Traffic control method, electronic equipment and storage medium

Abstract

The invention relates to the technical field of communication, and discloses a flow control method, electronic equipment and a storage medium, and the method comprises the steps: responding to data transmission of a target node, configuring a first-level cache of the target node, and configuring a data cache position corresponding to the target node as the first-level cache; under the condition that the data volume in the first-level cache reaches a preset back pressure threshold value, switching a data cache position corresponding to the target node to a second-level cache, and performing flow control on the target node; wherein the second-level cache is configured to be a shared cache of all nodes transmitting data to the local node. According to the invention, the problem of cache overflow caused by traffic burst of the switching network node can be solved.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a flow control method, an electronic device, and a storage medium. Background Art

[0002] In today's digital age, data traffic in switching networks is rapidly increasing. When a node in a switching network receives data from a remote node, it typically caches the data in a local buffer pending processing. However, if a local node receives data from multiple remote nodes simultaneously, traffic bursts can easily cause the local buffer to overflow. Summary of the Invention

[0003] In view of this, the present disclosure provides a flow control method, an electronic device, and a storage medium to solve the problem of buffer overflow caused by traffic bursts in switching network nodes.

[0004] In a first aspect, the present disclosure provides a traffic control method applicable to a node in a switching network, the method comprising:

[0005] In response to data transmission from a target node, configuring a first-level cache of the target node, and configuring a data cache location corresponding to the target node as the first-level cache;

[0006] When the amount of data in the first-level cache reaches a preset back pressure threshold, the data cache location corresponding to the target node is switched to the second-level cache, and flow control is performed on the target node; wherein the second-level cache is configured as a shared cache for all nodes transmitting data to the local node.

[0007] In this method, a separate first-level cache is configured for the target node to cache the target node's data. This ensures that data processing for different target nodes is independent of each other, allowing for targeted processing of data from different target nodes. When the amount of data in the first-level cache reaches a preset backpressure threshold, the target node's data cache location is switched from the first-level cache to the shared second-level cache. This allows data about to overflow from the first-level cache to be stored in the second-level cache, centrally allocating cache resources for local nodes during peak traffic periods. This improves the overall utilization of the local node's cache space, effectively preventing cache overflows in switching network nodes due to traffic bursts, and thus reducing data loss and transmission delays caused by cache overflows.

[0008] In an optional implementation, switching the data cache location corresponding to the target node to the secondary cache and performing flow control on the target node includes:

[0009] Acquire target network performance information of the target node and target transmission resource information of the local node;

[0010] Based on the target network performance information, the target transmission resource information, and a target prediction model, target flow control information for the target node is obtained; wherein the target prediction model is obtained by training an initial prediction model based on associated historical network performance information, historical transmission resource information, and historical flow control information; the target flow control information includes back pressure control information and configuration information of the secondary cache;

[0011] Switching the data cache location corresponding to the target node to the secondary cache, and caching the data of the data stream sent by the target node in the secondary cache according to the configuration information;

[0012] Flow control is performed on the target node based on the back pressure control information.

[0013] In this method, a target prediction model is used to determine backpressure control information and L2 cache configuration information for the target node based on traffic characteristics reflected by both target network performance information and target transmission resource information. When the data cache location corresponding to the target node is switched to the L2 cache, data from the data stream sent by the target node is cached in the L2 cache according to the configuration information, and traffic control is performed on the target node based on the backpressure control information. This ensures that during the target node's backpressure control process, the local node can fully process the target node's cached data, preventing data backlogs and loss.

[0014] In an optional implementation, the acquiring target network performance information of the target node includes:

[0015] Obtaining a traffic bandwidth corresponding to the target node and sending first indication information to the target node; wherein the first indication information is used to instruct the target node to periodically insert a first idle data packet during data transmission with the local node; the first idle data packet carries status information of the target node;

[0016] Receive the first idle data packet sent by the target node, and extract the status information of the target node from the first idle data packet to obtain the target network performance information; wherein the target network performance information includes the traffic bandwidth and the status information.

[0017] In this method, when the amount of data in the target node's first-level cache reaches a preset backpressure threshold, the local node sends a first indication message to the target node, causing the target node to periodically insert a first idle data packet during data transmission. Because the local node does not cache the first idle data packet, the local node's information processing efficiency is improved. Furthermore, the first idle data packet carries the target node's status information. Therefore, upon receiving the first idle data packet, the local node can extract the target node's status information from it and, combined with the target node's traffic bandwidth calculated in real time, provide rich reference information for a target prediction model to predict target traffic control information, thereby improving the accuracy of the target prediction model's prediction results.

