CSIG congestion notification method and device, electronic equipment and medium
By introducing an active feedback mechanism on the forwarding device to generate CSIG Reflection messages, the problem of delayed congestion feedback in long-path or high-latency networks by CSIG technology is solved, and faster and more accurate congestion information is delivered.
Patent Information
- Application Number
- CN202411335211.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-24
- Publication Date
- 2024-11-12
AI Technical Summary
Existing CSIG technology suffers from long congestion feedback delays in long-path or high-latency networks, making it impossible to adjust transmission strategies in a timely manner, resulting in increased packet loss rate and latency. Furthermore, it lacks compatibility with devices that do not support CSIG.
An active feedback mechanism is introduced for the forwarding device. Congestion information is monitored in real time along the path using CSIG Reflection messages and directly fed back to the sender, bypassing the traditional feedback path that relies on the receiver.
It improves the real-time performance and accuracy of congestion feedback, shortens the feedback path, reduces latency, and ensures complete information transmission even in situations where CSIG devices are not supported.
Smart Images

Figure CN118945121A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of network communication technology, and in particular relates to a CSIG congestion notification method, device, electronic equipment and medium. Background Art
[0002] With the development of network technology, network traffic is increasing day by day. How to effectively manage and control network congestion has become a key technical issue. In existing network congestion control schemes, typical congestion notification mechanisms rely on feedback from the receiving end. For example, TCP-based congestion control schemes determine whether the network is congested by detecting packet loss or increased latency at the receiving end, and alleviate congestion by adjusting the sending rate of the data sender. However, this passive congestion control method usually has a long feedback delay and is not suitable for scenarios with high real-time requirements.
[0003] In order to solve the above problems, an end-to-end congestion notification method based on CSIG (Congestion Signaling) has been proposed in recent years. In this method, the sender encapsulates CSIG information in the message and sends it along the data transmission path so that the forwarding devices on the path can detect the bandwidth and delay of the node.
[0004] In the prior art, the receiving end usually feeds back CSIG congestion information to the sending end, which means that the sending end can only obtain congestion information on the network path after the message reaches the receiving end. For long paths or high-latency networks, this method causes a large delay in congestion notification and fails to adjust the sending strategy in time, thereby increasing the packet loss rate and latency, and reducing network transmission performance.
[0005] Therefore, it is necessary to provide a new solution to the above technical problems. Summary of the invention
[0006] The object of the present invention is to provide a CSIG congestion notification method, device, electronic device and medium, which can improve the real-time and accuracy of congestion feedback by introducing an active feedback mechanism of a forwarding device.
[0007] To achieve the above purpose, the technical solution provided by the present invention is as follows:
[0008] In a first aspect, the present invention provides a CSIG congestion notification method, which includes:
[0009] The sending end sends a CSIG message along a transmission path; the forwarding device on the transmission path obtains bottleneck information of the local jump device according to the CSIG message; the forwarding device determines whether the local jump device generates a CSIGReflection message according to a predetermined condition, and the CSIGReflection message encapsulates CSIG congestion information; if so, the forwarding device generates a CSIGReflection message and sends the CSIGReflection message to the sending end.
[0010] In one or more embodiments, the method includes: the forwarding device determines whether the congestion level of the current jump device is higher than the bottleneck information carried by the CSIG message, and the congestion information includes the export available bandwidth, the export available bandwidth occupancy rate and the delay; if so, the forwarding device generates a CSIG Reflection message and sends the CSIG Reflection message to the sending end.
[0011] In one or more implementations, the method includes: the forwarding device determines whether the next-hop device supports CSIG; if not, the forwarding device generates a CSIG Reflection message, and sends the CSIG Reflection message to the sending end.
[0012] In one or more implementations, the method includes: the forwarding device determines whether the CSIG message is discarded in the local jump device; if so, the forwarding device generates a CSIG Reflection message and sends the CSIG Reflection message to the sending end.
[0013] In one or more embodiments, the method further includes: when the forwarding device receives the CSIG message for the first time, parsing the source IP, destination IP and bottleneck information of the CSIG message, and storing these fields as key values in the CSIG flow table.
[0014] In one or more embodiments, the method further includes: the forwarding device compares the bottleneck information of the current jump device with the bottleneck information in the CSIG flow table; if the congestion level of the current jump device is higher than the bottleneck information in the CSIG flow table, the bottleneck information in the CSIG flow table is updated.
[0015] In one or more implementations, the method further includes: the forwarding device periodically scanning the CSIG flow table, and if a flow table entry does not receive a corresponding CSIG message within multiple cycles, deleting the flow table entry.
[0016] In a second aspect, the present invention provides a CSIG congestion notification device, comprising:
[0017] A sending module is used to enable a sending end to send a CSIG message along a transmission path; an obtaining module is used to enable a forwarding device on the transmission path to obtain bottleneck information of a local jump device according to the CSIG message; a judging module is used to enable the forwarding device to judge whether the local jump device generates a CSIG Reflection message according to a predetermined condition, wherein the CSIG Reflection message is encapsulated with CSIG congestion information; and a generating module is used to enable the forwarding device to generate a CSIG Reflection message when the judging condition is met, and send the CSIG Reflection message to the sending end.
[0018] In a third aspect, the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the CSIG congestion notification method as described above when executing the program.
