Method for limiting fragment message traffic rate, DPU fragment message transmission method and electronic device
The method improves fragment message traffic rate limiting by pre-staining and evaluating fragments with a non-colorblind token bucket algorithm, addressing inefficiencies in DPU data processing and ensuring reliable data reconstruction.
Patent Information
- Application Number
- JP2024532904
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2022-10-28
- Filing Date
- 2023-10-30
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2043-10-30
AI Technical Summary
Current traffic rate limiting methods for fragmented messages in data processing units (DPUs) result in the discarding of essential fragments, leading to inefficient use of network bandwidth and reduced data processing reliability due to the inability to reconstruct raw data messages at the receiving terminal.
A method involving a pre-staining process using a message fragment staining table to assign discard or retain colors to fragments, followed by a non-colorblind token bucket algorithm for traffic evaluation, ensuring accurate discard or forwarding of fragments based on their color indicators.
This approach enhances traffic evaluation efficiency, reduces resource consumption, and improves the reliability and effectiveness of fragment message transmission by minimizing the transfer of invalid fragments, thereby optimizing network bandwidth usage and ensuring complete data reconstruction at the receiving terminal.
Smart Images

Figure 2024542768000001_ABST
Abstract
Description
[Technical field]
[0001] The present application relates to the field of data processing technology, and in particular to a method for limiting fragment message traffic rate, a method and apparatus for forwarding DPU fragment message. [Background technology]
[0002] In the network transmission process, when the IP layer protocol transmits data with a relatively large number of bytes, due to the limitation of the maximum transmission unit (MTU) of the data link layer, the transmitting end needs to perform fragmentation processing on the upper layer message, divide the message into multiple fragments whose length is less than the maximum message length of the link, and then use a data processing device such as a DPU to perform traffic rate limiting on these fragments, and transmit the fragments that meet the rate limiting rules to the receiving terminal. After the receiving terminal receives the fragments, it reconstructs each fragment belonging to the same message and restores it to a raw data message.
[0003] Currently, traffic rate limiting for fragment messages that a data processing device such as a DPU needs to transfer is mainly performed by a token bucket algorithm. When each fragment message arrives, the token bucket performs a traffic evaluation on it. If the committed traffic or maximum allowable burst traffic range is not exceeded, a color mark corresponding to the fragment is attached and then the message is transferred. If not, a subsequent module discards the message fragment that exceeds the promised rate.
[0004] However, in the process of transmitting a fragmented message, if a fragment in the fragment sequence of an IP message is discarded in the transmission process of a data processing device, the receiving terminal cannot reconstruct the fragment and cannot restore the raw data of the IP message. Furthermore, a situation will occur in which invalid fragments are transmitted in the transmission process performed by the data processing device, which will waste a large amount of network bandwidth resources and affect the data processing efficiency and application reliability of data processing devices such as DPUs. Summary of the Invention [Problem to be solved by the invention]
[0005] In view of this, embodiments of the present application provide a method for limiting fragment message traffic rate to overcome or ameliorate one or more deficiencies existing in the prior art. [Means for solving the problem]
[0006] A first aspect of the present application provides a method for limiting fragment message traffic rate, comprising: performing a pre-dyeing process on the target fragment message according to a preset message fragment dyeing table, thereby providing the target fragment message with a color indicator including a discard color and a keep color; performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on the color signature of the double dyed target fragment message; If the current color indication of the target fragment message is the discard color, performing a rate limiting process on the target fragment message, and if the current color indication of the target fragment message is the retain color, performing a forwarding process on the target fragment message.
[0007] In some embodiments of the present application, the above-mentioned performing the preliminary staining process for each fragment message based on the preset message fragment staining table includes: creating a message fragment staining table for storing a correspondence between the color indicators and the indicators of the raw data messages to which the fragment messages belong; Here, the raw data message identifier includes a source IP, a destination IP, a protocol type and a unique identity identifier.
[0008] In some embodiments of the present application, the above-mentioned pre-dyeing process is performed on the target fragment message based on the preset message fragment dyeing table, so that the target fragment message is provided with a color mark. Obtaining an indication of a raw data message to which a currently received target fragment message belongs; The method includes searching the message fragment dyeing table for whether it contains an indicator of the raw data message to which the target fragment message belongs, and if so, performing a preliminary dyeing process on the target fragment message based on a color indicator corresponding to the indicator of the raw data message.
[0009] In some embodiments of the present application, If an indicator including the raw data message to which the target fragment message belongs is not found in the message fragment staining table, the target fragment message is pre-stained with the retained color as a default color, and the correspondence between the indicator of the raw data message to which the target fragment message belongs and the color indicator which is the retained color is stored in the message fragment staining table.
[0010] In some embodiments of the present application, performing traffic evaluation on the target fragment message that is pre-dyed based on the token bucket algorithm of the non-color blind mode as described above, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on the color signature of the twice-dyed target fragment message, According to the token bucket algorithm of the non-color blind mode, firstly recognize the color mark of the current target fragment message, and if the color mark of the recognized and obtained target fragment message is the discard color, confirm that the color mark of the double-stained target fragment message does not change; When the color indicator of the target fragment message is the retained color, determine whether the number of tokens in the corresponding token bucket meets the message length of the target fragment message, and if so, confirm that the double-dyed color indicator of the target fragment message does not change, and reduce the number of tokens in the token bucket that matches the message length of the target fragment message; If it is determined that the number of tokens in the token bucket does not meet the message length of the target fragment message, changing the double-staining color indicator of the target fragment message to the discard color, and updating the color indicator of the raw data message to which the target fragment message belongs in the message fragment staining table.
[0011] In some embodiments of the present application, before performing the forwarding process on the target fragment message, determining whether the traffic-evaluated target fragment messages are the last fragment messages of the respective raw data messages; If so, delete the record of the raw data message to which the target fragment message belongs in the message fragment staining table.
[0012] A second aspect of the present application provides a DPU fragment message transfer method, which includes: receiving locally at the DPU a target fragment message to be forwarded; The method includes performing processing on the received target fragment message based on the fragment message traffic rate limiting method according to the first aspect, and performing forwarding processing on the target fragment message if the current color indication of the target fragment message is the retained color.
