Business data network transmission management method and DPU

Abstract

The present application provides a business data network transmission management method and DPU, the method is executed in the DPU, including: according to the business type to which each data packet to be transmitted on the network belongs, respectively searching for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the business type, and outputting each data packet to the target network for transmission based on the QoS policy data of each data packet; real-time status monitoring of the target network and each data packet, and determining whether to dynamically adjust the QoS policy data of the data packet according to the corresponding monitoring results. The present application can realize the accurate identification of different business flows and the allocation of QoS policies to provide high-quality network services, and can improve the flexibility of business data network transmission management, and thus can be applicable to diverse and highly variable business flow transmission scenarios.

Description

Technical Field

[0001] The present application relates to the technical field of traffic management and processing, and in particular to a business data network transmission management method and a DPU. Background Art

[0002] In network communications, QoS (Quality of Service) refers to the classification and management of different types of network traffic to ensure the reliability of network services, the rational use of bandwidth, and the stability of quality. Currently, business traffic management in data centers is an important task. In order to ensure the quality of network services (QoS), data centers need to ensure the priority and bandwidth of different business traffic.

[0003] At present, CPU-based traffic management solutions usually use static QoS policies to manage business traffic. However, CPU-based traffic management solutions cannot effectively and accurately distinguish and optimize various business traffic. And when dealing with highly variable and diverse business traffic, static QoS policies lack flexibility and adaptability and cannot quickly adapt to changes in different business types and network conditions, which may lead to performance degradation and QoS cannot be guaranteed. In addition, when dealing with high-speed traffic in large-scale data centers, traditional CPU-based traffic management solutions cannot meet the real-time processing requirements of a large number of data packets due to the limited computing and processing capabilities of the CPU, resulting in increased latency and decreased throughput.

[0004] In other words, the CPU-based traffic management solution uses static QoS policies, which are insufficiently flexible, unable to adapt to diverse business traffic, unable to handle highly variable traffic and performance bottlenecks, etc. These shortcomings limit the precise management and optimization of different business types and network conditions, leading to instability in performance and QoS. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a service data network transmission management method and a DPU to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present application provides a service data network transmission management method, which is executed in a DPU, and the service data network transmission management method includes:

[0007] According to the service type of each data packet to be transmitted on the network, the QoS policy data of each data packet is searched in the comparison relationship data between the current QoS policy and the service type, and each data packet is output to the target network for transmission based on the QoS policy data of each data packet;

[0008] The target network and each of the data packets are monitored in real time, and it is determined whether to dynamically adjust the QoS policy data of the data packets according to the corresponding monitoring results.

[0009] In some embodiments of the present application, before searching for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the service type according to the service type to which each data packet to be transmitted on the network belongs, the method further includes:

[0010] receiving data packets to be transmitted over the network;

[0011] Parsing the data packet to obtain header information of the data packet;

[0012] Based on a preset service type identification method, the service type to which the data packet belongs is identified according to the header information of the data packet.

[0013] In some embodiments of the present application, the business type identification method includes: feature matching, machine learning or deep data packet inspection;

[0014] Correspondingly, the method of identifying the service type based on the preset service type, identifying the service type to which the data packet belongs according to the header information of the data packet, includes:

[0015] If the service type identification method is the feature matching, extracting the service feature from the header information of the data packet, and based on the service feature, searching the service type to which the data packet belongs in the comparison relationship data between the preset service feature and the service type;

[0016] If the service type identification method is the machine learning, the header information of the data packet is input into a preset machine learning model for identifying the service type, so that the service type identification model outputs the service type to which the data packet belongs correspondingly;

[0017] If the service type identification method is the deep data packet inspection, deep data packet inspection is performed on the header information of the data packet to identify the service type to which the data packet belongs.

[0018] In some embodiments of the present application, before searching for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the service type according to the service type to which each data packet to be transmitted on the network belongs, the method further includes:

[0019] Receive the comparison relationship data between the preset QoS policy and the service type and store it locally;

[0020] Correspondingly, the service data network transmission management method further includes:

[0021] Adaptively adjust the comparison relationship data between the QoS policy and the service type according to the network status of the target network, and / or, if user-defined QoS policy data is received, adjust the comparison relationship data between the QoS policy and the service type based on the user-defined QoS policy data.

[0022] In some embodiments of the present application, the QoS policy data includes: priority and bandwidth allocation weight;

[0023] Correspondingly, according to the service type to which each data packet to be transmitted on the network belongs, searching for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the service type, and outputting each data packet to the target network for transmission based on the QoS policy data of each data packet, including:

[0024] According to the service type of each data packet to be transmitted on the network, the priority and bandwidth allocation weight corresponding to each data packet are respectively searched in the comparison relationship data between the current QoS policy and the service type;

[0025] Sorting each of the data packets in descending order of priority;

[0026] And, allocating the current bandwidth resources of the target network to each of the data packets according to the bandwidth allocation weight of each of the data packets;

[0027] Based on the bandwidth resources corresponding to each of the data packets, the sorted data packets are output to the target network in sequence for transmission.

[0028] In some embodiments of the present application, the priority includes: a high priority, a medium priority and a low priority divided from high to low, wherein the medium priority includes a plurality of intermediate priorities divided from high to low, and each of the intermediate priorities includes a plurality of sub-priorities divided from high to low;

[0029] Correspondingly, the data packets are sorted in descending order of priority, including:

[0030] According to the priorities corresponding to each of the data packets, each of the data packets is placed in a queue corresponding to the high priority, each of the medium priority and the low priority, respectively, wherein the data packets placed in the queue corresponding to each of the medium priority are sorted in descending order according to the sub-priority to which they belong.

[0031] In some embodiments of the present application, allocating the current bandwidth resources of the target network to each of the data packets according to the bandwidth allocation weight of each of the data packets includes:

[0032] Obtaining the current total bandwidth resources of the target network;

[0033] The total bandwidth resources are divided into percentages according to the bandwidth allocation weights of the respective data packets, and the current bandwidth resources of the target network are allocated to the respective data packets based on the corresponding percentage division results.