[0019] In a fourth aspect, the present invention provides a computer-readable medium, wherein the computer-readable medium carries computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the CSIG congestion notification method as described above.
[0020] Compared with the prior art, the CSIG congestion notification method, device, electronic device and medium provided by the present invention improve the real-time and accuracy of congestion feedback by introducing an active feedback mechanism of a forwarding device; the forwarding device is no longer just an information node that passively transmits messages, but can actively generate and send congestion feedback messages according to the network status of the node; by directly feeding back bottleneck information in the path from the forwarding device to the sending end when congestion occurs, the feedback path is shortened and the delay of congestion feedback is reduced. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0022] Figure 1 Flow chart of a CSIG congestion notification method according to an embodiment of the present invention;
[0023] Figure 2 This is a CSIG message forwarding process in a network in an embodiment of the present invention;
[0024] Figure 3 This is a CSIG message forwarding process in a network in an embodiment of the present invention;
[0025] Figure 4 This is a CSIG message forwarding process in a network in an embodiment of the present invention;
[0026] Figure 5 It is a structure in which CSIG Reflection is encapsulated as TCP Option in one embodiment of the present invention;
[0027] Figure 6 The encapsulation structure of the CSIG Reflection message as the payload part in the UDP header in one embodiment of the present invention;
[0028] Figure 7 A schematic diagram of a CSIG congestion notification device according to an embodiment of the present invention;
[0029] Figure 8 FIG. 4 is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0030] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0031] The technical implementation idea of the present invention is mainly derived from the limitations and shortcomings of the existing CSIG (congestion signaling) technology in congestion notification. With the expansion of network scale and the increase of complexity, the congestion situation in the network is becoming increasingly complicated, and the traditional congestion control mechanism that relies on feedback from the receiving end can no longer adapt to the needs of efficient and real-time congestion control. Although the existing CSIG technology can provide assistance for end-to-end congestion control to a certain extent, it has obvious deficiencies in feedback delay, information loss and compatibility with devices that do not support CSIG, and cannot fully meet the requirements of efficient and fast congestion feedback.
[0032] In the analysis of the prior art, the inventors found that the current CSIG solution mainly relies on the receiving end to feedback the congestion information in the path to the sending end. This method is prone to large delays due to the long feedback path, especially in high-latency or long-distance networks, and the feedback efficiency is significantly reduced. In addition, when dealing with devices that do not support the CSIG protocol, the prior art usually causes the CSIG message to be stripped or discarded, resulting in the sender being unable to accurately know the congestion status in the path, affecting the network's traffic regulation capabilities. At the same time, if the CSIG message is discarded in the path, the prior art also lacks an effective response mechanism and cannot timely feedback the collected congestion information.
[0033] In order to overcome these defects, the present invention proposes a new technical idea, which greatly improves the real-time and accuracy of congestion feedback by introducing an active feedback mechanism of the forwarding device. Under this new idea, the forwarding device is no longer just an information node that passively transmits messages, but can actively generate and send congestion feedback messages based on the network status of this node. By directly feedback the bottleneck information in the path from the forwarding device to the sending end when congestion occurs, the feedback path is shortened and the delay of congestion feedback is reduced. In addition, the present invention also generates feedback messages at devices that do not support CSIG on the path, ensuring that in the presence of unsupported devices, the congestion information in the path can still be fully notified.
[0034] Please refer to Figure 1 FIG. 1 is a flow chart of a CSIG congestion notification method in an embodiment of the present invention. The CSIG congestion notification method specifically includes the following steps:
[0035] S101: The sending end sends a CSIG message along a transmission path.
[0036] The sender sends CSIG messages containing network status information periodically or on demand through the network transmission path to trigger congestion status detection in the forwarding devices on the path and eventually form a feedback mechanism. The specific implementation method may vary depending on the network environment and application requirements.
[0037] In an exemplary network scenario, the sender can be a server, router or other network device that is responsible for data transmission. The sender will generate a CSIG message and inject it into the network transmission path according to the preset congestion monitoring mechanism. The CSIG message can contain some initial network status information, such as the current available bandwidth, end-to-end delay, etc. This information provides a benchmark value for the forwarding devices on the path so that the devices can compare when detecting their own status.
[0038] The sending of CSIG messages can be periodic or triggered based on specific conditions. For example, when network traffic reaches a certain threshold, or when potential congestion risks are predicted based on historical data, the sender can choose to send CSIG messages for monitoring. This flexible sending mechanism not only adapts to different network topologies and traffic patterns, but also dynamically adjusts the sending frequency and content of CSIG messages according to changes in the network environment.
[0039] S102: The forwarding device on the transmission path obtains bottleneck information of the local jump device according to the CSIG message.
[0040] It should be noted that each network node (forwarding device) on the transmission path can evaluate the network status of the current device (local jump device) based on the information in the CSIG message, especially key parameters related to congestion, such as bandwidth, latency, etc.
[0041] In the exemplary network path, the forwarding devices include routers, switches or other network nodes with forwarding functions. When these devices receive the CSIG message sent by the sender, they will parse the message and extract the key congestion information in the message. This information usually includes the initial bandwidth, delay, egress bandwidth occupancy, etc. The purpose of this step is to allow each node to obtain and compare its current status based on the information carried in the message.