[0013] In some embodiments of the present application, the message fragment staining table is used to store the correspondence between the keyword information of the raw data message to which the fragment message belongs and the color mark; Here, the keyword information of the fragment message includes a unique identity identifier and a local stream label that is preset based on the identifier of the raw data message.
[0014] Another aspect of the present application provides a fragment message traffic rate limiting device, comprising: a pre-dyeing module for performing a pre-dyeing process on the target fragment message according to a preset message fragment dyeing table, so that the target fragment message is provided with a color indicator including a discard color and a keep color; a traffic evaluation module for performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on a color signature of the double dyed target fragment message; The present invention also includes a fragment rate limiting module for performing rate limiting processing on the target fragment message if the current color indication of the target fragment message is the discard color, and for performing forwarding processing on the target fragment message if the current color indication of the target fragment message is the retain color.
[0015] Another aspect of the present application provides a DPU fragment message forwarding device, which includes: a fragment receiving module for locally receiving a target fragment message to be forwarded in the DPU; a traffic rate limiting module for performing processing on the received target fragment message based on the fragment message traffic rate limiting method according to the first aspect; A fragment forwarding module for performing a forwarding process on the target fragment message if the current color signature of the target fragment message is the retained color.
[0016] Another aspect of the present application provides an electronic device including a memory, a processor, and a computer program stored in the memory and operable to run on the processor, the electronic device being used to realize the fragment message traffic rate limiting method according to the first aspect or the DPU fragment message forwarding method according to the second aspect when the processor executes the computer program.
[0017] Another aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is used to realize the fragment message traffic rate limiting method according to the first aspect or the DPU fragment message forwarding method according to the second aspect. Effect of the Invention
[0018] The fragment message traffic rate limiting method of the present application first performs a pre-dyeing process on the received fragment message before adopting a token bucket algorithm to perform traffic evaluation of the fragment message, so that the traffic evaluation process can perform a discard process on the fragment message pre-dyed to a discard color, and there is no need to perform token evaluation in the token bucket for each fragment message, which can effectively reduce the resource consumption of the traffic evaluation process and improve the traffic evaluation efficiency; performs traffic evaluation on the fragment message pre-dyed by a token bucket algorithm based on a color blind mode, so that a double-dyeing process is performed on the fragment message, improving the traffic evaluation efficiency, and then discards the fragment message pre-dyed to a retained dye but whose size exceeds the excess area in the token bucket, thereby ensuring the effectiveness and reliability of the fragment message traffic rate limiting; performs a rate limiting process on the fragment message whose color mark of the pre-dyed and double-dyed color mark is a discard color before transmission, thereby effectively avoiding the transmission of invalid fragment messages and improving the transmission efficiency of the fragment message, and effectively reducing the waste of network bandwidth resources, so that the receiving terminal equipment can restore the raw data message based on each received fragment message, and further improving the effectiveness and reliability of the fragment message transmission.
[0019] Additional advantages, objects and features of the present application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art after examination of the following or may be learned by practice of the present application. The objectives and other advantages of the present application may be realized and obtained by the structure particularly pointed out in the description and the accompanying drawings.
[0020] It will be appreciated by those skilled in the art that the objects and advantages that may be achieved by the present application are not limited to those specifically set forth above, and these and other objects that may be achieved by the present application will be more clearly understood from the following detailed description. [Brief description of the drawings]
[0021] The accompanying drawings described herein are used to provide further understanding of the present application, constitute a part of the present application, and do not constitute limitations thereon. The parts in the accompanying drawings are not drawn to scale, but are drawn only to illustrate the principles of the present application. To facilitate the illustration and explanation of some parts of the present specification, the corresponding parts in the accompanying drawings may be enlarged, i.e., made larger than other parts in an exemplary device actually manufactured according to the present specification. The accompanying drawings are as follows:
[0022] [Figure 1] 1 is a general flowchart of a method for limiting fragmented message traffic rate in one embodiment of the present application; [Diagram 2] 2 is a specific flowchart of a method for limiting the rate of fragmented message traffic in one embodiment of the present application; [Diagram 3] 4 is a flowchart of a DPU fragment message forwarding method in another embodiment of the present application; [Figure 4] FIG. 2 is a structural schematic diagram of a fragment message traffic rate limiting device in another embodiment of the present application; [Diagram 5] FIG. 2 is a structural schematic diagram of a DPU fragment message forwarding device in another embodiment of the present application; [Figure 6]4 is an exemplary flowchart of a method for a DPU to perform traffic rate limiting on a fragment message according to an application example of the present application; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0023] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be described in more detail in the following in combination with the embodiments and accompanying drawings, where the exemplary embodiments and the description thereof are for the purpose of illustrating the present application, and do not constitute limitations to the present application.
[0024] It should be further explained here that in order to avoid obscuring the present application with unnecessary details, only structures and / or process steps closely related to aspects according to the present application are shown in the accompanying drawings, and other details not related to the present application are omitted.
[0025] It should be emphasized that the term "comprising" as used herein denotes the presence of a feature, element, step or component, but does not exclude the presence or addition of one or more other features, elements, steps or components.
[0026] Unless otherwise specified, the term "connection" in this specification can mean not only a direct connection but also an indirect connection where an intermediate object is present.
[0027] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the embodiments of the present application will be described in detail with reference to the accompanying drawings, in which the same reference numerals represent the same or similar components or steps.
[0028] In response to the problem that invalid fragments are transmitted in the transmission process performed by existing data processing devices, which wastes a large amount of network bandwidth resources and affects the data processing efficiency and application reliability of data processing devices such as DPUs, the present application first considers that when a fragment in a fragment sequence is discarded, the subsequent fragments do not need to undergo traffic evaluation and transmission; otherwise, when these subsequent fragments pass through the token bucket, they will consume the remaining tokens in the token bucket, preempting the chances of other messages to obtain tokens, which may result in a waste of a large amount of network bandwidth resources.
[0029] When the inventors researched fragment message rate limiting, they found that the above-mentioned defects in the prior art exist because the token bucket only has the ability to perform traffic evaluation, but does not have the ability to identify whether the subsequent fragment message does not need to be forwarded. The inventors found that, when optimizing the fragment message traffic evaluation result, by introducing a message fragment dyeing table, before the fragment message passes through the token bucket, each fragment is first pre-dyed, and based on the pre-dyed result, the non-color-blind mode token bucket algorithm is further used to perform traffic evaluation, and it is possible to accurately recognize which subsequent fragment messages should be dyed red, and discard them in subsequent processing.