[0034] In some embodiments of the present application, the step of outputting the sorted data packets in sequence to the target network for transmission based on the bandwidth resources corresponding to the data packets includes:

[0035] Based on the bandwidth resources corresponding to each of the data packets, the data packets in the queues corresponding to the high priority, each of the medium priority and the low priority are output to the target network in sequence for transmission in a first-in-first-out manner.

[0036] In some embodiments of the present application, the real-time status monitoring of the target network and each of the data packets, and determining whether to dynamically adjust the QoS policy data of the data packets according to the corresponding monitoring results, includes:

[0037] Performing real-time status monitoring on the target network and each of the data packets, and judging whether the data packets in any queue exceed a preset number threshold, whether the current network status of the target network meets a preset congestion condition, or whether the broadband resource of any of the data packets is less than a preset resource threshold corresponding to the data packet according to the corresponding monitoring results;

[0038] If the number of packets in any queue exceeds a preset threshold, a packet is extracted from the queue, and the priority corresponding to the extracted packet is modified so that the packet with the modified priority is re-added to the corresponding queue;

[0039] If the current network status of the target network meets the preset congestion condition, the priority and bandwidth allocation weight of the data packet that has not yet been output to the target network are dynamically adjusted, and / or a network congestion prompt message is sent to the sender of the data packet that has not yet been output to the target network;

[0040] If the bandwidth resource of any of the data packets currently exists and is less than the preset resource threshold corresponding to the data packet, the bandwidth allocation weight of the data packet is dynamically adjusted to reallocate the bandwidth resource of the data packet.

[0041] Another aspect of the present application provides a DPU, wherein the DPU is provided with a service data network transmission management module;

[0042] The business data network transmission management module is used in the business data network transmission management method;

[0043] The business data network transmission management module is communicatively connected with the data center to obtain data packets currently sent to the data center from the data center.

[0044] The business data network transmission management method provided by the present application is executed in the DPU, and according to the business type to which each data packet to be transmitted on the network belongs, the QoS policy data of each data packet is searched in the corresponding relationship data between the current QoS policy and the business type, and each data packet is output to the target network for transmission based on the QoS policy data of each data packet; the target network and each data packet are monitored in real time, and it is determined whether to dynamically adjust the QoS policy data of the data packet according to the corresponding monitoring results. By introducing the DPU, the present application can improve the QoS capability of the data center, realize the accurate identification of different business flows and the allocation of QoS policies, so as to provide high-quality network services; by real-time status monitoring of the target network and each data packet and dynamically adjusting the QoS policy, the flexibility of business data network transmission management can be improved, and it can be applied to diverse and highly variable business traffic transmission scenarios; compared with the traditional CPU-based solution, the present application has obvious advantages such as high performance, low latency, high parallel processing capability, flexibility and programmability.

[0045] Additional advantages, purposes, and features of the present application will be partially described in the following description, and will become partially apparent to those skilled in the art after studying the following, or may be learned from the practice of the present application. The purposes and other advantages of the present application can be achieved and obtained by the structures specifically pointed out in the specification and the drawings.

[0046] Those skilled in the art will understand that the purposes and advantages that can be achieved by the present application are not limited to the above specific description, and the above and other purposes that can be achieved by the present application will be more clearly understood based on the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide a further understanding of the present application, constitute a part of the present application, and do not constitute a limitation of the present application. The components in the drawings are not drawn to scale, but are only for the purpose of illustrating the principles of the present application. In order to facilitate the illustration and description of some parts of the present application, the corresponding parts in the drawings may be enlarged, that is, they may become larger relative to other components in the exemplary device actually manufactured according to the present application. In the drawings:

[0048] Figure 1 This is a schematic diagram of a first flow chart of a business data network transmission management method in an embodiment of the present application.

[0049] Figure 2 This is a second flow chart of the business data network transmission management method in one embodiment of the present application.

[0050] Figure 3 This is a third flow chart of the business data network transmission management method in one embodiment of the present application.

[0051] Figure 4 This is a schematic diagram of the execution logic of a DPU-based QoS policy implementation method in an application example of the present application.

[0052] Figure 5 Schematic diagram of the structure of the business data network transmission management module in one embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the implementation modes and the accompanying drawings. Here, the illustrative implementation modes and descriptions of the present application are used to explain the present application, but are not intended to limit the present application.

[0054] It should also be noted here that in order to avoid obscuring the present application due to unnecessary details, only the structures and / or processing steps closely related to the scheme according to the present application are shown in the accompanying drawings, while other details that are not very relevant to the present application are omitted.

[0055] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0056] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0057] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0058] Traditional CPU-based traffic management solutions use static QoS policies but have the following disadvantages:

[0059] 1. Lack of flexibility: Static QoS policies are pre-defined and fixed during design, and cannot be dynamically adjusted according to real-time network conditions and business needs. This results in the inability to quickly adapt and adjust QoS policies when facing different business types and changes in network traffic, resulting in performance degradation and QoS cannot be guaranteed.

[0060] 2. Unable to adapt to diverse business traffic: Static QoS policies are usually based on generalized assumptions and rules, and cannot fully consider and respond to the differences and special needs of different business types. Due to the diversity of business traffic, static QoS policies may not be able to provide personalized optimization and service guarantees for each business type.

[0061] 3. Unable to handle highly variable traffic: Static QoS policies cannot respond to rapid changes and fluctuations in network traffic in a timely manner. When faced with sudden high traffic or congestion, static QoS policies cannot adjust priority and bandwidth allocation in real time, resulting in performance degradation and QoS instability.

[0062] 4. Performance bottleneck: CPU-based traffic management solutions are limited by the computing and processing capabilities of the CPU and cannot meet the real-time processing requirements of high-speed traffic in large-scale data centers. This leads to increased latency, decreased throughput, and performance bottlenecks.