[0042] In an exemplary embodiment, when the forwarding device receives a CSIG message for the first time, it parses the source IP, destination IP and bottleneck information of the CSIG message, and stores these fields as key values in the CSIG flow table, thereby establishing a basis for subsequent CSIG message processing and congestion feedback. By storing these fields, the forwarding device can quickly find matching flow table entries, thereby comparing and updating congestion information.
[0043] In the specific implementation, when the forwarding device (such as a switch or router) receives the first CSIG message, it will first parse the header information of each layer of the message. Specifically, the CSIG message contains the L2 layer, IP layer, transport layer (such as TCP or UDP), and CSIG congestion signaling information. The forwarding device extracts the source IP address and destination IP address of the message by reading the L2 header and IP header. These two information are used to identify the path of the data flow. In addition, the bottleneck information in the message (such as latency and bandwidth) will be used to monitor and compare network performance. The source IP and destination IP can usually uniquely identify a data flow in the network, and the bottleneck information is used to determine whether the CSIG message needs to be updated.
[0044] These parsed fields are then stored as key values in the CSIG flow table of the forwarding device. The CSIG flow table is a data structure used in the forwarding device to manage and track the status of CSIG messages. It contains fields such as source IP, destination IP, bottleneck information, etc. Through this flow table management mechanism, when the device receives subsequent CSIG messages, it can quickly obtain the associated historical congestion information by searching for matching flow table entries, and perform comparisons and updates.
[0045] For example, a company's internal server is transmitting data with a remote data center. The server's source IP is 192.168.1.10 and the destination IP is 10.0.0.20. When the forwarding device receives the CSIG message on this path for the first time, it parses the source IP and destination IP in the message and extracts the initial bottleneck information, such as bandwidth 500Mbps and latency 20ms. The device stores this information in the CSIG flow table and records it as a flow table entry. In the future, when the device receives a CSIG message on the same path again, it can quickly find the corresponding record in the flow table and compare the current bottleneck information with the stored historical value to determine whether it is necessary to update the CSIG message or generate a CSIG Reflection message.
[0046] In a specific embodiment, the CSIG flow table structure may be set as follows:
[0047] FlowId Source IP Destination IP CSIG Type CSIG signal type CAIG signal value 1 SourceIP1 DestIP1 Compact 0 xxx 2 SourceIP2 DestIP2 Expanded 1 xxx 3 SourceIP3 DestIP3 Expanded 2 xxx …… …… …… …… …… ……
[0048] Among them, in the CSIG signal type: 0 represents the minimum available bandwidth, 1 represents the minimum available bandwidth utilization, and 2 represents the maximum hop-by-hop delay.
[0049] The main function of the CSIG flow table mechanism is to improve the efficiency and accuracy of forwarding devices in processing CSIG messages. By maintaining the CSIG flow table, forwarding devices do not need to reparse and process the entire message link every time, but can quickly find the corresponding data flow and compare congestion information. This not only reduces the processing time of the device, but also ensures that the congestion information of each path can accurately reflect the current network status.
[0050] In an exemplary embodiment, the forwarding device compares the bottleneck information of the current jump device with the bottleneck information in the CSIG flow table. If the congestion level of the current jump device is higher than the bottleneck information in the CSIG flow table, the bottleneck information in the CSIG flow table is updated.
[0051] By comparing historical records with real-time data, the bottleneck information of the device is dynamically adjusted and updated to ensure that the congestion information of each node can reflect the true status of the current network. If the congestion level of the current device exceeds the historical value in the flow table, the flow table needs to be updated to continue monitoring and feedback more accurate bottleneck information in the future.
[0052] In the specific implementation, the forwarding device parses the CSIG message to obtain the bottleneck information it carries, such as bandwidth, latency, etc. This bottleneck information reflects the status of the previous node on the path. At the same time, the forwarding device monitors its own export bandwidth, latency, and other network performance indicators. These current device states reflect the performance of this node in data transmission. When comparing these data, the forwarding device compares the actual bandwidth and latency information of this node with the bottleneck information stored in the flow table. If it is found that the network performance of the current device (such as reduced bandwidth or increased latency) is worse than the value in the CSIG flow table, it means that the network status of the current node has deteriorated, and the bottleneck information in the CSIG flow table needs to be updated to ensure that the congestion information on the path can accurately reflect the bottleneck status of the current node.
[0053] This process can be achieved through hardware modules that regularly monitor network performance parameters, such as egress bandwidth monitors and latency detection modules. These modules can provide real-time feedback on the current network status of the device and automatically compare these statuses with historical data in the CSIG flow table. Once the current device status is found to be worse than the bottleneck information in the flow table, the device will automatically trigger an update operation and store the new bottleneck information in the flow table.
[0054] The purpose of this mechanism is to ensure that the bottleneck information carried in the CSIG message can be updated in real time, thereby reflecting the latest status of each node in the network. Without such a mechanism, the bottleneck information of earlier nodes in the network may be retained in the flow table. Even if the congestion of subsequent nodes is more serious, the sender cannot be informed of the latest bottleneck status in time, resulting in the inability to effectively adjust the transmission strategy. By updating the flow table in real time, the latest bottleneck information on each path can be accurately recorded and fed back, thereby improving the congestion control efficiency of the entire network.