[0030] If a method of sending the fragment stream to the matching sub-token bucket for special processing is adopted, it would require processing each fragment of the same message simultaneously and splitting the bucket, but this method is unlikely to be used. First, each fragment processing is generally serial unless there is an extra buffer to cache the fragments. In addition, the size of the original message for each stream is often not the same, and the number of slices is not the same. Calculating by splitting the buckets would consume a large amount of computing resources and does not match the actual message processing scenario.
[0031] The fragment message traffic rate limiting method according to the present application meets the actual message processing scene, and conforms to the same stream serial, or at least the same message fragment serial (and is not compulsorily required), does not need to cache messages, and does not require consistency in the number of message fragments.
[0032] The present invention will be described in detail with reference to the following examples.
[0033] Based on this, an embodiment of the present application provides a fragment message traffic rate limiting method that can be performed by a fragment message traffic rate limiting device, and referring to FIG. 1, the fragment message traffic rate limiting method specifically includes the following content:
[0034] Step 100: A pre-dyeing process is performed on the target fragment message based on a preset message fragment dyeing table, so that the target fragment message is provided with a mark including a discard color and a keep color.
[0035] In step 100, the data processing device for performing the fragment message traffic rate limiting method can first receive each fragment message from different sending end devices, and determine the currently processed fragment message as a target message, and each of these fragment messages carries at least the identifier of the raw data message to which it belongs, the unique identifier of the fragment message itself, and the unique identifier of the receiving end device, which may include, for example, source IP, destination IP, protocol type, and identity identifier.
[0036] As can be understood, the message fragment dyeing table is a data table for storing the correspondence between the label and color label of each raw data message. For example, when a fragment message A113 is received, the label A100 of the raw data message to which the fragment message A113 belongs can be first obtained, and then the message fragment dyeing table can be searched based on A100 to see whether the raw data message contains a color label corresponding to the raw data message. If it is found based on A100 that the message fragment dyeing table definitely contains a color label belonging to the raw data message, then the fragment message A113 can be pre-dyed with the same color label based on the color label. Specifically, the message fragment dyeing table will be described in detail in the subsequent embodiments.
[0037] In one or more embodiments of the present application, the color indicator may include at least a discard color and a keep color, and the two colors can be selected according to practical application, as long as the color difference between the two colors is ensured. In a preferred embodiment, the discard color is defined as red and the keep color is defined as green, so as to better match the application practice of the token bucket algorithm, and the application reliability and efficiency of the fragment message traffic rate limiting method can be improved.
[0038] In another example, the color indicators may further include more than two color indicators, such as green, yellow and red, for application to a three-color token bucket or the like.
[0039] Step 200: Perform traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color-blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on the color signature of the double-dyed target fragment message.
[0040] In step 200, the token bucket can be considered as a container that holds tokens with a certain preset capacity. The system places tokens into the bucket at a predetermined rate, and when the bucket is full, the excess tokens overflow. The token bucket algorithm in non-color-blind mode (also called Color-Aware or "color-sensitive mode") is as follows: (1) first recognize the color mark of the pre-dyeing of the fragment message, and if it is a discard color, discard the fragment message directly, and confirm that the color mark of the second dyeing of the fragment message does not change; (2) if the color mark of the preliminary dye of the fragment message is a retained color, first determine whether the token bucket contains a number of tokens corresponding to the size of the fragment message based on the size of the fragment message; If there is, extract a number of tokens corresponding to the data size and transmit the data, and confirm that the color mark of the double staining of the fragment message does not change; If not, discard the fragment message, and change the color indicator of the double-stained fragment message from the retain color to the discard color, thereby limiting the message traffic to below the token generation speed, thereby achieving the purpose of limiting the traffic.
[0041] Step 300: if the current color indicator of the target fragment message is the discard color, perform a rate limiting process on the target fragment message; if the current color indicator of the target fragment message is the keep color, perform a forwarding process on the target fragment message.
[0042] In step 300, the method of performing rate limiting processing on fragment messages whose color mark is the discard color may be set according to the needs of actual application. For example, fragment messages whose color mark is the discard color may be directly discarded (deleted), so that all remaining fragment messages are used for subsequent forwarding, or fragment messages whose color mark is the discard color may be stored in a non-forwarding group, and forwarding processing may be performed on other fragment messages that are not stored in the group. In this way, it is easy to search for fragment messages that have not been forwarded in the non-forwarding group so that the user can later perform processing such as lowering the priority and forwarding them.
[0043] As can be seen from the above description, the fragment message traffic rate limiting method according to the embodiment of the present application first performs a pre-dyeing process on the received fragment message before adopting a token bucket algorithm to perform traffic evaluation of the fragment message, so that the traffic evaluation process can perform a discard process on the fragment message pre-dyed to a discard color, and there is no need to perform token evaluation in the token bucket for each fragment message, which can effectively reduce the resource consumption of the traffic evaluation process and improve the traffic evaluation efficiency; performs traffic evaluation on the fragment message pre-dyed by the token bucket algorithm based on the non-color blind mode, so as to perform a double-dyeing process on the fragment message, improving the traffic evaluation efficiency, and then discards the fragment message pre-dyed to a retained dye but whose size exceeds the excess area in the token bucket, thereby ensuring the effectiveness and reliability of the fragment message traffic rate limiting; performs a rate limiting process on the fragment message whose color mark of the pre-dyed and double-dyed color indicators is a discard color before transmission, thereby effectively avoiding the transmission of invalid fragment messages and improving the transmission efficiency of the fragment message, effectively reducing the waste of network bandwidth resources, so that the receiving terminal equipment can restore the raw data message based on each received fragment message, and further improving the effectiveness and reliability of the fragment message transmission.
[0044] In order to further improve the application reliability and effectiveness of the message fragment staining table, in the fragment message traffic rate limiting method according to an embodiment of the present application, referring to FIG. 2, before step 100 in the fragment message traffic rate limiting method, specifically, the following content is further included:
[0045] Step 010: Create a message fragment coloring table for storing the correspondence between the identifier of the raw data message to which the fragment message belongs and a color identifier, where the identifier of the raw data message includes a source IP, a destination IP, a protocol type and a unique identification identifier (i.e., Identification).