[0063] Therefore, in order to solve the above problems, the present application introduces a data processing unit (DPU) as a key technology for business traffic management. The embodiments of the present application respectively provide a business data network transmission management method, a business data network transmission management module for executing the business data network transmission management method, and a DPU including the business data network transmission management module, which can realize accurate identification of different business traffic and allocation of QoS policies to provide high-quality network services, and can improve the flexibility of business data network transmission management, and thus can be suitable for diverse and highly variable business traffic transmission scenarios.

[0064] The details are described in detail through the following examples.

[0065] Based on this, the embodiment of the present application provides a service data network transmission management method that can be implemented in the DPU, see Figure 1 The business data network transmission management method specifically includes the following contents:

[0066] Step 100: According to the service type of each data packet to be transmitted over the network, the QoS policy data of each data packet is searched in the comparison relationship data between the current QoS policy and the service type, and based on the QoS policy data of each data packet, each data packet is output to the target network for transmission.

[0067] In one or more embodiments of the present application, DPU (Data Processing Unit) refers to a data processing unit, that is, a dedicated hardware accelerator with high performance and low latency, which can effectively process and manage different types of business traffic. The advantage of DPU lies in its highly parallel processing capabilities and dedicated hardware acceleration, which can achieve efficient traffic identification, classification and processing. Compared with traditional CPU-based solutions, DPU can capture and parse the header information of data packets more quickly, and accurately determine the business type through feature extraction and matching algorithms. At the same time, DPU has parallel processing capabilities and can process multiple data packets at the same time, improving processing efficiency.

[0068] In addition, DPU is flexible and programmable, and can define and adjust QoS policies according to the specific needs of the data center. By pre-defining and storing QoS policies for different service types, DPU can select and apply corresponding policies according to the service type of the data packet. At the same time, DPU supports dynamic adjustment, and can make timely adjustments to priority and bandwidth allocation according to the real-time network status and service traffic conditions to ensure the execution of QoS policies and optimize network performance.

[0069] In step 100, the DPU obtains data packets sent to the data center. These data packets can be obtained by the DPU simultaneously or individually in real time. The specific settings can be made according to the actual application scenario. It is understandable that the data packets can refer to compressed packets of business data.

[0070] In one or more embodiments of the present application, the mapping relationship data between QoS policies and service types refers to data used to store the mapping relationship between each QoS policy data and each service type.

[0071] The QoS policy data may include at least one of policy data such as priority, bandwidth allocation weight, delay requirement, and packet loss rate requirement.

[0072] Step 200: Real-time status monitoring is performed on the target network and each of the data packets, and according to the corresponding monitoring results, it is determined whether to dynamically adjust the QoS policy data of the data packets.

[0073] In step 200, the DPU needs to dynamically monitor the network status and service traffic. According to the real-time network status and service requirements, the DPU can make timely adjustments to the priority and bandwidth allocation. For example, according to the network load and the priority of the service type, the bandwidth occupancy ratio of different service types can be adjusted.

[0074] From the above description, it can be seen that the business data network transmission management method provided in the embodiment of the present application can improve the QoS capability of the data center by introducing DPU, and can realize accurate identification of different business traffic and allocation of QoS policies to provide high-quality network services; through real-time status monitoring of the target network and each data packet and dynamic adjustment of QoS policies, the flexibility of business data network transmission management can be improved, and thus it can be applicable to diverse and highly variable business traffic transmission scenarios; compared with traditional CPU-based solutions, the present application has obvious advantages such as high performance, low latency, high parallel processing capability, flexibility and programmability.

[0075] In order to further improve the effectiveness and reliability of obtaining the service type to which each data packet belongs, in a service data network transmission management method provided in an embodiment of the present application, see Figure 2 , the business data network transmission management method further specifically includes the following contents before step 100:

[0076] Step 010: Receive a data packet to be transmitted over the network.

[0077] In step 010, once the data packet arrives at the data center, the DPU obtains the data packet to begin identification and classification.

[0078] Step 020: Parse the data packet to obtain header information of the data packet.

[0079] In step 020, the DPU captures and parses the header information of the data packet, which may include the source address, destination address, transmission protocol type, data packet size, and request header.

[0080] Step 030: Based on a preset service type identification method, the service type to which the data packet belongs is identified according to the header information of the data packet.

[0081] In step 030, by analyzing this information, the DPU can determine the service type to which the data packet belongs. For example, some specific source and destination addresses may be associated with specific services, or a specific transmission protocol may be used for a certain service, and even the size and request header of the data packet may indicate the service type.

[0082] That is to say, the embodiment of the present application can utilize the data packet capture and analysis library supported by the DPU, and use the data plane development kit DPDK (Data Plane Development Kit) library to realize the capture and analysis of data packets. Then all available Ethernet devices are started, and then data packets are received from these devices, and the header information of these data packets is parsed. Finally, the user adds his own code to identify and classify the data packets based on the parsed header information. The DPU outputs the classification results, which can be marked business types or other forms of representation.

[0083] In order to further improve the accuracy, efficiency and effectiveness of identifying the service type to which the data packet belongs according to the header information of the data packet, in a service data network transmission management method provided in an embodiment of the present application, the service type identification method includes: feature matching, machine learning or deep data packet inspection; see Figure 3 , step 030 in the business data network transmission management method specifically includes the following contents:

[0084] Step 031: If the service type identification method is the feature matching, the service feature is extracted from the header information of the data packet, and based on the service feature, the service type to which the data packet belongs is searched in the comparison relationship data between the preset service feature and the service type.

[0085] Specifically, the DPU can extract features related to the service type, i.e., service features, by analyzing the header information obtained through parsing. For example, it can extract specific identifiers in the source and destination addresses to determine whether they are associated with a specific service; it can check whether the transmission protocol type matches a known service protocol; and it can infer the service type based on the size of the data packet and the request header information.