[0055] In an exemplary embodiment, the forwarding device periodically scans the CSIG flow table. If a flow table entry does not receive a corresponding CSIG message within multiple cycles, the flow table entry is deleted to ensure that the items in the flow table can be kept updated and efficiently managed, preventing expired or no longer active data from occupying system resources, thereby improving the performance of the device and the efficiency of message processing.
[0056] In the specific implementation, the forwarding device will set up a periodic scanning and aging mechanism for each flow table entry stored in the CSIG flow table. This mechanism periodically scans all items in the CSIG flow table to determine whether each item has received the corresponding CSIG message within a specified time period. If a flow table entry does not receive the corresponding message within multiple periods (such as several time units or polling periods), this usually indicates that the data flow is no longer active or the network path has changed. At this time, the forwarding device will mark the flow table entry as expired and delete it to free up storage space and computing resources. This process is called the "aging" mechanism of the flow table, which can ensure that the device's flow table only retains active message path information.
[0057] This periodic scanning can be implemented through the device's built-in timer or scheduled scanning tasks. For example, the forwarding device can set a specific time period (such as every 10 seconds) to check all CSIG flow table entries. When each flow table entry is generated, it will be accompanied by a timestamp to record the last time the item was updated or matched a CSIG message. The scanning task will periodically check the timestamp of each flow table item. If it is found that the timestamp of an item exceeds the set threshold (for example, no new CSIG messages have been matched for more than three time periods), the item will be marked as no longer active and deleted from the flow table.
[0058] S103: The forwarding device determines whether the current jump device generates a CSIG Reflection message according to a predetermined condition, where the CSIG Reflection message is encapsulated with CSIG congestion information.
[0059] The forwarding device obtains bottleneck information such as bandwidth, latency and other parameters by parsing the CSIG message. These parameters constitute the benchmark values for comparison, and the actual current status of the local device (this node) can be obtained through the real-time monitoring module of the device. The judgment logic of the predetermined conditions is mainly to compare these values. When it is found that the congestion level of this node exceeds the benchmark value carried in the CSIG message, it will trigger the process of generating a CSIG Reflection message. These predetermined conditions may include the following situations:
[0060] ① Bandwidth change: If the forwarding device finds that the egress bandwidth of this node is less than the bandwidth value in the CSIG message, for example, the bandwidth in the message is 500Mbps, but the egress bandwidth of this node is only 300Mbps, it means that the bandwidth resources of this node have become a bottleneck. At this time, a CSIG Reflection message should be generated to notify the sender of the current network bottleneck.
[0061] ② Increase in delay: The forwarding device will also monitor its internal forwarding delay. If it detects that the delay of this node is greater than the reference value specified in the CSIG message, for example, the delay carried in the CSIG message is 20ms, and the actual delay of this node has reached 35ms, this means that the transmission delay of this node is increasing and the network may be in a congested state. At this time, it is also necessary to generate a CSIG Reflection message.
[0062] ③ CSIG support of the next-hop device: If the forwarding device detects that the next node does not support the CSIG protocol, it means that the CSIG message cannot continue to be transmitted in the path. In this case, the forwarding device will strip the CSIG header and generate a CSIG Reflection message based on the existing congestion information, and directly feed it back to the sender. This ensures that even if some devices do not support CSIG, the bottleneck information in the network can still be transmitted in a timely manner.
[0063] By judging these predetermined conditions, the forwarding device can flexibly decide whether to generate a CSIG Reflection message. For example, suppose in a video conference, the video stream is transmitted from the sender to multiple participant devices. The CSIG message propagates along the path. When a forwarding device detects that its egress bandwidth has dropped due to other high-traffic tasks, the device will determine that the bandwidth is insufficient according to the predetermined conditions and actively generate a CSIG Reflection message to feed back the bottleneck information to the sender. After receiving this information, the sender can immediately adjust the transmission rate of the video stream to avoid further network congestion.
[0064] The role of this step is to enhance the intelligent feedback capability of the network, so that the forwarding device can actively generate feedback messages based on real-time conditions, rather than passively waiting for feedback from the receiving end. Through this mechanism, congestion information in the network can be transmitted to the sending end more promptly, shortening the feedback path, reducing latency, and improving the real-time and accuracy of feedback.
[0065] S104: If yes, the forwarding device generates a CSIG Reflection message, and sends the CSIG Reflection message to the sending end.
[0066] The forwarding device will decide whether to generate a CSIG Reflection message based on the judgment result in the previous step S103. If it is detected that the congestion of this node (such as export bandwidth or delay) has exceeded the benchmark value set in the CSIG message, the forwarding device will immediately construct a CSIG Reflection message. The message encapsulates the bottleneck information of this node, including export bandwidth, delay and other key parameters that may affect network performance. After generating the message, the forwarding device sends it directly back to the sending end, bypassing the traditional path that relies on feedback from the receiving end.
[0067] Traditional congestion control mechanisms usually rely on the receiver to feedback the congestion situation in the network, but this passive feedback method is prone to long feedback delays. Especially in long-distance networks, the adjustment measures taken by the sender often lag behind the actual congestion situation. By actively generating CSIG Reflection messages through forwarding devices, the network can pass bottleneck information to the sender in the shortest time, helping the sender to adjust the transmission strategy in time and improving the network response speed.
