A lossless traffic congestion adaptive method, system and network device

Abstract

The present application discloses a lossless flow congestion adaptive method, system and network device, belonging to the field of communication technology. The lossless flow congestion adaptive method comprises: analyzing the transmitted network traffic and obtaining traffic analysis parameters, wherein the traffic analysis parameters include the traffic incast value, the delay-sensitive traffic ratio, and the throughput-sensitive traffic ratio; dynamically adjusting the threshold value of the display congestion notification ECN according to the obtained traffic analysis parameters; and the source end automatically adjusts the traffic sending window according to the ECN threshold value. It is applied in the process of network traffic control to achieve the purpose of dynamically adjusting the ECN threshold value.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communications, and in particular to a lossless traffic congestion adaptation method, system, and network device. Background Art

[0002] When congestion occurs in a multi-level network, the current common solution is to deploy both the Priority-based Flow Control (PFC) and the Explicit Congestion Notification (ECN) functions. PFC is a queue congestion-based function. When the congestion length of a queue reaches its threshold, PFC is triggered and a backpressure packet is sent to the upstream device until the source device receives the backpressure packet, at which point the sending rate of the corresponding priority traffic is reduced. ECN is when congestion occurs in the lossless queue of a network device, that is, when the queue's used buffer exceeds the ECN threshold, the network device adds an ECN tag to the forwarded message. After receiving the message with the ECN congestion tag, the receiving end sends a congestion notification message to the source end. The source end reduces its sending rate after receiving the message.

[0003] However, the traditional ECN function manually sets static thresholds. This solution easily triggers the PFC threshold when the network is congested, causing the network congestion to worsen. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to provide a lossless traffic congestion adaptive method, system and network device, which can dynamically adjust the ECN threshold value and avoid the problem of triggering the PFC function due to the long back pressure duration of the ECN response message.

[0005] To achieve the above objectives, an embodiment of the present application provides a lossless traffic congestion adaptation method, comprising: performing statistical analysis on transmitted network traffic to obtain traffic analysis parameters, wherein the traffic analysis parameters include a traffic incast value, a delay-sensitive traffic ratio, and a throughput-sensitive traffic ratio; dynamically adjusting a threshold value for displaying an ECN (Emergency Congestion Notification) based on the obtained traffic analysis parameters; and automatically adjusting a traffic sending window at a source end based on the ECN threshold value.

[0006] To achieve the above objectives, the present invention further provides a lossless traffic congestion adaptive system, including:

[0007] A traffic analysis module is used to analyze the transmitted network traffic, obtain traffic analysis parameters, and send the traffic analysis parameters to the traffic queue management module, wherein the traffic analysis parameters include traffic incast value, delay-sensitive traffic ratio, and throughput-sensitive traffic ratio;

[0008] a traffic queue management module, configured to dynamically adjust a threshold value for displaying a congestion notification (ECN) based on the traffic analysis parameters obtained by the traffic analysis module;

[0009] The traffic window adjustment module is used to automatically adjust the traffic sending window according to the ECN threshold value dynamically adjusted by the traffic queue management module.

[0010] To achieve the above objectives, an embodiment of the present application further provides a lossless traffic congestion adaptive network device, comprising:

[0011] at least one processor; and,

[0012] a memory communicatively connected to the at least one processor; wherein,

[0013] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned lossless traffic congestion adaptation method.

[0014] The present application proposes a lossless traffic congestion adaptation method, system, and network device. By analyzing the transmitted network traffic and obtaining traffic analysis parameters, the ECN threshold is dynamically adjusted based on the traffic analysis parameters, including the traffic incast value, the delay-sensitive traffic ratio, and the throughput-sensitive traffic ratio. This avoids the problem of triggering the PFC function due to the long backpressure duration of the ECN response message. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flowchart of the lossless traffic congestion adaptation method provided by the first embodiment of the present application;

[0016] Figure 2 is a flow chart of a lossless traffic congestion adaptation method provided by the second embodiment of the present application;

[0017] Figure 3 is a flowchart of a lossless traffic congestion adaptation method provided by a third embodiment of the present application;

[0018] Figure 4 is a flowchart of a lossless traffic congestion adaptation method provided by a fourth embodiment of the present application;

[0019] Figure 5 is a flowchart of a lossless traffic congestion adaptation method provided by a fifth embodiment of the present application;

[0020] Figure 6 is a flowchart of a lossless traffic congestion adaptation method provided by a sixth embodiment of the present application;

[0021] Figure 7 2 is a schematic structural diagram of a lossless flow congestion adaptive system provided by a seventh embodiment of the present application;