[0046] Specifically, the identifier of the raw data message to which the fragment message belongs is taken as an example to be a source IP, a protocol type, a destination IP (a unique identifier of the receiving terminal device) and an Identification. The structure of the message fragment staining table shown in Table 1 is taken as an example.
[0047] Table 1 [Table 1]
[0048] In order to further improve the application reliability and effectiveness of the pre-dyeing process, in the fragment message traffic rate limiting method according to the embodiment of the present application, step 100 in the fragment message traffic rate limiting method specifically includes the following content:
[0049] Step 110: Obtain an identification of the raw data message to which the currently received target fragment message belongs.
[0050] Step 120: searching whether the message fragment dye table contains the identifier of the raw data message to which the target fragment message belongs; if so, performing a pre-dyeing process on the target fragment message according to the color identifier corresponding to the identifier of the raw data message;
[0051] As can be seen, the process of pre-dying fragments by employing a message fragment dyeing table is actually to associate each fragment belonging to one original IP message with the color of the nearest bucket fragment in the fragment sequence, and further optimize the traffic evaluation result of the non-color-blind mode token bucket.
[0052] In order to further improve the application reliability and effectiveness of the pre-dyeing process, in the fragment message traffic rate limiting method according to the embodiment of the present application, step 120 in the fragment message traffic rate limiting method specifically further includes the following content:
[0053] Step 121: If an indicator including the raw data message to which the target fragment message belongs is not found in the message fragment staining table, pre-staining is performed on the target fragment message using the retained color as the default color, and the correspondence between the indicator of the raw data message to which the target fragment message belongs and the color indicator that is the retained color is stored in the message fragment staining table.
[0054] For example, when the first fragment of an IP message is received (if out of order, it may not be the first fragment of a fragment), the message information (the above KEY value) is extracted and the message fragment staining table is queried. If no corresponding entry is found, a new entry record is added and the staining value is set to green by default. Then, when traffic rate limiting is performed on the fragment message, it is evaluated by the non-color blind mode token bucket algorithm (multiple classical token bucket algorithms can be selected). If it is evaluated as green, there is no need to modify the fragment information table, and the message is processed with the action corresponding to green (send by default). If it is evaluated as red, the staining value of the corresponding fragment information table is changed to red, and the message is processed with the action corresponding to red (discard by default). When a subsequent fragment of the IP message is received, the fragment message staining information table is queried in the same way, and the corresponding entry is queried, and if the first fragment has already set the staining value to red before, the message is pre-stained red directly. Then, when traffic rate limiting is performed on the fragment message, it is directly processed red by the non-color blind mode token bucket algorithm (discard by default).
[0055] In order to further improve the application reliability and effectiveness of the traffic evaluation process, in the fragment message traffic rate limiting method according to the embodiment of the present application, step 200 in the fragment message traffic rate limiting method specifically further includes the following content:
[0056] Step 210: Based on the token bucket algorithm in non-color blind mode, first recognize the color signature of the current target fragment message, and if the color signature of the recognized and obtained target fragment message is the discard color, confirm that the double-stained color signature of the target fragment message does not change.
[0057] Step 220: If the color indicator of the target fragment message is the retained color, determine whether the number of tokens in the corresponding token bucket meets the message length of the target fragment message, and if so, confirm that the double-dyed color indicator of the target fragment message does not change, and reduce the number of tokens in the token bucket that matches the message length of the target fragment message.
[0058] Step 230: If it is determined that the number of tokens in the token bucket does not meet the message length of the target fragment message, change the double-staining color indicator of the target fragment message to the discard color, and update the color indicator of the raw data message to which the target fragment message belongs in the message fragment staining table.
[0059] For example, there are the following:
[0060] Taking the single-speed single-bucket token bucket as an example, the content of the non-colorblind mode token bucket algorithm will be briefly described.
[0061] The algorithm follows these rules when performing traffic assessment on arriving messages:
[0062] If a message is dyed green, but the length of the message (say B) does not exceed the number of tokens remaining in the token bucket (say Tc), then the message is dyed green and Tc=Tc-B.
[0063] If the message is dyed green and the length of the message exceeds the number of tokens remaining in the token bucket, Tc, then the message is dyed red and Tc remains the same.
[0064] If the message is stained red and the message remains red, then Tc remains.
[0065] In this method, traffic evaluation is performed on the fragment message using the non-color-blind mode token bucket algorithm based on the preliminary staining result of the fragment message in key point 1 above.
[0066] The token bucket algorithm is used to evaluate traffic for subsequent dyed fragments to avoid excessive bursts of fragments due to excessively large messages (large number of fragments) or malicious fragmentation attacks, which can invalidate the actual rate limiting and ultimately cause uncontrollable bandwidth preemption and packet loss. That is, it avoids sending invalid fragments as much as possible, while also taking into account the rate limiting effect and avoiding malicious bursts (all green transmission).
[0067] Here, for various classical token bucket algorithms, the final output is to dye the message, which is generally three colors: green (sufficient tokens), yellow (intermediate, the judgment differs depending on the algorithm, for example, C bucket has no tokens, E bucket has tokens, or P bucket has tokens and C bucket has no tokens), and red (insufficient tokens). The partially simplified token bucket algorithm (single-speed two-color) only has green and red. For different dyeing results, the message processing method (user configuration) is different, and green pass and red drop are usually adopted by default. Therefore, when describing the token bucket algorithm in the embodiments of this application, it is mostly described by dyeing.
[0068] In order to further improve the application reliability and effectiveness of the traffic evaluation process, in the fragment message traffic rate limiting method according to an embodiment of the present application, referring to FIG. 2, between steps 200 and 300 in the fragment message traffic rate limiting method, specifically, the following content is further included:
[0069] Step 020: Determine whether the traffic-evaluated target fragment message is the last fragment message of the raw data message to which it belongs, and if so, delete the record of the raw data message to which the target fragment message belongs in the message fragment staining table.