[0068] In an exemplary embodiment, the forwarding device determines whether the congestion level of the current jump device is higher than the bottleneck information carried by the CSIG message, and the congestion information includes the available export bandwidth, the available export bandwidth occupancy rate and the delay; if so, the forwarding device generates a CSIG Reflection message and sends the CSIG Reflection message to the sending end.
[0069] The forwarding device evaluates the network congestion status in real time by monitoring the usage of its own network resources, especially key congestion parameters such as the available bandwidth at the egress, the available bandwidth occupancy rate, and latency. If it finds that its own congestion level exceeds the bottleneck information carried in the CSIG message, it means that this node may become a network bottleneck. The forwarding device will immediately generate a CSIG Reflection message and feed back this real-time network status information to the sender to help the sender adjust the data transmission strategy to avoid further congestion.
[0070] In the specific implementation, the forwarding device will first parse the CSIG message to obtain key information from it. Assume that the information carried in the CSIG message shows that the allowed export bandwidth in the path is 500Mbps, the export bandwidth utilization rate is 80%, and the delay is 20 milliseconds. At this time, the forwarding device needs to obtain its own real-time network status. For example, through the built-in bandwidth monitoring module, it detects that the current export bandwidth is only 300Mbps, the bandwidth utilization rate is 90%, and the delay has reached 35 milliseconds. This shows that the actual network status of the forwarding device is more serious than the bottleneck information defined in the CSIG message, indicating that the current node may have become the congestion bottleneck of the network. In order to prevent the network from further deteriorating, the forwarding device will trigger the generation of the CSIG Reflection message at this time, encapsulate the latest bottleneck information in it, and send it directly back to the data sender.
[0071] In practical applications, such as video streaming or online games, which require high network performance, it is crucial to promptly discover and feedback network congestion information. If the forwarding device monitors its own bandwidth and latency and finds that the network load of this node has exceeded the set threshold, the sender can quickly obtain this bottleneck information by generating a CSIG Reflection message, and adjust the video bit rate, transmission rate and other strategies accordingly to reduce the subsequent packet loss, delay or freeze. Compared with the traditional mechanism that relies on the receiver to feedback congestion information, this method can significantly shorten the congestion feedback path, allowing the sender to make adjustments more quickly.
[0072] Please refer to Figure 2 The figure shows the forwarding process of CSIG message in the network in one embodiment of the present invention. The sending end (Server) sends the CSIG message, and the sending end server generates and sends a message with CSIG Tag (CSIGTag0), which arrives at the switch after being transmitted through the network. The message contains L2 Header (data link layer header), IP Header (network layer header), IP Header (may contain UDP or TCP protocol header) and Payload. These headers carry relevant information about the transmission path and CSIG signaling data, which are used to detect network performance along the way, such as bandwidth and latency.
[0073] The switch receives and processes the CSIG message. After receiving the message with the CSIG Tag, the switch will parse the message. When the switch finds that the delay (PD) of the local device is greater than the delay value carried in the CSIG message, or the available bandwidth (ABW) of the egress is smaller than the value in the message, it means that the current device is in a congested state and becomes a bottleneck node in the path. At this time, the switch will update the bottleneck information of this node based on the information in the CSIG message.
[0074] When the switch detects congestion, it updates the bottleneck information of the current node into the CSIG message. The updated CSIG message carries a new CSIGTag1, reflecting the latest bandwidth and delay information.
[0075] At the same time, a CSIG Reflection message is generated and sent. To ensure that the source server can obtain detailed information about the congested node in a timely manner, the switch will actively generate a CSIG Reflection message. The structure of the CSIG Reflection message includes L2 Header, IP Header, UDP Header, and Payload. The Payload encapsulates the bottleneck information of the current node, such as egress bandwidth and latency. The CSIG Reflection message is sent directly to the source server through the source information learned by the flow table.
[0076] The switch continues to forward the updated CSIG message. After generating the CSIG Reflection message, the switch continues to forward the updated CSIG message (CSIGTag1) to the next node. The message continues to be transmitted along the transmission path until it reaches the destination server (Server).
[0077] The destination server processes the CSIG message. After receiving the final CSIG message, the destination server can also generate a CSIG Reflection message based on the latest bottleneck information in the message and return it to the source server. This process ensures that the status of all nodes in the path can be fed back to the sender to help it optimize the data transmission strategy.
[0078] In an exemplary embodiment, the forwarding device determines whether the next-hop device supports CSIG; if not, the forwarding device generates a CSIG Reflection message and sends the CSIG Reflection message to the sending end.
[0079] When the forwarding device detects that the next-hop device does not support the CSIG protocol, it generates a CSIG Reflection message and sends it back to the sender. This process ensures that when there are nodes that do not support CSIG in the path, the congestion status of the current path can still be fed back in time to prevent the loss of congestion information. This mechanism is designed to improve the compatibility of the CSIG protocol and the integrity of network feedback.