[0022] Figure 8 It is a structural diagram of the network device provided in the eighth embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, each embodiment of the present application will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present application, many technical details are proposed to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present application. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0024] The first embodiment of the present invention relates to a lossless flow congestion adaptive method, the specific process is as follows Figure 1 Shown, including:

[0025] Step 101: Analyze the transmitted network traffic and obtain traffic analysis parameters, wherein the traffic analysis parameters include traffic incast value, delay-sensitive traffic ratio, and throughput-sensitive traffic ratio;

[0026] In this embodiment, step 101 can analyze the transmitted network traffic through an Ethernet device. For example, a switch can perform statistical analysis on the network packets entering and leaving each port to obtain traffic analysis parameters. The traffic analysis parameters include the incast value of traffic, the ratio of delay-sensitive traffic, the ratio of throughput-sensitive traffic, etc. Of course, the above is only a specific example. In actual use, the traffic analysis parameters may also include other parameters, which are not detailed here.

[0027] It should be noted that this embodiment does not limit the method of analyzing network traffic. In actual use, any existing traffic analysis method can be used, and they are not described in detail here.

[0028] Step 102: Dynamically adjust the threshold value for displaying congestion notification (ECN) based on the acquired traffic analysis parameters.

[0029] Step 103: The source end automatically adjusts the traffic sending window according to the ECN threshold.

[0030] In this embodiment, step 103 includes sending an ACK response message carrying an ECE flag to the source end according to the ECN threshold value; after the source end receives the ACK response message, checking the ECE flag; if the ECE flag is marked as a congested state, the source end starts to automatically adjust the traffic sending window.

[0031] Compared with the prior art, the embodiments of the present invention analyze the transmitted network traffic to obtain traffic analysis parameters. Based on the traffic analysis parameters, including the traffic incast value, the delay-sensitive traffic ratio, and the throughput-sensitive traffic ratio, the ECN threshold is dynamically adjusted, thereby avoiding the problem of triggering the PFC function due to the long back pressure duration of the ECN response message.

[0032] The second embodiment of the present invention relates to a lossless flow congestion adaptive method, which is basically the same as the lossless flow congestion adaptive method provided by the first embodiment of the present invention, except that: Figure 2 As shown, step 101 includes:

[0033] Step 201: Statistically analyze the transmitted network traffic to obtain the traffic incast value.

[0034] In this embodiment, the traffic incast value changes dynamically in real time. The larger the traffic incast value is, the more serious the network congestion is.

[0035] Step 202: Differentiate the types of network traffic according to the attributes of the data packets, and obtain the ratio of delay-sensitive traffic and the ratio of throughput-sensitive traffic.

[0036] In this embodiment, the method for obtaining the ratio of delay-sensitive traffic and throughput-sensitive traffic is not limited. In actual use, any existing method for obtaining the ratio of traffic types can be used, and this is not described in detail here. In addition, the attributes of the data packet may include the length of the data packet, the type of the data packet, etc., which are not described in detail here.

[0037] Compared with the prior art, the embodiments of the present invention, while achieving the beneficial effects of the first embodiment, differentiate network traffic types based on data packet attributes, obtain the ratio of delay-sensitive traffic and throughput-sensitive traffic, and ensure that the network's requirements for delay and throughput are fully considered during the subsequent dynamic adjustment of the ECN threshold.

[0038] The third embodiment of the present invention relates to a lossless flow congestion adaptive method, which is basically the same as the lossless flow congestion adaptive method provided by the first embodiment of the present invention, except that: Figure 3 As shown, step 102 includes:

[0039] Step 301: Determine whether the flow incast value is greater than a preset incast value.

[0040] In this embodiment, the network congestion level is determined by comparing the traffic incast value with a preset incast value, and an appropriate control method is selected. If the traffic incast value exceeds the preset incast value, step 302 is executed to lower the ECN threshold. If the traffic incast value is less than the preset incast value, step 303 is executed to raise the ECN threshold.

[0041] Step 302: Reduce the ECN threshold value using the formula Th=E-incast*a.

[0042] Step 303: Increase the ECN threshold value using the formula Th=E+incast*a.

[0043] In this embodiment, Th represents the dynamic ECN threshold, E represents the default ECN threshold set in the initial state, and a represents the coefficient of influence of the incast value on the ECN threshold. It should be noted that the preset incast value, the default ECN threshold set in the initial state, and the coefficient of influence of the incast value on the ECN threshold can be flexibly set based on user needs, network environment, and actual application scenarios.

[0044] Compared with the prior art, the embodiments of the present invention, while achieving the beneficial effects of the first embodiment, compare the current traffic incast value with the preset incast value to determine the network congestion situation, and select different methods to adjust the ECN threshold according to different situations, thereby ensuring dynamic adjustment of the ECN threshold.