[0086] The DPU then matches the extracted features with predefined service features, which may include a mapping table of source and destination addresses, known service protocols and related features, and association rules between service types and data packet sizes and request headers.

[0087] In order to further improve the automation and intelligence level of identifying the service type to which the data packet belongs based on the header information of the data packet, see Figure 3 Step 030 in the business data network transmission management method may also include the following content:

[0088] Step 032: If the business type identification method is machine learning, the header information of the data packet is input into a preset machine learning model for identifying the business type, so that the business type identification model outputs the business type to which the data packet belongs.

[0089] Specifically, by training the model, the DPU can automatically learn and identify the service type based on the characteristics of the data packet without predefined mapping relationships or rules. This method can provide more accurate and adaptive traffic classification to adapt to the ever-changing service types and traffic patterns. The machine learning model for identifying the service type can select classification models such as decision tree models, support vector machine models, and logistic regression models.

[0090] In order to further improve the accuracy of identifying the service type to which the data packet belongs based on the header information of the data packet, see Figure 3 Step 030 in the business data network transmission management method may also include the following content:

[0091] Step 033: If the service type identification method is the deep data packet inspection, deep data packet inspection is performed on the header information of the data packet to identify the service type to which the data packet belongs.

[0092] Specifically, deep learning and deep packet inspection technologies are used to conduct in-depth analysis of the contents of data packets to identify and classify business traffic. Through deep packet inspection, the DPU can obtain more detailed information and improve the accuracy of identification and classification.

[0093] Among them, deep packet inspection is a technology used to inspect and analyze the contents of packets in real time as they flow through the network, enabling real-time analysis and decision-making. It is used for a variety of purposes, including security, traffic management, data leakage, policy violations, malware, and quality of service. Deep packet inspection allows packet data to be inspected at the application layer of the network, rather than just the header information, which provides more information about the contents of the packet. It involves looking at the actual payload or content of the packet, including the data being transmitted and the application that generated it.

[0094] In order to further improve the efficiency, flexibility and reliability of searching for the QoS policy data of each data packet in the corresponding relationship data between the current QoS policy and the service type, see Figure 2 or Figure 3 , the business data network transmission management method further specifically includes the following contents before step 100:

[0095] Step 040: Receive the comparison relationship data between the preset QoS policy and the service type and store it locally.

[0096] Specifically, QoS policies are predefined and stored in the DPU. Each service type corresponds to one or more QoS policies. These policies can be set according to the characteristics of the service type and the needs of the data center. For example, for services that require low latency, their priority can be set higher; for services with large amounts of data transmission, their bandwidth usage weight can be set higher. When the DPU identifies and classifies data packets, it selects and applies the corresponding QoS policy based on the service type of the data packet.

[0097] In the DPU, QoS policies for different service types are predefined and stored. Each service type corresponds to one or more QoS policies. QoS policies may include but are not limited to the following aspects: priority setting, bandwidth allocation, delay requirements, packet loss rate requirements, etc. These policies are set according to the characteristics of the service type and the needs of the data center.

[0098] Correspondingly, in order to further improve the flexibility and wide applicability of the business data network transmission management, the business data network transmission management method further includes step 050 before step 100 or after step 100, see Figure 2 or Figure 3 , taking step 050 executed before step 100 as an example, step 050 specifically includes the following contents:

[0099] Step 050: Adaptively adjust the comparison relationship data between the QoS policy and the service type according to the network status of the target network, and / or, if user-defined QoS policy data is received, adjust the comparison relationship data between the QoS policy and the service type based on the user-defined QoS policy data.

[0100] Specifically, the adaptive QoS policy setting method can adjust QoS parameters according to the real-time network status and business needs. By monitoring indicators such as network load, latency, and bandwidth utilization, the DPU can dynamically adjust priority and bandwidth allocation to adapt to different network environments and business needs.

[0101] At the same time, users can also customize QoS policies according to their own needs. DPU provides a user interface or configuration interface, allowing users to flexibly set parameters such as priority, bandwidth allocation, and delay requirements. This method can meet the personalized needs of different users or businesses.

[0102] In order to further improve the efficiency, flexibility and reliability of business data network transmission management, in the business data network transmission management method provided in the embodiment of the present application, the QoS policy data may at least include priority and bandwidth allocation weight; correspondingly, see Figure 2, step 100 in the business data network transmission management method specifically includes the following contents:

[0103] Step 110: According to the service type to which each data packet to be transmitted on the network belongs, the priority and bandwidth allocation weight corresponding to each data packet are searched in the comparison relationship data between the current QoS policy and the service type.

[0104] Specifically, when the DPU identifies and classifies a data packet, it first determines the service type to which the data packet belongs based on the aforementioned traffic identification and classification method. For example, if the feature match is successful, the data packet can be marked as a specific service type, such as video stream, audio stream, web request, etc.

[0105] When the DPU identifies and classifies a data packet, it determines the service type to which the data packet belongs based on the aforementioned traffic identification and classification method. Then, based on the service type of the data packet, the DPU selects the corresponding QoS policy. The QoS policy to be used can be determined based on a predefined mapping relationship or rule. For example, a mapping table of service type and QoS policy can be matched, or the QoS policy to be applied can be determined based on the characteristics of the service type. This allows the DPU to subsequently apply the selected QoS policy to the data packet or related service traffic. Based on the selected QoS policy, the DPU can adjust the priority, bandwidth allocation weight, or other related parameters of the data packet.

[0106] In order to distinguish different types of network traffic, QoS introduces some classification tags, among which EF, AF and BE are the three most common tags.

[0107] EF (Expedited Forwarding) is a high-priority tag, mainly used for network traffic with high real-time requirements, such as VoIP (Voice over IP) and video communications. The EF tag indicates that the data packet has the highest priority, the transmission speed should be as fast as possible, and the delay and jitter should be small to ensure real-time performance and stability.