[0080] In actual network transmission, the transmission path may pass through multiple different network nodes, and the devices at these nodes may have different protocol support capabilities. Some devices may not be able to parse or process CSIG messages. When such a device is used as the next hop, if the CSIG message continues to be forwarded, the CSIG information will not be processed correctly, resulting in the message header being stripped or the message being discarded, and congestion information being lost. To avoid this situation, the forwarding device will detect whether the device supports the CSIG protocol before forwarding to the next hop. If it is found that the next hop device does not support CSIG, the forwarding device will not continue to forward the CSIG message, but will immediately generate a CSIG Reflection message and feed back the bottleneck information collected along the way to the sender.
[0081] In specific implementation, the forwarding device needs to have the ability to detect the protocols supported by the next-hop device, which can be achieved by consulting the configuration table or protocol identification mechanism of the next-hop device. Once it is detected that the next-hop device does not support the CSIG protocol, the forwarding device will intercept the CSIG message, extract the congestion information collected in it, such as egress bandwidth, latency, etc., and generate a CSIG Reflection message, encapsulate this information in it, and send it back to the sender.
[0082] Please refer to Figure 3 FIG. 1 is a flow chart of forwarding CSIG messages in a network according to an embodiment of the present invention. When a device that does not support CSIG appears in a path, the forwarding device can generate a CSIG Reflection message to feedback the collected bottleneck information, thereby avoiding the loss of important congestion information.
[0083] The source server first sends a message with a CSIG Tag (CSIGTag0), which is transmitted through the network path and contains the L2 (data link layer) header, IP header, and Payload. The original intention of the CSIG message is to collect and monitor the network status (such as bottleneck information such as latency and bandwidth) at nodes along the way, so as to help the sender adjust the transmission strategy in time.
[0084] The message arrives at a forwarding device that supports CSIG (Switch 1). When the message is transmitted to a forwarding device that supports the CSIG protocol, the device parses the information in the message, extracts and compares the network bottleneck status of the current device. For example, the device obtains its egress bandwidth and latency, and compares it with the information in the CSIG message. If the latency or bandwidth of this node is worse, the device updates the bottleneck information in the CSIG message and continues to forward the message to the next hop device.
[0085] Since the next hop device (Switch 2) does not support the CSIG protocol. Figure 3As shown in the figure, when the message reaches a device that does not support the CSIG protocol, the device cannot correctly parse the CSIG message and cannot continue to forward the message. In order to avoid losing the collected network bottleneck information, switch 1 will strip the CSIG header of the CSIG message and generate a feedback message through the following steps:
[0086] ① Generate CSIG Reflection message: After stripping the CSIG header, switch 1 actively generates a CSIGReflection message. The format of this message is different from that of the CSIG message. Its purpose is to feed back the collected network bottleneck information to the source server in another encapsulation method. The CSIG Reflection message contains the L2 header, IP header, UDP header and Payload. The Payload encapsulates the network bottleneck information collected by the forwarding device, including key information such as latency and bandwidth.
[0087] ② Feedback information to the source: The CSIG Reflection message is returned to the source server through the network. After receiving this feedback message, the server can adjust the transmission strategy according to the feedback content, such as reducing the sending rate or reducing data traffic to avoid further congestion.
[0088] ③ Update flow table: While generating and sending CSIG Reflection messages, switch 1 will update the latest bottleneck information to the flow table. Through this flow table management mechanism, the device can record the network status of this node and quickly find relevant historical information during future message transmission, reducing processing time.
[0089] In an exemplary embodiment, the forwarding device determines whether the CSIG message is discarded in the local jump device; if so, the forwarding device generates a CSIG Reflection message and sends the CSIG Reflection message to the sending end.
[0090] The forwarding device usually tracks the processing status of the CSIG message when processing it. If the CSIG message is discarded on the local device due to a full queue, insufficient buffer, or other network problems, the forwarding device needs to detect and record this event. To ensure timely and accurate feedback of network congestion information, the forwarding device will immediately generate a CSIG Reflection message when it detects that the CSIG message is discarded. This message contains congestion information that the forwarding device has previously collected, such as the current egress bandwidth, latency, and other parameters. By generating a CSIG Reflection message, this important network information can be delivered to the sender in a timely manner, helping the sender adjust its transmission strategy to avoid further deterioration of congestion.
[0091] For example, a forwarding device may process a large number of data packets at the same time, and its buffer may be saturated, causing the received CSIG message to be unable to be forwarded and discarded. In this case, the forwarding device will immediately detect the packet loss event. In order to prevent the loss of network bottleneck information, the forwarding device will generate a CSIG Reflection message to feed back the export bandwidth and latency information previously collected by this node to the sender. After receiving this feedback information, the sender can immediately adjust the data transmission rate to reduce the transmission pressure and ensure that no serious delays or disconnections occur during the game.
[0092] The core function of this mechanism is to ensure the integrity of congestion information in the network. Even if the CSIG message is discarded, the key bottleneck information can still be delivered to the sender in a timely manner, helping the network to quickly adapt to the dynamically changing transmission environment. By allowing the forwarding device to actively generate feedback messages, the delay problem of the traditional passive feedback method that relies on the receiving end is avoided. In contrast, without this mechanism, once the CSIG message is lost, the congestion information will not be fed back to the sender, causing the sender to continue to transmit data at the current rate, further exacerbating network congestion.