[0045] The fourth embodiment of the present invention relates to a lossless flow congestion adaptive method, which is basically the same as the lossless flow congestion adaptive method provided by the first embodiment of the present invention, except that: Figure 4 As shown, step 102 includes:

[0046] Step 401 : Determine whether the delay-sensitive traffic ratio or the throughput-sensitive traffic ratio is greater than a preset condition.

[0047] In this embodiment, when the delay-sensitive traffic ratio is greater than a preset condition, step 402 is executed to lower the ECN threshold; when the throughput-sensitive traffic ratio is greater than a preset condition, step 403 is executed to increase the ECN threshold.

[0048] Step 402, using the formula Th=ER S*b, by increasing the delay-sensitive traffic ratio's impact coefficient b on the ECN threshold, the ECN threshold value is reduced.

[0049] Step 403: Use the formula Th=E+R H *c, by increasing the coefficient c of the impact of the throughput-sensitive traffic ratio on the ECN threshold, the ECN threshold value is increased.

[0050] In this embodiment, b represents the influence coefficient of the delay-sensitive traffic ratio on the ECN threshold, c represents the influence coefficient of the throughput-sensitive traffic ratio on the ECN threshold, and R S Indicates the proportion of delay-sensitive traffic, R H Indicates the proportion of throughput-sensitive traffic.

[0051] Specifically, when increasing the coefficient b of the delay-sensitive traffic ratio's impact on the ECN threshold and the coefficient c of the throughput-sensitive traffic ratio's impact on the ECN threshold, the ECN threshold can be dynamically adjusted by increasing b and c based on the traffic type ratio. Of course, the above is merely an example; in actual use, any existing adjustment method or standard can be used, and this is not detailed here.

[0052] Compared to the prior art, the embodiments of the present invention, while achieving the beneficial effects of the first embodiment, take into account both delay-sensitive traffic and throughput-sensitive traffic. The coefficient b for the impact of the delay-sensitive traffic ratio on the ECN threshold and the coefficient c for the impact of the throughput-sensitive traffic ratio on the ECN threshold are set according to different user needs and actual application environments. This ensures the requirement for dynamic adjustment of the ECN threshold while also taking into account the latency and throughput requirements of different traffic types.

[0053] The fifth embodiment of the present invention relates to a lossless flow congestion adaptive method, which is basically the same as the lossless flow congestion adaptive method provided by the first embodiment of the present invention, except that: Figure 5 As shown, step 102 includes:

[0054] Step 501: Whether the flow incast value is greater than the preset incast value.

[0055] In this embodiment, the network congestion level is determined by comparing the traffic incast value with a preset incast value, and an appropriate control method is selected. If the traffic incast value is greater than the preset incast value, step 502 is executed to lower the ECN threshold. If the traffic incast value is less than the preset incast value, step 503 is executed to raise the ECN threshold.

[0056] Step 502: Th = E-incast*aR S *b+RH *cLower the ECN threshold.

[0057] Step 503, by the formula Th=E+incast*aR S *b+R H *c Increase ECN threshold;

[0058] Compared to the prior art, this embodiment of the present invention, while achieving the beneficial effects of the first embodiment, determines network congestion by comparing the current traffic incast value with a preset incast value, while also taking into account latency and throughput issues. When the incast value is too high, a low ECN threshold is set to meet the low latency requirements of the traffic in the queue. When the incast value is too low, the ECN threshold is increased to ensure high throughput requirements. Furthermore, the method for dynamically adjusting the ECN threshold has been optimized.

[0059] The sixth embodiment of the present invention relates to a lossless flow congestion adaptive method, which is basically the same as the lossless flow congestion adaptive method provided by the first embodiment of the present invention, except that: Figure 6 As shown, step 102 includes:

[0060] Step 601 : Determine whether the delay-sensitive traffic ratio or the throughput-sensitive traffic ratio is greater than a preset condition.

[0061] In this embodiment, when the delay-sensitive traffic ratio is greater than a preset condition, step 602 is executed to lower the ECN threshold; when the throughput-sensitive traffic ratio is greater than a preset condition, step 603 is executed to increase the ECN threshold.

[0062] Step 602: Use the formula Th = E-incast*aR S *b+R H *c, by increasing the coefficient b of the delay-sensitive traffic ratio on the ECN threshold and decreasing the coefficient c of the throughput-sensitive traffic ratio on the ECN threshold, the ECN threshold value is lowered.