[0070] DPU (Data Processing Unit) is a dedicated processor built around data, which adopts the software-defined technology route to support the virtualization of infrastructure layer resources and supports infrastructure layer services such as storage, security, and service quality management. DPU can be a core component of data centers, 5G edge computing, and cloud computing. In the process of DPU forwarding data, when the IP layer protocol transmits data with a relatively large number of bytes, due to the limitation of the maximum transmission unit (MTU) of the data link layer, the transmitting end needs to perform fragmentation processing on the upper layer raw data message, divide the message into multiple fragments whose length is less than or equal to the link maximum message length, and send each obtained fragment to DPU, perform traffic rate limiting on the fragments that need to be forwarded by DPU, and forward the fragments that meet the rate limiting rules to the receiving terminal, and after the receiving terminal receives each fragment, it reconstructs each fragment belonging to the same message and returns it to a raw data message.
[0071] Currently, the traffic rate limiting for fragment messages that the DPU needs to transmit is mainly performed by a token bucket algorithm. When each fragment message arrives, it performs traffic evaluation for it using the token bucket. If it does not exceed the committed traffic or the maximum allowed burst traffic range, it marks the corresponding fragment with a color mark and then transmits it. If not, the subsequent module discards the fragments that exceed the promised rate.
[0072] However, the current DPU fragment message forwarding policy has obvious defects: in the process of forwarding a fragment message, if a fragment in the fragment sequence of an IP message is discarded in the DPU forwarding process, the receiving terminal cannot reconstruct the fragment, and cannot recover the raw data of the IP message; moreover, the process of invalid traffic rate limit evaluation and invalid fragment forwarding appears in the DPU forwarding process, which wastes a large amount of network bandwidth resources and affects the data processing efficiency and application reliability of the DPU.
[0073] Based on this, based on the above embodiment of the fragment message traffic rate limiting method, the present application further provides an embodiment of a DPU fragment message forwarding method. Referring to FIG. 3, the DPU fragment message forwarding method specifically includes the following content:
[0074] Step 400: Locally receive a target fragment message to be forwarded in a DPU.
[0075] Step 500: Processing the received target fragment message according to the fragment message traffic rate limiting method.
[0076] Specifically, the execution content of the fragment message traffic rate limiting method in step 500 may be referred to the embodiment of the fragment message traffic rate limiting method described above, and will not be further described.
[0077] Step 600: If the current color indication of the target fragment message is the retained color, a forwarding process is performed on the target fragment message.
[0078] In order to further improve the efficiency and effectiveness of the DPU fragment message forwarding process, in the DPU fragment message forwarding method according to the embodiment of the present application, the message fragment dyeing table in the DPU fragment message forwarding method is used to store the correspondence between the keyword information and the color mark of the raw data message to which the fragment message belongs; Here, the keyword information of the fragment message includes a unique identity identifier and a local stream label that is preset based on the identifier of the raw data message.
[0079] Specifically, in the DPU scenario, a part of the IP message fragment is unloaded to the DPU, and in order to facilitate the recognition of the stream information, a stream label (generally 2 bytes, used as the local metadata of the message) may be set for each stream (generally a 5-tuple, including source IP, destination IP, protocol type, source port, and destination port), and the stream label only needs to be unique within the device at the same time. In such a scenario, the fragment dyeing table structure of the DPU can be optimized and adjusted, and only the stream label and the identification field are KEY, and the speed limit of the fragment of the own device can be accelerated. Using the stream label is more convenient than the traditional general-purpose message triple (source IP, destination IP, protocol type) information extraction, has better performance, and occupies less table space.
[0080] At the software level, the present application further provides a fragment message traffic rate limiting device for implementing all or part of the fragment message traffic rate limiting method. Referring to FIG. 4, the fragment message traffic rate limiting device specifically includes: a pre-dyeing module 10 for performing a pre-dyeing process on the target fragment message based on a preset message fragment dyeing table, so that the target fragment message is provided with a mark including a discard color and a keep color; a traffic evaluation module 20 for performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on a color signature of the double dyed target fragment message; The present invention also includes a fragment rate limiting module 30 for performing rate limiting processing on the target fragment message if the current color indication of the target fragment message is the discard color, and for performing forwarding processing on the target fragment message if the current color indication of the target fragment message is the retain color.
[0081] The embodiment of the fragment message traffic rate limiting device of the present application can specifically be used to execute the processing flow of the embodiment of the fragment message traffic rate limiting method in the above embodiment, and its function will not be described further, but reference can be made to the detailed description of the embodiment of the fragment message traffic rate limiting method.
[0082] The part of the fragment message traffic rate limiting device performing fragment message traffic rate limiting may be executed in a server, for example, an edge server, and in another practical application situation, all operations may be completed in a client device. Specifically, it may be selected based on the processing capability of the client device and the limitations of the user's usage scenario. This application is not limited thereto. All operations may be completed in the client device, and the client device may further include a processor for specific processing of fragment message traffic rate limiting.
[0083] The above-mentioned client device may have a communication module (i.e., a communication unit) and may be communicatively connected to a remote server to realize data transmission with the server. The server may include a server on the side of a task scheduling center, and in other implementation scenarios may include a server on an intermediate platform, for example, a server on a third-party server platform that is communicatively linked with the task scheduling center server. The server may include a single computer device, a server cluster composed of multiple servers, or a server structure of a distributed device.
[0084] The server and the client device may communicate using any suitable network protocol, including network protocols not yet developed as of the filing date of this application. The network protocols may include, for example, TCP / IP, UDP / IP, HTTP, HTTPS, etc. Of course, the network protocols may further include, for example, RPC (Remote Procedure Call Protocol), REST (Representational State Transfer), etc., which are used on top of the above protocols.
[0085] As can be seen from the above description, the fragment message traffic rate limiting device according to the embodiment of the present application first performs a pre-dyeing process on the received fragment message before adopting a token bucket algorithm to perform traffic evaluation of the fragment message, so that the traffic evaluation process can perform a discard process on the fragment message pre-dyed to a discard color, and there is no need to perform token evaluation in the token bucket for each fragment message, which can effectively reduce the resource consumption of the traffic evaluation process and improve the traffic evaluation efficiency; performs traffic evaluation on the fragment message pre-dyed by the token bucket algorithm based on the non-color blind mode, so as to perform a double-dyeing process on the fragment message, improving the traffic evaluation efficiency, and then discards the fragment message pre-dyed to a retained dye but whose size exceeds the excess area in the token bucket, thereby ensuring the effectiveness and reliability of the fragment message traffic rate limiting; performs a rate limiting process on the fragment message whose color mark of the pre-dyed and double-dyed color mark is a discard color before transmission, thereby effectively avoiding the transmission of invalid fragment messages and improving the transmission efficiency of the fragment message, effectively reducing the waste of network bandwidth resources, so that the receiving terminal equipment can restore the raw data message based on each received fragment message, and further improving the effectiveness and reliability of the fragment message transmission.