[0093] Please refer to Figure 4 As shown in FIG. 1 , the forwarding process of the CSIG message in the network in one embodiment of the present invention is shown. When the CSIG message is discarded at a node in the forwarding path, the source server generates a message with a CSIG Tag (CSIGTag0) and transmits it to the destination server through the network. The CSIG message contains the data transmitted in the network path and carries the bottleneck information about the network status (such as the egress bandwidth, delay, etc.), which is used to detect the network status of the nodes along the way.
[0094] The message passes through the forwarding device (switch 1) that supports the CSIG protocol. When the message reaches the switch that supports CSIG, the switch will parse the CSIG message, extract the bottleneck information, and compare it with its own network status. Assuming that the network status of this node (such as bandwidth and latency) is more serious than the bottleneck information recorded in the message, the switch will update the bottleneck information in the message and continue to forward the CSIG message. The updated message is as follows: Figure 4 The mark CSIGTag1 indicates that the current network bottleneck information has been updated.
[0095] The packet is discarded at the next hop device (Switch 2). When the packet passes through Switch 2, due to network congestion or other reasons of the device, the CSIG packet cannot be transmitted and is discarded. In this case, Switch 2 cannot pass the CSIG packet to the destination server, resulting in the bottleneck information in the path cannot be transmitted back to the source end through normal CSIG packets.
[0096] To avoid congestion information loss, switch 1 actively generates a CSIG Reflection message. When switch 1 detects that a CSIG message is discarded by a downstream device, it immediately generates a CSIGReflection message based on the previously stored flow table information. This reflection message is used to feed back bottleneck information on the current path to the source server. The CSIGReflection message contains an L2 header, an IP header, a UDP header, and a Payload. The Payload part encapsulates the network bottleneck information collected by the current device, including important parameters such as egress bandwidth and latency.
[0097] At the same time, switch 1 updates the flow table. While generating and sending the CSIG Reflection message, switch 1 will update the latest bottleneck information to the flow table so that it can quickly find historical bottleneck information when processing future messages and continue to monitor the current network status. Through this mechanism, the forwarding device can ensure that the bottleneck information is saved and fed back to the source end in a timely manner.
[0098] The CSIG Reflection message is fed back to the source end. The generated CSIG Reflection message is directly fed back to the source end server through the network in the defined format. After the source end receives the reflection message, it can analyze and adjust the data transmission strategy, such as reducing the transmission rate, reducing the video stream bit rate or reducing the amount of data sent to alleviate network congestion.
[0099] In an exemplary embodiment, the encapsulation of the CSIG Reflection message can be implemented through TCP Option and UDP Payload. The main purpose of the CSIG Reflection message is to feedback the bottleneck information (such as latency, bandwidth, etc.) in the network through four-layer protocol encapsulation for network telemetry or remote control.
[0100] Please refer to Figure 5 As shown in the figure, CSIG Reflection is encapsulated as a TCP Option structure. This encapsulation method uses the option field in the TCP protocol to create a new TCP Option for CSIG Reflection. TCPOption is an optional field in the TCP message, which is used to carry additional information outside the normal TCP message header. The specific structure is divided into the following parts:
[0101] Kind field: This is a unique code point that identifies the option as a CSIG option. The Kind field of a TCP option tells the receiving device the type of option. By assigning a unique code to CSIG, network devices are able to distinguish between CSIG Reflection and other TCP options (such as the timestamp option or the window scaling option).
[0102] Length field: indicates the length of the option. Since the length of the TCP option field is variable, the Length field is used to indicate the total number of bytes of the option. This ensures that the receiving end can correctly parse the option.
[0103] CSIG data fields: This is where the data part of CSIG Reflection is stored, i.e., the bottleneck information reflected from the receiving end to the sending end. These data fields usually contain congestion information such as network latency and egress bandwidth, which are used by the sending end to adjust its transmission strategy.
[0104] As a TCP Option, the CSIG Reflection message encapsulation method can use the existing TCP connection to carry additional network status information. This method has good compatibility because most network devices and transmission paths support the TCP option mechanism. Through this encapsulation method, the CSIG message can be directly embedded in the existing TCP connection without additional overhead or special processing, while ensuring that congestion information can be delivered to the sender.
[0105] Please refer to Figure 6 The figure shows the encapsulation structure of the CSIG Reflection message as the payload part in the UDP header. In this encapsulation structure, the CSIG Reflection message is placed in the payload area of the UDP message and transmitted in the standard UDP message format. The UDP protocol is different from the TCP protocol and does not have an option field, so CSIG Reflection exists directly in the form of payload. The specific structure is as follows:
[0106] Flags field: This field is usually used to carry some status or control information, such as whether further reflection is required or whether certain network parameters need to be adjusted. The Flags field can be used to define special control instructions for the CSIG Reflection message.
[0107] CSIG data fields: CSIG data fields carry the core data part of the CSIG Reflection message, including network bottleneck information, such as bandwidth, latency, packet loss rate, etc. This information is directly transmitted to the sender through UDP messages for network tuning.
[0108] The main purpose of using UDP payload to encapsulate CSIG Reflection messages is for connectionless application scenarios. Due to its lightweight and stateless characteristics, UDP protocol is often used in data transmission applications that require low latency and high performance, such as real-time video streaming, online games, etc. By encapsulating CSIG Reflection messages in UDP payload, it can be ensured that CSIG congestion information can also be effectively transmitted and fed back in these connectionless applications. Since the UDP protocol has no retransmission mechanism, using this encapsulation method can reduce network overhead and is suitable for scenarios with high real-time requirements.