[0063] Step 603, using the formula Th=E+incast*aR S *b+R H *c, by reducing the coefficient b of the impact of the delay-sensitive traffic ratio on the ECN threshold and increasing the coefficient c of the impact of the throughput-sensitive traffic ratio on the ECN threshold, the ECN threshold value is increased.

[0064] Compared with the prior art, the embodiments of the present invention, while achieving the beneficial effects of the first embodiment, analyze the traffic types in the network and adjust the ECN threshold according to the proportion of traffic types. This reduces the impact of congestion on traffic forwarding while meeting the latency and throughput requirements of different traffic types.

[0065] The seventh embodiment of the present invention relates to a lossless flow congestion adaptive system, such as Figure 7 Shown, including:

[0066] Traffic analysis module 701 is used to analyze the transmitted network traffic, obtain traffic analysis parameters, and send the traffic analysis parameters to traffic queue management module 702, wherein the traffic analysis parameters include traffic incast value, delay-sensitive traffic ratio, and throughput-sensitive traffic ratio;

[0067] The traffic queue management module 702 is configured to dynamically adjust a threshold value for displaying a congestion notification (ECN) based on the traffic analysis parameters obtained by the traffic analysis module 701;

[0068] The traffic window adjustment module 703 is configured to automatically adjust the traffic sending window according to the ECN threshold value dynamically adjusted by the traffic queue management module 702 .

[0069] An eighth embodiment of the present invention relates to a network device, such as Figure 8 Shown, including:

[0070] at least one processor 801; and,

[0071] A memory 802 in communication with the at least one processor 801; wherein,

[0072] The memory 802 stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor 801 to enable the at least one processor 801 to execute the lossless flow congestion adaptation method described in the first to sixth embodiments of the present invention.

[0073] The memory and processor are connected using a bus, which can include any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and are therefore not described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over a wireless medium via an antenna. Furthermore, the antenna receives data and transmits it to the processor.

[0074] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0075] Those skilled in the art will appreciate that the above-mentioned embodiments are specific examples for implementing the present invention, and that in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A lossless traffic congestion adaptive method, characterized in that: include: Performing statistical analysis on the transmitted network traffic to obtain traffic analysis parameters, wherein the traffic analysis parameters include traffic incast value, delay-sensitive traffic ratio, and throughput-sensitive traffic ratio; Dynamically adjust a threshold value of displaying a congestion notification (ECN) according to the acquired traffic analysis parameters; According to the ECN threshold value, the source end automatically adjusts the traffic sending window; The step of dynamically adjusting the threshold value of displaying the congestion notification ECN according to the acquired traffic analysis parameter includes at least one of the following: When the traffic incast value is greater than a preset incast value, the ECN threshold value is negatively correlated with the traffic incast value; When the traffic incast value is less than the preset incast value, the ECN threshold value is positively correlated with the traffic incast value; When the delay-sensitive traffic ratio is greater than a preset condition, the ECN threshold is negatively correlated with the delay-sensitive traffic ratio; When the throughput-sensitive traffic ratio is greater than a preset condition, the ECN threshold value is positively correlated with the throughput-sensitive traffic ratio.

2. The lossless flow congestion adaptive method according to claim 1, characterized in that: Analyze the transmitted network traffic and obtain traffic analysis parameters, including: Perform statistical analysis on the transmitted network traffic and obtain the traffic incast value; The types of the network traffic are distinguished according to the attributes of the data packets, and the delay-sensitive traffic ratio and the throughput-sensitive traffic ratio are obtained.

3. The lossless flow congestion adaptive method according to claim 1, characterized in that: The dynamically adjusting the threshold value of the display congestion notification ECN according to the acquired traffic analysis parameter includes: Comparing the flow incast value with a preset incast value; If the traffic incast value is greater than the preset incast value, the ECN threshold value is reduced by the formula T=E-incast*a; If the traffic incast value is less than the preset incast value, the ECN threshold is increased by the formula T=E+incast*a; Wherein, Th represents the dynamic ECN threshold value, E represents the default ECN threshold set in the initial state, and a represents the influence coefficient of the incast value on the ECN threshold.

4. The lossless flow congestion adaptive method according to claim 1, characterized in that: The dynamically adjusting the threshold value of the display congestion notification ECN according to the acquired traffic analysis parameter includes: Comparing the delay-sensitive traffic ratio, the throughput-sensitive traffic ratio and a preset condition; If the delay-sensitive traffic ratio is greater than the preset condition, the formula Th = ER is used. S *b, by increasing the influence coefficient b of the delay-sensitive traffic ratio on the ECN threshold, the ECN threshold value is reduced; If the throughput sensitive traffic ratio is greater than the preset condition, the formula Th=E+R is used H *c, by increasing the impact coefficient c of the throughput sensitive traffic ratio on the ECN threshold, the ECN threshold value is increased; Where Th represents the dynamic ECN threshold value, E represents the default ECN threshold set in the initial state, b represents the influence coefficient of the delay-sensitive traffic ratio on the ECN threshold, c represents the influence coefficient of the throughput-sensitive traffic ratio on the ECN threshold, and R S Indicates the proportion of delay-sensitive traffic, R H Indicates the proportion of throughput-sensitive traffic.