[0086] Based on the above-mentioned embodiment of the fragment message traffic rate limiting method and / or the DPU fragment message forwarding method, the present application further provides a DPU fragment message forwarding device for realizing the DPU fragment message forwarding method. Referring to FIG. 5, the DPU fragment message forwarding device specifically includes the following contents:
[0087] A fragment receiving module 40 for locally receiving a target fragment message to be forwarded in the DPU.
[0088] A traffic rate limiting module 50 for performing a forwarding process on the target fragment message if the current color signature of the target fragment message is the retained color.
[0089] Specifically, the execution contents of the fragment message traffic rate limiting method of the traffic rate limiting module 50 refer to the embodiment of the fragment message traffic rate limiting method described above, and will not be further described.
[0090] A fragment forwarding module 60 for performing forwarding processing on fragment messages whose color signature after traffic evaluation is the retained color.
[0091] Specifically, the DPU fragment message forwarding device can be implemented in the DPU as one functional module.
[0092] To further explain this aspect, the present application further provides a specific application example of a fragment message traffic rate limiting method applied to a DPU to solve the problem of unnecessary bandwidth loss when performing traffic rate limiting on a fragment message on a current DPU, and the specific application example of the fragment message traffic rate limiting method applied to the DPU specifically includes the following improvements:
[0093] (1) Introduce message fragment dyeing table to pre-dye fragment messages; The message fragment staining table is an index table. A fragment sequence belonging to one original IP message corresponds to one entry record in the index table. An entry has an additional information in the IP header as a key (e.g., the source IP, destination IP, protocol type, and identification fields in the IP header as a key), and the most recent fragment staining that has passed the token bucket as a value. The first fragment of a message corresponds to a scene where no entry has been queried, which will be described later, and an entry is created and stained green by default.
[0094] Before a fragmented message passes through the token bucket, it first uses the additional information in the message IP header as the KEY and queries the staining information of the entry corresponding to the fragment in the staining table; if the message fragment header is not queried, it creates one new entry record for the fragment sequence to which the message belongs in the entry, and it is stained green by default; if queried, it pre-stains the fragmented message with the queried result.
[0095] Therefore, the process of pre-dying fragments by the message fragment dyeing table is actually to associate each fragment belonging to one original IP message with the color of the nearest bucket fragment in the fragment sequence, further optimizing the traffic evaluation results of the non-color-blind mode token bucket.
[0096] (2) Based on the result of the preliminary staining, combine with the non-colorblind mode token bucket algorithm to perform traffic speed limiting; The color-blind mode is to base the traffic evaluation on the previous staining results when the current message or fragment passes through the token bucket.
[0097] To illustrate the effectiveness of the method, the content of the token bucket algorithm in non-color blind mode is briefly described here by taking the single-speed single-bucket token bucket as an example.
[0098] The algorithm follows these rules when performing traffic assessment on arriving messages:
[0099] 1. If a message is dyed green, but the length of the message (say B) does not exceed the number of tokens remaining in the token bucket (say Tc), then the message is dyed green and Tc=Tc-B.
[0100] 2. If the message is dyed green and its length exceeds the number of tokens remaining in the token bucket, Tc, then the message is dyed red and Tc remains the same.
[0101] 3. If the message is stained red and the message remains red, then Tc remains.
[0102] In this method, traffic evaluation is performed on message fragments using the non-color-blind mode token bucket algorithm, based on the preliminary staining results of the fragment messages in key point 1 above.
[0103] The token bucket algorithm is used for traffic evaluation of subsequent dyed fragments to prevent excessive bursts of fragments due to oversized messages (high number of fragments) or malicious fragmentation attacks from overriding the effective rate limiting, ultimately causing uncontrollable bandwidth preemption and packet loss.
[0104] Therefore, by performing traffic evaluation in conjunction with the preliminary staining results of fragment messages, it can be ensured that as long as one fragment of an IP message is stained red when passing through a bucket, the fragments that arrive later in the fragment sequence will also be stained red. In general, the practice is to forward green stained messages and discard red stained messages in subsequent processing, to avoid invalid fragments occupying bandwidth and causing losses.
[0105] With reference to FIG. 6, taking the single-rate single-bucket token bucket based on the color-blind mode as an example, the flow of the method for limiting the traffic rate of fragment messages by a DPU described in this method is described as follows.
[0106] 1. Start traffic rate limiting for fragment messages in the arriving traffic, and query the staining table using the source IP, destination IP, protocol type, and identification fields in the IP message header as keys; 1.1.If the query is successful, perform pre-staining of the fragment based on the staining information; 1.2. If the query fails, create an entry record in the fragment staining table for the fragment sequence to which the message belongs, using the source IP, destination IP, protocol type, and identification fields in the IP header as the key, and pre-stain the fragment green; 2. Based on the preliminary coloring of the fragment message, perform traffic evaluation on the fragment message using a token bucket algorithm in non-color blind mode; 3. If a fragment message changes from green to red after passing through a bucket, update the staining of the corresponding fragment sequence in the staining table to red; 4. If the current fragment message is the last fragment, delete the entry record corresponding to the current fragment sequence in the fragment staining table; 5. Based on the message staining result, the message is then forwarded or discarded.
[0107] In summary, the method according to the application example of the present application ensures that when the DPU performs traffic evaluation on each fragment of the same IP message received by the DPU, after the first red-stained fragment appears, subsequent fragments are stained red in the traffic evaluation and discarded, thereby avoiding bandwidth loss due to the transmission of invalid fragments in the network.
[0108] The embodiment of the present application further provides an electronic device (i.e., electronic equipment), for example, the electronic device may include a processor, a memory, a receiver, and a transmitter, the processor is used to perform the fragment message traffic rate limiting method or the DPU fragment message forwarding method mentioned in the above embodiment, where the processor and the memory are connected by a bus or other manner, for example, the connection by a bus. The receiver can be connected to the processor and the memory by wire or wirelessly. The electronic device can receive real-time motion data from sensors in the wireless multimedia sensor network and receive original video sequences from the video collection device.