[0108] AF (Assured Forwarding) is a medium-priority tag, mainly used for network traffic that requires a certain priority but does not require strict guarantees, such as online games, file transfers, etc. AF tags classify data packets into four different priority levels, each of which has three different sub-categories, for a total of 12 priorities. Specifically, the four performance categories in the AF tag are AF1, AF2, AF3, and AF4.

[0109] BE (Best Effort) is a low-priority tag, mainly used for non-real-time data streams, such as email, file downloads, etc. The BE tag indicates that the data packet has no special priority processing requirements. The network switch will give priority to data packets of other priorities, and the BE-tagged data packets will be processed with a lower priority.

[0110] In short, the three tags EF, AF, and BE are used to prioritize different types of network traffic to ensure that all types of data can be properly processed and protected during transmission.

[0111] For each service type, the corresponding bandwidth allocation weight can be calculated according to its priority and weight calculation formula.

[0112] EF (Expedited Forwarding): Since EF is a high-priority service type, its weight can be set to the highest value, such as 1.0.

[0113] AF (Assured Forwarding): Based on the four performance categories of AF (AF1, AF2, AF3, and AF4), each category has three different subcategories, for a total of 12 priorities. According to business needs and performance requirements, appropriate weights are assigned to each subcategory, such as 0.8, 0.6, 0.4, etc.

[0114] BE (Best Effort): Since BE is a low-priority service type, its weight can be set to the lowest value, such as 0.2.

[0115] Step 120: Sort the data packets in descending order of priority.

[0116] Specifically, the DPU may adjust the priority of the data packets in a specific manner as follows: sorting the data packets according to the priority rules defined in the QoS policy, and placing the high-priority data packets in front.

[0117] And, step 130: allocating the current bandwidth resources of the target network to each of the data packets according to the bandwidth allocation weight of each of the data packets.

[0118] Specifically, the DPU can allocate available bandwidth to each service flow in proportion to the weight according to the bandwidth allocation weight. Bandwidth allocation can be performed using weighted fair queuing or other appropriate algorithms.

[0119] Weighted Fair Queuing: Allocate bandwidth to each service type based on the weight ratio. For example, if the weight of EF is 1.0, the weight of AF is 0.8, and the weight of BE is 0.2, then EF traffic will get 50% of the bandwidth, AF traffic will get 40% of the bandwidth, and BE traffic will get 10% of the bandwidth.

[0120] Step 140: Based on the bandwidth resources corresponding to each of the data packets, the sorted data packets are output in sequence to the target network for transmission.

[0121] Specifically, the specific way in which the DPU manages and transmits business traffic according to the QoS policy settings can be: based on the results of priority and bandwidth allocation, the DPU processes data packets in the set priority order and allocates appropriate bandwidth to each business traffic.

[0122] That is to say, the DPU processes data packets in order of priority and sends them according to the bandwidth allocation results. High-priority data packets will be processed and sent first, and each service flow will occupy a corresponding proportion of bandwidth for transmission according to its bandwidth allocation weight.

[0123] In steps 110 to 140, the DPU performs priority adjustment and bandwidth allocation according to the QoS policy corresponding to each data packet. Specifically, the DPU first sorts the data packets by priority. High-priority data packets will be processed and sent first. At the same time, the DPU will also control the bandwidth usage of each business flow. This is usually achieved through weights, and the bandwidth usage weight of each business flow determines its available bandwidth ratio. In this process, the DPU needs to dynamically monitor the network status and the status of the business flow in order to make timely adjustments to the priority and bandwidth allocation, which is the content of the aforementioned step 200.

[0124] In order to further improve the efficiency, flexibility and reliability of business data network transmission, in the business data network transmission management method provided in the embodiment of the present application, the priority includes: a high priority EF, a medium priority AF and a low priority BE divided from high to low, wherein the medium priority AF includes a plurality of intermediate priorities divided from high to low, such as AF1, AF2, AF3 and AF4, each of which includes a plurality of sub-priorities divided from high to low; correspondingly, see Figure 3 , step 120 in the business data network transmission management method specifically includes the following contents:

[0125] Step 121: According to the priorities corresponding to each of the data packets, each of the data packets is placed in a queue corresponding to the high priority, each of the medium priority and the low priority, respectively, wherein the data packets placed in the queue corresponding to each of the medium priority are sorted in descending order according to the sub-priority to which they belong.

[0126] For example, a queuing algorithm is used for different types of services, for priority scheduling of three different services (EF, AF, BE):

[0127] A) Three queues are defined: EF queue, AF queue and BE queue. The AF queue includes AF1 queue to AF4 queue, which are used to store data packets of corresponding priorities.

[0128] B) Packets enter queues: Packets are placed in corresponding queues according to their tags. Packets marked with EF are placed in EF queues, packets marked with AF are placed in AF queues, and packets marked with BE are placed in BE queues.

[0129] The above queuing algorithm can reasonably schedule and process different types of services according to the priority of the data packets, ensure the real-time and stability of the EF data packets, provide a certain priority service for the AF data packets, and process the low priority requirements of the BE data packets. The algorithm can be adjusted and optimized according to specific needs to meet the priority requirements of different services and changes in network conditions.

[0130] In order to further improve the efficiency, flexibility and reliability of business data network transmission, in the business data network transmission management method provided in the embodiment of the present application, see Figure 3 , step 130 in the business data network transmission management method specifically includes the following contents:

[0131] Step 131: Obtain the current total bandwidth resources of the target network.

[0132] Step 132: Divide the total bandwidth resources by percentage according to the bandwidth allocation weight of each of the data packets, and allocate the current bandwidth resources of the target network to each of the data packets based on the corresponding percentage division results.

[0133] Specifically, the pseudo code corresponding to step 132 is shown in Table 1.