[0109] Please refer to Figure 7 As shown, based on the same inventive concept as the aforementioned CSIG congestion notification method, the present invention provides a CSIG congestion notification method 700 , which includes: a sending module 701 , an acquiring module 702 , a judging module 703 and a generating module 704 .
[0110] The sending module 701 is used to enable the sending end to send a CSIG message along the transmission path. The acquisition module 702 is used to enable the forwarding device on the transmission path to obtain the bottleneck information of the current jump device according to the CSIG message. The judgment module 703 is used to enable the forwarding device to judge whether the current jump device generates a CSIG Reflection message according to a predetermined condition, and the CSIGReflection message is encapsulated with CSIG congestion information. The generation module 704 is used to enable the forwarding device to generate a CSIG Reflection message when the judgment condition is met, and send the CSIG Reflection message to the sending end.
[0111] Please refer to Figure 8 As shown, an embodiment of the present invention further provides an electronic device 800, which includes at least one processor 801, a memory 802 (for example, a non-volatile memory), a memory 803, and a communication interface 804, and the at least one processor 801, the memory 802, the memory 803, and the communication interface 804 are connected together via a bus 805. The at least one processor 801 is used to call at least one program instruction stored or encoded in the memory 802, so that the at least one processor 801 performs various operations and functions of the CSIG congestion notification method described in various embodiments of this specification.
[0112] In an embodiment of the present specification, the electronic device 800 may include, but is not limited to, a personal computer, a server computer, a workstation, a desktop computer, a laptop computer, a notebook computer, a mobile electronic device, a smart phone, a tablet computer, a cellular phone, a personal digital assistant (PDA), a handheld device, a messaging device, a wearable electronic device, a consumer electronic device, and the like.
[0113] An embodiment of the present invention further provides a computer-readable medium carrying computer-executable instructions. When the computer-executable instructions are executed by a processor, they can be used to implement various operations and functions of the CSIG congestion notification method described in various embodiments of this specification.
[0114] The computer-readable medium in the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.
[0115] In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.
[0116] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0117] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses, systems, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0118] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
[0119] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.
Claims
1. A CSIG congestion notification method, characterized in that: include: The sender sends the CSIG message along the transmission path; The forwarding device on the transmission path obtains the bottleneck information of the local jump device according to the CSIG message; The forwarding device determines, according to a predetermined condition, whether the current jump device generates a CSIG Reflection message, wherein the CSIGReflection message is encapsulated with CSIG congestion information; If so, the forwarding device generates a CSIG Reflection message, and sends the CSIG Reflection message to the sending end.
2. The CSIG congestion notification method according to claim 1, characterized in that: The method comprises: The forwarding device determines whether the congestion level of the current jump device is higher than the bottleneck information carried by the CSIG message, where the congestion information includes the available bandwidth of the egress, the occupancy rate of the available bandwidth of the egress, and the delay; If so, the forwarding device generates a CSIG Reflection message, and sends the CSIG Reflection message to the sending end.
3. The CSIG congestion notification method according to claim 1, characterized in that: The method comprises: The forwarding device determines whether the next-hop device supports CSIG; If not, the forwarding device generates a CSIG Reflection message, and sends the CSIG Reflection message to the sending end.
4. The CSIG congestion notification method according to claim 1, characterized in that: The method comprises: The forwarding device determines whether the CSIG message is discarded in the local jump device; If so, the forwarding device generates a CSIG Reflection message, and sends the CSIG Reflection message to the sending end.
5. The CSIG congestion notification method according to claim 1, characterized in that: The method further comprises: When the forwarding device receives a CSIG message for the first time, it parses the source IP, destination IP and bottleneck information of the CSIG message, and stores these fields as key values in the CSIG flow table.
6. The CSIG congestion notification method according to claim 5, characterized in that: The method further comprises: The forwarding device compares the bottleneck information of the current jump device with the bottleneck information in the CSIG flow table. If the congestion level of the current jump device is higher than the bottleneck information in the CSIG flow table, the bottleneck information in the CSIG flow table is updated.
7. The CSIG congestion notification method according to claim 5, characterized in that: The method further comprises: The forwarding device periodically scans the CSIG flow table, and if a flow table entry does not receive a corresponding CSIG message within multiple cycles, the flow table entry is deleted.
8. A CSIG congestion notification device, characterized in that: include: A sending module, used to enable the sending end to send the CSIG message along the transmission path; An acquisition module, configured to enable a forwarding device on the transmission path to acquire bottleneck information of a local jump device according to the CSIG message; A judgment module, used to enable the forwarding device to judge whether the current jump device generates a CSIGReflection message according to a predetermined condition, wherein the CSIGReflection message is encapsulated with CSIG congestion information; The generating module is used to enable the forwarding device to generate a CSIG Reflection message when a judgment condition is met, and send the CSIG Reflection message to the sending end.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the CSIG congestion notification method according to any one of claims 1 to 7 is implemented.
10. A computer-readable medium, characterized in that The computer-readable medium carries computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the CSIG congestion notification method according to any one of claims 1 to 7.
Citation Information
Cited By
Message transmission methods, apparatus, electronic device, medium and program product
WO2026118691A1