[0109] The processor may be a DPU, or may be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or other chips, or a combination of the above chips.
[0110] The memory can be used as a non-transient computer-readable storage medium to store non-transient software programs, non-transient computer executable programs and modules, such as program commands / modules corresponding to the fragment message traffic rate limiting method or the DPU fragment message forwarding method in the embodiments of the present application. The processor executes the non-transient software programs, commands and modules stored in the memory to perform various functional applications and data processing of the processor, that is, to realize the fragment message traffic rate limiting method or the DPU fragment message forwarding method in the embodiments of the above methods.
[0111] The memory may include a storage program area and a storage data area, where the storage program area may store an operating system, an application program required for at least one function, and the storage data area may store data created by the processor. In addition, the memory may include high-speed random access memory and may further include non-transient state memory, such as at least one disk storage device, flash memory, or other non-transient state solid storage device. In some embodiments, the memory may optionally include memory located remotely with respect to the processor, and these remote memories may be connected to the processor by a network. Examples of such networks include, but are not limited to, the Internet, a corporate intranet, a local area network, a mobile communication network, and combinations thereof.
[0112] When the one or more modules are stored in the memory and executed by the processor, they perform a fragment message traffic rate limiting method or a DPU fragment message forwarding method in an embodiment.
[0113] In some embodiments of the present application, the user equipment may include a processor, a memory, and a transceiver unit, which may include a receiver and a transmitter, and the processor, memory, receiver and transmitter may be connected by a bus system, the memory being used to store computer commands, and the processor being used to execute the computer commands stored in the memory to control the transceiver unit to transmit and receive signals.
[0114] In one implementation, the functions of the receiver and transmitter in this application may be realized by a transmitting / receiving circuit or a dedicated chip for transmitting and receiving, and the processor may be realized by a dedicated processing chip, a processing circuit or a general-purpose chip.
[0115] In another implementation, it is contemplated that a server according to an embodiment of the present application may be implemented using a general purpose computer, i.e., program code implementing the processor, receiver and transmitter functions is stored in memory, and the general purpose processor implements the processor, receiver and transmitter functions by executing the code in memory.
[0116] An embodiment of the present application further provides a computer-readable storage medium having stored thereon a computer program that, when executed by a processor, implements the steps of the fragment message traffic rate limiting method or the DPU fragment message forwarding method. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the art.
[0117] It should be understood by those skilled in the art that the combination of each exemplary component, system, and method described in the embodiments disclosed herein can be realized in hardware, software, or a combination of both. Specifically, whether to implement in hardware or software depends on the specific application and design constraints of the technical proposal. Those skilled in the art can realize the described functions using different methods for each specific application, but such realization should not be considered beyond the scope of this application. When implemented in hardware, it may be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of this application are programs or code segments used to perform the desired tasks. The programs or code segments may be stored in a machine-readable medium or transmitted over a transmission medium or a communication link-up by a data signal borne by a carrier.
[0118] It should be clear that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. Detailed descriptions of well-known methods are omitted here for the sake of brevity. In the above-mentioned embodiments, some specific steps are described and shown as examples. However, the method procedure of the present application is not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps, while understanding the spirit of the present application.
[0119] In this application, features described and / or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments and / or may be combined with or substituted for features of the other embodiments.
[0120] The above description is merely a preferred embodiment of the present application, and is not used to limit the present application. Those skilled in the art may make various modifications and changes to the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application should be included within the protection scope of the present application.
[0121] (Additional Note) (Appendix 1) performing a pre-dyeing process on the target fragment message according to a preset message fragment dyeing table, thereby providing the target fragment message with a color indicator including a discard color and a keep color; performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on the color signature of the double dyed target fragment message; A method for limiting fragment message traffic rate, comprising: performing a rate limiting process on the target fragment message if the current color indication of the target fragment message is the discard color; and performing a forwarding process on the target fragment message if the current color indication of the target fragment message is the keep color.
[0122] (Appendix 2) The above-mentioned preliminary dyeing process for the target fragment message based on the preset message fragment dyeing table is creating a message fragment staining table for storing a correspondence between the color indicators and the indicators of the raw data messages to which the fragment messages belong; 2. The method for limiting fragment message traffic rate as described in claim 1, wherein the raw data message identifier includes a source IP, a destination IP, a protocol type and a unique identity identifier.
[0123] (Appendix 3) The above-mentioned pre-dyeing process is performed on the target fragment message based on the preset message fragment dyeing table, so that the target fragment message is provided with a color mark. Obtaining an indication of a raw data message to which a currently received target fragment message belongs; The fragment message traffic rate limiting method of claim 2, further comprising: searching whether the message fragment dyeing table contains an indicator of the raw data message to which the target fragment message belongs; and if so, performing a preliminary dyeing process on the target fragment message based on a color indicator corresponding to the indicator of the raw data message.
[0124] (Appendix 4) The fragment message traffic rate limiting method of claim 3, further comprising: if an indicator including the raw data message to which the target fragment message belongs is not found in the message fragment staining table, pre-staining the target fragment message with the retained color as a default color; and storing in the message fragment staining table a correspondence between the indicator of the raw data message to which the target fragment message belongs and a color indicator that is the retained color.
[0125] (Appendix 5) performing traffic evaluation on the pre-dyed target fragment message based on the token bucket algorithm of the non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on the color signature of the double dyed target fragment message; According to the token bucket algorithm of the non-color blind mode, firstly recognize the color mark of the current target fragment message, and if the color mark of the recognized and obtained target fragment message is the discard color, confirm that the color mark of the double-stained target fragment message does not change; When the color indicator of the target fragment message is the retained color, determine whether the number of tokens in the corresponding token bucket meets the message length of the target fragment message, and if so, confirm that the double-dyed color indicator of the target fragment message does not change, and reduce the number of tokens in the token bucket that matches the message length of the target fragment message; The fragment message traffic rate limiting method of claim 1, further comprising: if it is determined that the number of tokens in the token bucket does not meet the message length of the target fragment message, changing the double-staining color indicator of the target fragment message to the discard color, and updating the color indicator of the raw data message to which the target fragment message belongs in the message fragment staining table.