[0134] Table 1

[0135]

[0136]

[0137] In the above pseudo code, a coefficient variable is added to adjust the weight ratio. This coefficient can be set according to actual needs to achieve more flexible bandwidth allocation.

[0138] In Table 1, EF_weight stands for the weight of the expedited forwarding queue, indicating that the data packet has the highest priority and needs to be transmitted quickly. AF stands for the guaranteed forwarding queue, indicating medium-priority data packets, which are used for network traffic that requires a certain priority but does not need to be strictly guaranteed. There are four queues in the AF queue, AF1_weight represents the weight of guaranteed forwarding queue 1, AF2_weight represents the weight of guaranteed forwarding queue 2, AF3_weight represents the weight of guaranteed forwarding queue 3, and AF4_weight represents the weight of guaranteed forwarding queue 4.

[0139] BE is a best-effort queue, which indicates low-priority packets and is mainly used for non-real-time data flows. BE_weight indicates the best-effort queue weight. The BE queue is based on the AF queue value. n_weight , where n ranges from 1 to 4.

[0140] Total_Bandwidth indicates the total bandwidth, which represents the total transmission capacity of the entire network or a specific part. This is calculated based on the actual situation and is related to the performance of the device after leaving the factory.

[0141] coefficient is a constant coefficient used to adjust the weight of calculating the bandwidth of each service category. By adjusting this coefficient, the bandwidth allocation ratio of each service category can be affected.

[0142] EF_bandwidth indicates the bandwidth of the accelerated forwarding queue; AF1_bandwidth indicates the bandwidth of the guaranteed forwarding queue 1; AF2_bandwidth indicates the bandwidth of the guaranteed forwarding queue 2; AF3_bandwidth indicates the bandwidth of the guaranteed forwarding queue 3; AF4_bandwidth indicates the bandwidth of the guaranteed forwarding queue 4; BE_bandwidth indicates the bandwidth of the best-effort queue.

[0143] In order to further improve the efficiency, flexibility and reliability of business data network transmission, in the business data network transmission management method provided in the embodiment of the present application, see Figure 3 , step 140 in the business data network transmission management method specifically includes the following contents:

[0144] Step 141: Based on the bandwidth resources corresponding to each of the data packets, the data packets in the queues corresponding to the high priority, each of the medium priority, and the low priority are output to the target network in sequence for transmission in a first-in-first-out manner.

[0145] In one example, the specific process of processing data packets in the priority order may be: taking a data packet from the EF queue for processing and transmission. If the EF queue is empty, taking a data packet from the AF queue for processing and transmission. If the AF queue is also empty, taking a data packet from the BE queue for processing and transmission.

[0146] Each queue processes packets in a first-in-first-out (FIFO) manner, that is, in each queue, the packets that enter the queue first will be processed and transmitted first.

[0147] In order to further improve the efficiency, flexibility and reliability of business data network transmission, in the business data network transmission management method provided in the embodiment of the present application, see Figure 2 or Figure 3 , step 200 in the business data network transmission management method specifically includes the following contents:

[0148] Step 210: Perform real-time status monitoring on the target network and each of the data packets, and determine whether the data packets in any queue exceed a preset number threshold, whether the current network status of the target network meets a preset congestion condition, or whether there is any data packet whose broadband resource is less than the preset resource threshold corresponding to the data packet based on the corresponding monitoring results.

[0149] Step 220: If the number of packets in any queue exceeds a preset threshold, extract packets from the queue and modify the priority of the extracted packets so that the packets with modified priority are re-added to the corresponding queue.

[0150] Step 230: If the current network status of the target network meets the preset congestion condition, the priority and bandwidth allocation weight of the data packet that has not yet been output to the target network are dynamically adjusted, and / or a network congestion prompt message is sent to the sender of the data packet that has not yet been output to the target network.

[0151] Step 240: If the bandwidth resource of any of the data packets currently exists and is smaller than the preset resource threshold corresponding to the data packet, the bandwidth allocation weight of the data packet is dynamically adjusted to reallocate the bandwidth resource of the data packet.

[0152] Specifically, the DPU dynamically monitors the network status and business traffic conditions. Based on the real-time network conditions and business needs, the DPU can make timely adjustments to the priority and bandwidth allocation. For example, if the network is congested or a certain business traffic requires more bandwidth, the DPU can recalculate the weight and reallocate the bandwidth.

[0153] That is to say, the priority scheduling strategy can be adjusted dynamically according to the real-time network status and business needs. For example, if there are too many packets in the EF queue, the priority scheduling weight can be adjusted so that the packets in the EF queue can be processed and transmitted faster. The number of packets and network status of each queue are monitored in real time. Based on the monitoring results, feedback information can be provided to the sender, such as sending a congestion signal to the sender through a congestion control mechanism to control the sending rate of the data packets.

[0154] At the same time, DPU has the ability to recover from faults and handle exceptions. When a fault or abnormal situation occurs, such as network interruption or packet loss, DPU can take corresponding measures, such as resending lost data packets, adjusting priorities or bandwidth allocation, etc., to ensure business continuity and reliability.

[0155] In order to further illustrate the service data network transmission management method provided by the above embodiment, the present application also provides an application example of a QoS policy implementation method based on DPU, which can guarantee the priority and bandwidth of different service flows. Specifically, DPU will first identify and classify the incoming traffic, and then set the priority and allocate bandwidth according to the preset QoS policy. Finally, DPU transmits and manages the service traffic according to the adjusted priority and bandwidth.

[0156] See also Figure 4 After the DPU captures the data packet from the data stream, it performs header information analysis, feature extraction, feature matching and service type judgment in sequence, and then matches the service type according to the obtained service type. In this process, the network status and service traffic are monitored in real time; then, according to the matched service type, the QoS policy corresponding to the service type is selected in the policy warehouse formed in advance based on the defined policy rules and RBAC access, and then the service policy execution point is distinguished, and then the packet priority is adjusted according to the packet classification model, and then the packet is added to each queue in the QoS policy queue (such as EF queue, AF1 queue, AF2 queue and BE queue) according to the priority, and then the bandwidth allocation weight is calculated, bandwidth is allocated, and data packets are processed and sent according to the priority, and fault recovery and exception handling are performed when faults are detected.

