the link that limits the transmission rate of the service message to be sent on the first path.
17. The method according to claim 16, characterized in that, The method further includes: The second control message is sent to the third communication device via the first link.
18. The method according to claim 16 or 17, characterized in that, The first link information includes the overload rate of the third path and / or the rate of the third path, wherein the third path is the portion of the first path from the head node to the second communication device; The second link information includes the link overload rate of the first link and / or the link speed of the first link.
19. The method according to claim 18, characterized in that, The first link information includes the overload rate and the speed of the third path; the second link information includes the link overload rate and the link speed of the first link; determining the first speed based on the first link information and the second link information of the first link includes: If the overload rate of the first link is greater than the overload rate of the third path, or if the overload rate of the first link is equal to the overload rate of the third path and the link rate of the first link is less than the rate of the third path, then the link rate of the first link is determined to be the first rate. If the overload rate of the first link is less than the overload rate of the third path, or if the overload rate of the first link is equal to the overload rate of the third path and the link rate of the first link is greater than or equal to the rate of the third path, then the rate of the third path is determined to be the first rate.
20. The method according to claim 19, characterized in that, If the first rate is the link rate of the first link, the second control message also includes the link overload rate of the first link; if the first rate is the rate of the third path, the second control message also includes the overload rate of the third path.
21. The method according to any one of claims 16-20, characterized in that, The method further includes: The first communication device receives a third control message, the third control message including the link identifier of the target link and the target rate, the target link being a link that limits the transmission rate of the service message to be sent on the first path; Based on the third control message, update the status information of the first link, the status information being used to indicate the traffic status of the first link; Based on the status information, update the link overload rate and / or the link speed of the first link, and / or send the third control message to the third communication device.
22. The method according to claim 21, characterized in that, The step of updating the status information of the first link based on the third control message includes: The target rate carried in the third control message is added to the sum of the target rates in all third control messages carried on the first link to obtain the updated sum of the target rates in all third control messages carried on the first link. And / or, If the link identifier of the target link is used to indicate the first link, then the number of third control messages with the first link as the target link is incremented by one to obtain the updated number of third control messages with the first link as the target link. If the link identifier of the target link does not indicate the first link, then the target rate carried in the third control message is added to the sum of the target rates in the third control messages carried on the first link that do not target the first link, so as to obtain the updated sum of the target rates in the third control messages carried on the first link that do not target the first link.
23. The method according to claim 21 or 22, characterized in that, The step of updating the link overload rate and / or link speed of the first link based on the status information includes: Based on the sum of the target rates in all third control messages carried on the updated first link and the link bandwidth of the first link, the updated link overload rate of the first link is obtained. And / or, Based on the link bandwidth of the first link, the updated link rate of the first link is obtained by summing the target rates of the third control messages carried on the first link that do not target the first link, and the updated number of the third control messages that target the first link.
24. The method according to claim 15, characterized in that, The second control message is a response message to the first control message, the first rate is the target rate, and the method further includes: Send the second control message to the head node of the first path.
25. The method according to claim 24, characterized in that, The first link information includes the overload rate of the target link and / or the target rate, wherein the target link is a link that limits the transmission rate of the service message to be sent on the first path.
26. The method according to claim 25, characterized in that, The first link information also includes the link identifier of the target link, and the second control message also includes the link identifier of the target link.
27. A communication device, characterized in that, The communication device includes a processor; The processor is configured to perform the method described in any one of claims 1-26.
28. A communication system, characterized in that... The communication system includes a first communication device and a second communication device, and the first path is the path from the first communication device to the second communication device; The first communication device is used to implement the method according to any one of claims 1-14; The second communication device is used to implement the method described in any one of claims 15-26.
29. A storage medium, characterized in that, The storage medium includes instructions that, when executed on a processor, cause the processor to perform the method described in any one of claims 1-26.
30. A program product, characterized in that, The program product includes a program that, when run on a processor, causes the processor to perform the method described in any one of claims 1-26.