[0126] (Appendix 6) Before performing the forwarding process on the target fragment message as described above, determining whether the traffic-evaluated target fragment messages are the last fragment messages of the respective raw data messages; If so, the record of the raw data message to which the target fragment message belongs is deleted from the message fragment staining table.
[0127] (Appendix 7) receiving locally at the DPU a target fragment message to be forwarded; A DPU fragment message forwarding method, comprising: performing processing on the received target fragment message based on the fragment message traffic rate limiting method described in any one of Supplementary Notes 1 to 6; and, if the current color indication of the target fragment message is the retained color, performing a forwarding process on the target fragment message.
[0128] (Appendix 8) The message fragment staining table is used for storing the correspondence between the keyword information of the raw data message to which the fragment message belongs and the color mark; The DPU fragment message forwarding method according to claim 7, wherein the keyword information of the fragment message includes a unique identity indicator and a local stream label that is preset according to the indicator of the raw data message.
[0129] (Appendix 9) a pre-dyeing module for performing a pre-dyeing process on the target fragment message according to a preset message fragment dyeing table, so that the target fragment message is provided with a color indicator including a discard color and a keep color; a traffic evaluation module for performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on a color signature of the double dyed target fragment message; A fragment message traffic rate limiting device comprising: a fragment rate limiting module for performing rate limiting processing on the target fragment message when the current color indication of the target fragment message is the discard color, and for performing forwarding processing on the target fragment message when the current color indication of the target fragment message is the retain color.
[0130] (Appendix 10) An electronic device comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, the processor being used to realize a fragment message traffic rate limiting method described in any one of appendices 1 to 6, or to realize a DPU fragment message forwarding method described in appendix 7, when executing the computer program.
Claims
1. performing a pre-dyeing process on the target fragment message according to a preset message fragment dyeing table, thereby providing the target fragment message with a color indicator including a discard color and a keep color; performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on a color signature of the double dyed target fragment message; A method for limiting fragment message traffic rate, comprising: performing a rate limiting process on the target fragment message if the current color indication of the target fragment message is the discard color; and performing a forwarding process on the target fragment message if the current color indication of the target fragment message is the keep color.
2. The above-mentioned preliminary dyeing process for the target fragment message based on the preset message fragment dyeing table is creating a message fragment staining table for storing a correspondence between the color indicators and the indicators of the raw data messages to which the fragment messages belong; 2. The method of claim 1, wherein the raw data message identifiers include a source IP, a destination IP, a protocol type, and a unique identity identifier.
3. The above-mentioned pre-dyeing process is performed on the target fragment message based on the preset message fragment dyeing table, so that the target fragment message is provided with a color mark; Obtaining an indication of a raw data message to which a currently received target fragment message belongs; The method for limiting fragment message traffic rate as described in claim 2, further comprising: searching whether the message fragment staining table contains an indicator of the raw data message to which the target fragment message belongs; and if so, performing a pre-staining process on the target fragment message based on a color indicator corresponding to the indicator of the raw data message.
4. The method for limiting fragment message traffic rate as described in claim 3, further comprising: when an indicator including the raw data message to which the target fragment message belongs is not found in the message fragment staining table, pre-staining the target fragment message with the retained color as a default color; and storing in the message fragment staining table a correspondence between the indicator of the raw data message to which the target fragment message belongs and a color indicator which is the retained color.
5. performing traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on the color signature of the double dyed target fragment message; According to the token bucket algorithm of the non-color blind mode, firstly recognize the color mark of the current target fragment message, and if the color mark of the recognized and obtained target fragment message is the discard color, confirm that the color mark of the double-stained target fragment message does not change; If the color indicator of the target fragment message is the retained color, determine whether the number of tokens in the corresponding token bucket meets the message length of the target fragment message, and if so, confirm that the double-dyed color indicator of the target fragment message does not change, and reduce the number of tokens in the token bucket that matches the message length of the target fragment message; 2. The method of claim 1, further comprising: if it is determined that the number of tokens in the token bucket does not meet the message length of the target fragment message, changing the color indicator of the target fragment message twice to the discard color, and updating the color indicator of the raw data message to which the target fragment message belongs in the message fragment staining table.
6. Before performing the forwarding process on the target fragment message as described above, determining whether the traffic-evaluated target fragment messages are the last fragment messages of the respective raw data messages; If so, deleting in the message fragment staining table the record of the raw data message to which the target fragment message belongs.
7. receiving locally at the DPU a target fragment message to be forwarded; A DPU fragment message forwarding method, comprising: performing processing on the target fragment message received based on the fragment message traffic rate limiting method described in any one of claims 1 to 6; and performing forwarding processing on the target fragment message if the current color indication of the target fragment message is the retained color.
8. The message fragment staining table is used for storing the correspondence between the keyword information of the raw data message to which the fragment message belongs and the color mark; The DPU fragment message forwarding method of claim 7, wherein the keyword information of the fragment message includes a unique identity indicator and a local stream label that is preset based on the indicator of the raw data message.
9. a pre-dyeing module for performing a pre-dyeing process on the target fragment message according to a preset message fragment dyeing table, so that the target fragment message is provided with a color indicator including a discard color and a keep color; a traffic evaluation module for performing a traffic evaluation on the pre-dyed target fragment message based on a token bucket algorithm in a non-color blind mode, thereby performing a double dyeing process on the target fragment message, and updating the message fragment dyeing table based on a color signature of the double dyed target fragment message; A fragment message traffic rate limiting device comprising: a fragment rate limiting module for performing rate limiting processing on the target fragment message when the current color indication of the target fragment message is the discard color, and for performing forwarding processing on the target fragment message when the current color indication of the target fragment message is the retain color.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, the processor being used to realize the fragment message traffic rate limiting method described in any one of claims 1 to 6 when executing the computer program, or to realize the DPU fragment message transfer method described in claim 7.
Citation Information
Patent Citations
Traffic supervision method, equipment and device, and computer storage medium
CN111064676A
Radio communication apparatus and radio communication method
JP2000349814A
Multiplexing device and packet transfer control method of multiplexing device
JP2008294535A
Relay device, communication system, and communication method
JP2015144390A
Band control system, station side subscriber termination device, aggregate switch, and band control method
JP2017152778A