[0018] In an optional implementation, when there are multiple target nodes, the receiving the first idle data packet sent by the target node, and extracting the state information of the target node from the first idle data packet to obtain the target network performance information includes:

[0019] receiving the first idle data packet sent by the target node, and extracting the status information of the target node from the first idle data packet;

[0020] Writing the status information of the target node into a preset information table;

[0021] If the amount of data in the first-level cache corresponding to multiple target nodes reaches the preset back pressure threshold, the traffic heat of each target node is obtained;

[0022] Target state information is obtained from the preset information table based on the traffic heat to obtain the target network performance information; wherein the target network performance information includes the target state information and the traffic bandwidth of the target node corresponding to the target state information.

[0023] In this method, target state information is retrieved from the state information of multiple target nodes stored in a preset information table based on the target node's traffic popularity. This information is then combined with the traffic bandwidth of the corresponding target node to obtain target network performance information. This allows local nodes to prioritize using a target prediction model to predict target flow control information for target nodes with high traffic popularity, allowing them to promptly adjust flow control for these target nodes, avoiding bandwidth utilization reductions caused by prolonged flow control, thereby ensuring stable operation of the switching network.

[0024] In an optional implementation, the configuration information includes a valid time range for the secondary cache to provide a cache service to the target node; and caching data of the data stream sent by the target node in the secondary cache according to the configuration information includes:

[0025] Within the valid time range, the data of the data stream sent by the target node is cached in the secondary cache.

[0026] In this method, when the amount of data in the first-level cache corresponding to the target node reaches a preset back pressure threshold, the target prediction model gives the effective time range for the second-level cache to provide cache services to the target node, and the data of the target node is cached in the second-level cache within the effective time range. Therefore, it can avoid the target node occupying the resources of the shared second-level cache for a long time, so as to centrally allocate the resources of the second-level cache and ensure the high efficiency of network data processing.

[0027] In an optional implementation, the configuration information further includes a cache size allocated to the target node in the secondary cache; and caching data of the data stream sent by the target node in the secondary cache within the valid time range includes:

[0028] Allocate a corresponding cache area for the target node in the secondary cache according to the cache size;

[0029] Within the valid time range, the data of the data stream sent by the target node is cached in the cache area.

[0030] In this method, since the second-level cache is shared and the traffic characteristics of each target node are different, when the amount of data in the first-level cache corresponding to the target node reaches the preset back pressure threshold, the target prediction model gives the corresponding cache size allocated to each target node in the second-level cache, thereby ensuring that the local node can process the cache data of the target node during the back pressure control process of the target node.

[0031] In an optional implementation, the back pressure control information includes a target back pressure period and a corresponding number of periods; and the performing flow control on the target node based on the back pressure control information includes:

[0032] Based on the back pressure control information, second indication information is sent to the target node to perform flow control on the target node; wherein the second indication information is used to instruct the target node to periodically insert a second idle data packet during data transmission with the local node according to the target back pressure period and the number of periods.

[0033] In this method, when the amount of data in the first-level cache corresponding to the target node reaches the preset back pressure threshold, the target prediction model gives the target back pressure period and its corresponding number of periods. Therefore, the back pressure period and number of periods of each target node can be dynamically adjusted according to the traffic characteristics of each target node, thereby avoiding the reduction of bandwidth utilization due to long-term back pressure and improving network performance.

[0034] In an optional embodiment, the back pressure control information also includes a target back pressure level; the target node is configured with a correspondence between at least one preset back pressure level and the number of idle data packets; the preset back pressure level is positively correlated with the number of idle data packets; the second indication information is used to instruct the target node to periodically insert a number of second idle data packets corresponding to the target back pressure level during data transmission with the local node based on the target back pressure period, the number of periods and the target back pressure level.

[0035] In this method, when the amount of data in the first-level cache corresponding to the target node reaches the preset back pressure threshold, the target prediction model gives the target back pressure cycle, the number of cycles and the target back pressure level, and hierarchically controls the number of second idle data packets sent by the target node in each target back pressure cycle. Therefore, the traffic of each target node can be accurately adjusted according to its traffic characteristics to prevent congestion.

[0036] In a second aspect, the present disclosure provides an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions being stored in the memory, and the processor executing the flow control method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0037] In a third aspect, the present disclosure provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the flow control method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 is a schematic diagram of an architecture of a switching network according to an embodiment of the present disclosure;

[0040] Figure 2 is a flow chart of a flow control method according to an embodiment of the present disclosure;

[0041] Figure 3 is a flow chart of another flow control method according to an embodiment of the present disclosure;

[0042] Figure 4is a structural block diagram of a flow control device according to an embodiment of the present disclosure;

[0043] Figure 5 It is a structural block diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0044] To make the purpose, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present disclosure.