5. The lossless flow congestion adaptive method according to claim 1, characterized in that: The dynamically adjusting the threshold value of the display congestion notification ECN according to the acquired traffic analysis parameter includes: Comparing the flow incast value with a preset incast value; If the flow incast value is greater than the preset incast value, then the formula T = E-incast*aR s *b+R H *c Reduce the ECN threshold; If the flow incast value is less than the preset incast value, then the formula T = E + incast * aR s *b+R H *cIncrease ECN threshold; Where Th represents the dynamic ECN threshold value, E represents the default ECN threshold set in the initial state, a represents the influence coefficient of the incast value on the ECN threshold, b represents the influence coefficient of the delay-sensitive traffic ratio on the ECN threshold, c represents the influence coefficient of the throughput-sensitive traffic ratio on the ECN threshold, and R S Indicates the proportion of delay-sensitive traffic, R H Indicates the proportion of throughput-sensitive traffic.

6. The lossless flow congestion adaptive method according to claim 1, characterized in that: The dynamically adjusting the threshold value of the display congestion notification ECN according to the acquired traffic analysis parameter includes: If the delay-sensitive traffic ratio is greater than the preset condition, the formula T = E-incast*aR is used. s *b+R H *c, by increasing the coefficient b of the delay-sensitive traffic ratio on the ECN threshold and reducing the coefficient c of the throughput-sensitive traffic ratio on the ECN threshold, the ECN threshold value is reduced; If the throughput sensitive traffic ratio is greater than the preset condition, the formula T = E + incast * aR is used s *b+R H *c, the ECN threshold value is improved by reducing the coefficient b of the delay-sensitive traffic ratio on the ECN threshold and increasing the coefficient c of the throughput-sensitive traffic ratio on the ECN threshold; Where Th represents the dynamic ECN threshold value, E represents the default ECN threshold set in the initial state, a represents the influence coefficient of the incast value on the ECN threshold, b represents the influence coefficient of the delay-sensitive traffic ratio on the ECN threshold, c represents the influence coefficient of the throughput-sensitive traffic ratio on the ECN threshold, and R S Indicates the proportion of delay-sensitive traffic, R H Indicates the proportion of throughput-sensitive traffic.

7. The lossless flow congestion adaptive method according to any one of claims 3 to 6, characterized in that: The influence coefficient of incast value on ECN threshold, the influence coefficient of delay-sensitive traffic ratio on ECN threshold, and the influence coefficient of throughput-sensitive traffic ratio on ECN threshold are all set according to user requirements.

8. The lossless flow congestion adaptive method according to claim 1, characterized in that: The source end automatically adjusts the traffic sending window according to the ECN threshold value, including: According to the ECN threshold value, an ACK response message carrying the ECE flag is sent to the source end; The source end receives the ACK response message and checks the ECE flag bit; If the ECE flag is marked as a congestion state, the source end starts to automatically adjust the traffic sending window.

9. A lossless traffic congestion adaptive system, characterized in that: include: A traffic analysis module, used to analyze the transmitted network traffic, obtain traffic analysis parameters and send the traffic analysis parameters to the traffic queue management module, wherein the traffic analysis parameters include traffic incast value, delay-sensitive traffic ratio, and throughput-sensitive traffic ratio; A traffic queue management module, configured to dynamically adjust a threshold value of a display congestion notification ECN according to the traffic analysis parameters obtained by the traffic analysis module; A flow window adjustment module is used to automatically adjust the flow sending window according to the ECN threshold value dynamically adjusted by the flow queue management module; Wherein, when the traffic incast value is greater than a preset incast value, the ECN threshold value is negatively correlated with the traffic incast value; When the traffic incast value is less than the preset incast value, the ECN threshold value is positively correlated with the traffic incast value; When the delay-sensitive traffic ratio is greater than a preset condition, the ECN threshold is negatively correlated with the delay-sensitive traffic ratio; When the throughput-sensitive traffic ratio is greater than a preset condition, the ECN threshold is positively correlated with the delay-sensitive traffic ratio.

10. A network device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the lossless traffic congestion adaptation method as described in any one of claims 1 to 8.

Images