[0157] The specific application examples of the DPU-based QoS policy implementation method include the following:

[0158] (1) Based on the results of priority and bandwidth allocation, the DPU transmits and manages service traffic. Specifically, data packets are sent in order of priority. Each service traffic also occupies a corresponding proportion of bandwidth according to its bandwidth usage weight. Throughout the process, the DPU needs to keep monitoring the network status and service traffic to ensure the correct execution of QoS policies and respond quickly to any sudden network conditions.

[0159] (2) Data packet reception and sorting: The DPU receives data packets and sorts them according to their priority. High-priority data packets are sorted first, and low-priority data packets are sorted last.

[0160] (3) Bandwidth allocation weight calculation: The DPU calculates the bandwidth allocation weight for each service flow based on the bandwidth allocation rules defined in the QoS policy. These weights determine the proportion of available bandwidth that each service flow occupies.

[0161] (4) Packet transmission and bandwidth management: The DPU sends packets in order of priority and manages bandwidth usage based on bandwidth allocation weights. High-priority packets are sent first, and each service flow occupies a corresponding proportion of bandwidth based on its bandwidth allocation weight. This ensures that high-priority services obtain more bandwidth resources.

[0162] (5) Network status and service flow monitoring: The DPU continuously monitors the network status and service flow. By real-time monitoring of indicators such as network latency and bandwidth utilization, as well as analyzing service flow, the DPU can understand the actual network status and service needs.

[0163] (6) QoS policy execution and response: The DPU ensures the correct execution of QoS policies and responds quickly to any unexpected network conditions. If network congestion or other abnormal conditions occur, the DPU can dynamically adjust the priority and bandwidth allocation based on the real-time monitoring of network status and service traffic to maintain QoS performance.

[0164] Fault recovery and exception handling: DPU has the ability to recover from faults and handle exceptions. When a fault or anomaly occurs, such as network interruption or packet loss, DPU can take appropriate measures, such as resending lost data packets, adjusting priority or bandwidth allocation, etc., to ensure business continuity and reliability.

[0165] The process describes the implementation steps of DPU-based priority adjustment and bandwidth allocation, including packet reception, QoS policy selection, packet priority adjustment, bandwidth allocation weight calculation, bandwidth allocation, monitoring and adjustment, and packet processing and sending.

[0166] In summary, the DPU-based service traffic management solution has obvious advantages over the traditional CPU-based solution, including high performance, low latency, high parallel processing capability, flexibility and programmability, etc. By introducing DPU, the QoS capability of the data center can be improved, and accurate identification, priority adjustment and bandwidth allocation of different service traffic can be achieved to provide high-quality network services.

[0167] That is to say, the business data network transmission management method provided in the embodiment of the present application adopts a queuing algorithm and priority scheduling mechanism for different business priorities by designing a DPU-based traffic identification and classification method, a QoS policy setting and implementation method, priority adjustment and bandwidth allocation operations, a business traffic transmission and management method, and a queuing algorithm and priority scheduling. By defining different queues and processing data packets in order of priority, effective management and scheduling of EF, AF and BE services are achieved. The core innovations of this part include the definition of queues and the mechanism for data packets to enter queues, as well as the process of scheduling data packets according to priority.

[0168] By using the technical solution of this application, the business traffic of the data center can be managed more effectively to ensure the service quality of different businesses. At the same time, the traffic management solution based on DPU can more flexibly respond to changes in business traffic, thereby improving the overall performance and efficiency of the data center.

[0169] The present application also provides a business data network transmission management module for executing all or part of the business data network transmission management method, see Figure 5 The service data network transmission management module set in the DPU specifically includes the following contents:

[0170] The QoS policy acquisition module 10 is used to search for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the service type according to the service type of each data packet to be transmitted on the network, and output each data packet to the target network for transmission based on the QoS policy data of each data packet;

[0171] The QoS policy dynamic adjustment module 20 is used to monitor the status of the target network and each of the data packets in real time, and determine whether to dynamically adjust the QoS policy data of the data packets according to the corresponding monitoring results.

[0172] The embodiment of the business data network transmission management module provided in the present application can be specifically used to execute the processing flow of the embodiment of the business data network transmission management method in the above-mentioned embodiment. Its functions are not repeated here, and reference can be made to the detailed description of the embodiment of the above-mentioned business data network transmission management method.

[0173] The part of the business data network transmission management module that performs business data network transmission management can be completed in the client device. It can be selected based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor for specific processing of business data network transmission management.

[0174] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0175] The server and the client device may communicate with each other using any suitable network protocol, including network protocols that have not yet been developed on the date of filing this application. The network protocols may include, for example, TCP / IP, UDP / IP, HTTP, HTTPS, etc. Of course, the network protocols may also include, for example, RPC (Remote Procedure Call Protocol) and REST (Representational State Transfer) protocols used on top of the above protocols.

[0176] From the above description, it can be seen that the business data network transmission management module provided in the embodiment of the present application can improve the QoS capability of the data center by introducing DPU, and can realize accurate identification of different business traffic and allocation of QoS policies to provide high-quality network services; through real-time status monitoring of the target network and each data packet and dynamic adjustment of QoS policies, it can improve the flexibility of business data network transmission management, and thus can be suitable for diverse and highly variable business traffic transmission scenarios; compared with traditional CPU-based solutions, the present application has obvious advantages such as high performance, low latency, high parallel processing capability, flexibility and programmability.