[0045] In today's digital age, data traffic on switching networks is growing rapidly, and traffic management technology faces the following challenges:

[0046] 1. Regarding cache management, when a local node receives data from multiple remote nodes, related technologies often use a single cache strategy. However, if traffic bursts occur, the cache can overflow, leading to data loss or transmission delays, which in turn affects network service quality. For example, when a video conference is running concurrently with a file download, video freezes may occur.

[0047] 2. In terms of flow control, back pressure technology lacks flexibility. It usually stops the sender from sending data when the cache reaches the threshold. This not only wastes bandwidth but also causes network jitter, reducing network utilization and stability.

[0048] 3. In terms of information exchange, information between nodes is not timely and comprehensive. Local nodes struggle to keep up with the status of remote nodes, making it difficult to adjust strategies based on real-time network changes. Furthermore, the predictive models used to predict network status lack accurate predictions of traffic changes due to their limited data dimensions and outdated algorithms, leading to lagging traffic management.

[0049] In summary, current switching network technology has many shortcomings, and there is an urgent need for a technical solution that optimizes cache management, accurately controls traffic, and efficiently interacts to improve the overall performance of the switching network and meet the growing needs of network applications.

[0050] In view of this, according to an embodiment of the present disclosure, a flow control method embodiment is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0051] See also Figure 1 , Figure 1 This is a schematic diagram of the architecture of a switching network provided by the present disclosure. The switching network includes multiple nodes and switches, and multiple nodes are interconnected through switches. Among them, the host can connect to the target node (see Figure 1 The remote nodes 1 to 3 in the network are interconnected to receive the data stream sent by the target node. The local node caches the data in the received data stream locally, waiting to be read and processed.

[0052] In this embodiment, a flow control method is provided, which can be used for nodes in the above switching network. Figure 2 is a flow chart of a flow control method according to an embodiment of the present disclosure, such as Figure 2 As shown, the process includes the following steps:

[0053] Step S201 : in response to data transmission of a target node, configuring a first-level cache of the target node, and configuring a data cache location corresponding to the target node as a first-level cache.

[0054] Optionally, the target node is a remote node that has data to transmit to the local node.

[0055] In practical applications, local nodes can set up independent L1 caches for each target node based on its traffic characteristics. The L1 cache capacity is configured based on the target node's traffic characteristics. The L1 cache is used to cache data from the corresponding target node, ensuring independent data processing at different target nodes and preventing interference.

[0056] Step S202: When the amount of data in the first-level cache reaches a preset back pressure threshold, the data cache location corresponding to the target node is switched to the second-level cache, and flow control is performed on the target node; wherein the second-level cache is configured as a shared cache for all nodes transmitting data to the local node.

[0057] In actual applications, the local node can configure a preset backpressure threshold for each L1 cache based on the traffic characteristics of each target node. When the amount of data in any L1 cache reaches the corresponding preset backpressure threshold, the data cache location corresponding to the target node is switched to the L2 cache. Alternatively, a preset percentage of the L1 cache capacity can be used as the preset backpressure threshold, for example, 95% of the L1 cache capacity.

[0058] Understandably, when the amount of data in the L1 cache reaches a preset backpressure threshold, indicating that the L1 cache is about to be saturated, the shared L2 cache is immediately activated to store the data that is about to overflow. It should be noted that all nodes transmitting data to the local node described in step S202 above include the target node.

[0059] This embodiment employs a two-level cache architecture. The first-level cache operates independently, enabling targeted processing based on the traffic characteristics of each target node. The second-level cache is shared, allowing centralized resource allocation during peak traffic times, improving overall cache space utilization. This effectively reduces data loss and transmission delays, ensuring efficient and stable network data processing.

[0060] The flow control method provided in this embodiment configures a separate first-level cache for the target node to cache the data of the target node. Therefore, it can ensure that the data processing of different target nodes is independent of each other, so as to perform targeted processing on the data of different target nodes. When the amount of data in the first-level cache reaches a preset back pressure threshold, the data cache location of the target node is switched from the first-level cache to the shared second-level cache. Therefore, data that is about to overflow from the first-level cache can be stored in the second-level cache, so that the cache resources of the local node can be centrally allocated during traffic peaks, thereby improving the overall utilization rate of the cache space of the local node, thereby effectively avoiding cache overflows caused by traffic bursts in the switching network node, and further reducing data loss and transmission delays caused by cache overflows.