[0177] The embodiment of the present application also provides a DPU, which may include a service data network transmission management module;

[0178] The service data network transmission management module is used to execute the service data network transmission management method provided in the above embodiment;

[0179] The business data network transmission management module is communicatively connected to the data center, and can also be arranged in the data center to obtain data packets currently sent to the data center from the data center.

[0180] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned business data network transmission management method are implemented. The computer-readable storage medium can 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 technical field.

[0181] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is specifically performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. When implemented in hardware, it can 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 required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0182] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0183] In the present application, features described and / or illustrated for one embodiment may be used in the same manner or in a similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0184] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the embodiments of the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for managing business data network transmission, characterized in that: Executed in the DPU, the service data network transmission management method includes: receiving data packets to be transmitted over the network; Parsing the data packet to obtain header information of the data packet; Based on a preset service type identification method, the service type to which the data packet belongs is identified according to the header information of the data packet; the service type identification method includes: feature matching, machine learning or deep data packet detection; Correspondingly, the method of identifying the service type based on the preset service type, identifying the service type to which the data packet belongs according to the header information of the data packet, includes: If the service type identification method is the feature matching, extracting the service feature from the header information of the data packet, and based on the service feature, searching the service type to which the data packet belongs in the comparison relationship data between the preset service feature and the service type; If the service type identification method is the machine learning, the header information of the data packet is input into a preset machine learning model for identifying the service type, so that the service type identification model outputs the service type to which the data packet belongs correspondingly; If the service type identification method is the deep data packet inspection, performing deep data packet inspection on the header information of the data packet to identify the service type to which the data packet belongs; Adaptively adjust the comparison relationship data between QoS strategy and service type according to the network status of the target network; According to the service type of each data packet to be transmitted on the network, the QoS policy data of each data packet is searched in the comparison relationship data between the current QoS policy and the service type, and each data packet is output to the target network for transmission based on the QoS policy data of each data packet; Performing real-time status monitoring on the target network and each of the data packets, and determining whether to dynamically adjust the QoS policy data of the data packets according to the corresponding monitoring results; The real-time status monitoring of the target network and each of the data packets, and determining whether to dynamically adjust the QoS policy data of the data packets according to the corresponding monitoring results, includes: Performing real-time status monitoring on the target network and each of the data packets, and judging whether the data packets in any queue exceed a preset number threshold, whether the current network status of the target network meets a preset congestion condition, or whether the broadband resource of any of the data packets is less than a preset resource threshold corresponding to the data packet according to the corresponding monitoring results; If the number of packets in any queue exceeds a preset threshold, a packet is extracted from the queue, and the priority corresponding to the extracted packet is modified so that the packet with the modified priority is re-added to the corresponding queue; If the current network status of the target network meets the preset congestion condition, the priority and bandwidth allocation weight of the data packet that has not yet been output to the target network are dynamically adjusted, and / or a network congestion prompt message is sent to the sender of the data packet that has not yet been output to the target network; If the bandwidth resource of any of the data packets currently exists and is less than the preset resource threshold corresponding to the data packet, the bandwidth allocation weight of the data packet is dynamically adjusted to reallocate the bandwidth resource of the data packet.

2. The business data network transmission management method according to claim 1, characterized in that: Before searching for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the service type according to the service type to which each data packet to be transmitted on the network belongs, the method further includes: Receive the comparison relationship data between the preset QoS policy and the service type and store it locally; If user-defined QoS policy data is received, the mapping relationship data between the QoS policy and the service type is adjusted based on the user-defined QoS policy data.

3. The business data network transmission management method according to claim 1, characterized in that: The QoS policy data includes: priority and bandwidth allocation weight; Correspondingly, according to the service type to which each data packet to be transmitted on the network belongs, searching for the QoS policy data of each data packet in the comparison relationship data between the current QoS policy and the service type, and outputting each data packet to the target network for transmission based on the QoS policy data of each data packet, including: According to the service type of each data packet to be transmitted on the network, the priority and bandwidth allocation weight corresponding to each data packet are respectively searched in the comparison relationship data between the current QoS policy and the service type; Sorting each of the data packets in descending order of priority; And, allocating the current bandwidth resources of the target network to each of the data packets according to the bandwidth allocation weight of each of the data packets; Based on the bandwidth resources corresponding to each of the data packets, the sorted data packets are output to the target network in sequence for transmission.

4. The business data network transmission management method according to claim 3, characterized in that: The priorities include: a high priority, a medium priority and a low priority divided from high to low, wherein the medium priority includes a plurality of intermediate priorities divided from high to low, and each of the intermediate priorities includes a plurality of sub-priorities divided from high to low; Correspondingly, the data packets are sorted in descending order of priority, including: According to the priorities corresponding to each of the data packets, each of the data packets is placed in a queue corresponding to the high priority, each of the medium priority and the low priority, respectively, wherein the data packets placed in the queue corresponding to each of the medium priority are sorted in descending order according to the sub-priority to which they belong.

5. The business data network transmission management method according to claim 3, characterized in that: Allocating the current bandwidth resources of the target network to each of the data packets according to the bandwidth allocation weight of each of the data packets includes: Obtaining the current total bandwidth resources of the target network; The total bandwidth resources are divided into percentages according to the bandwidth allocation weights of the respective data packets, and the current bandwidth resources of the target network are allocated to the respective data packets based on the corresponding percentage division results.

6. The business data network transmission management method according to claim 4, characterized in that: The step of outputting the sorted data packets in sequence to the target network for transmission based on the bandwidth resources corresponding to the data packets includes: Based on the bandwidth resources corresponding to each of the data packets, the data packets in the queues corresponding to the high priority, each of the medium priority and the low priority are output to the target network in sequence for transmission in a first-in-first-out manner.

7. A DPU, characterized in that: The DPU is provided with a service data network transmission management module; The service data network transmission management module is used to execute the service data network transmission management method according to any one of claims 1 to 6; The business data network transmission management module is communicatively connected with the data center to obtain data packets currently sent to the data center from the data center.

Images