[0061] In this embodiment, a flow control method is provided, which can be used for nodes in the above switching network. Figure 3 is a flow chart of a flow control method according to an embodiment of the present disclosure, such as Figure 3 As shown, the process includes the following steps:

[0062] Step S301: In response to the data transmission of the target node, configure the L1 cache of the target node and configure the data cache location corresponding to the target node as L1 cache. Please refer to the relevant description of step S201 above, and no further details will be given here.

[0063] Step S302: When the amount of data in the first-level cache reaches a preset back pressure threshold, the data cache location corresponding to the target node is switched to the second-level cache, and flow control is performed on the target node; wherein the second-level cache is configured as a shared cache for all nodes transmitting data to the local node.

[0064] Specifically, the above step S302 includes:

[0065] Step S3021: Acquire target network performance information of the target node and target transmission resource information of the local node.

[0066] Optionally, the target network performance information includes one or more of the traffic bandwidth of the target node and the state information of the target node, which can be adjusted according to actual conditions. The state information includes one or more of the load, voltage, and temperature of the target node, and may also include other state information of the target node, which is not limited here.

[0067] Optionally, the target transmission resource information includes one or more of the cache space capacity of the local node (including the sum of the cache capacity of the second-level cache, all first-level caches and other caches) and the bandwidth utilization of the local node, which can be adjusted according to actual conditions.

[0068] Step S3022, based on the target network performance information, target transmission resource information and target prediction model, obtain the target flow control information for the target node; wherein, the target prediction model is obtained by training the initial prediction model based on the associated historical network performance information, historical transmission resource information and historical flow control information; the target flow control information includes back pressure control information and secondary cache configuration information.

[0069] Optionally, the back pressure control information includes one or more of a target back pressure period, a number of periods corresponding to the target back pressure period, and a target back pressure level. The target back pressure period is used to indicate a period during which the target node inserts idle packets into the transmitted data stream, the number of periods is used to indicate the number of periods during which the target node inserts idle packets into the transmitted data stream, and the target back pressure level is used to indicate the number of idle packets inserted per period into the transmitted data stream by the target node.

[0070] Optionally, the configuration information of the second-level cache includes one or more of a valid time range for the second-level cache to provide cache services to the target node and a cache size allocated to the target node in the second-level cache, which can be adjusted according to actual conditions. The valid time range can be represented by the start time and the stop time of the second-level cache providing cache services to the target node.

[0071] Optionally, the historical network performance information includes one or more of the following: historical traffic bandwidth and historical status information of the first sample node, which can be adjusted based on actual circumstances. The historical status information includes one or more of historical load, historical voltage, and historical temperature, and may also include other historical status information of the sample node, without limitation. The first sample node may be the target node, another node in the switching network that has sent data to the local node, or a node in the switching network that has sent data to a specific node, without limitation.

[0072] Optionally, the historical transmission resource information includes one or more of the cache space capacity and historical bandwidth utilization of the second sample node, which can be adjusted based on actual conditions. The second sample node is the data recipient corresponding to the first sample node. For example, when the first sample node is the target node, the second sample node can be a local node or another data recipient of the target node in the switching network. Alternatively, the second sample node is the local node, and the first sample node is another node in the switching network that has sent data to the local node. Alternatively, the second sample node is any node in the switching network, and the first sample node is a node in the switching network that sends data to the second sample node.

[0073] Optionally, the historical flow control information includes historical back pressure control information of the second sample node for the first sample node and historical configuration information of the secondary cache of the second sample node. The historical back pressure control information includes one or more of the historical back pressure period, the number of historical back pressure periods and the historical back pressure level. The historical back pressure period is used to characterize the period in which the first sample node inserts idle data packets into the transmitted data stream, the number of historical back pressure periods is used to characterize the number of periods in which the first sample node inserts idle data packets into the transmitted data stream, and the historical back pressure level is used to indicate the number of idle data packets inserted by the first sample node into the transmitted data stream per period. The historical configuration information includes the effective time range in which the secondary cache of the second sample node provides cache services to the first sample node, and one or more of the cache sizes allocated to the first sample node in the secondary cache of the second sample node, which can be adjusted according to actual conditions.

[0074] It should be noted that historical traffic control information can be obtained by debugging the traffic control information of the second sample node relative to the first sample node during the process of data transmission from the first sample node to the second sample node, and by correlating historical network performance information, historical transmission resource information, and historical traffic control information based on the time of information collection. Furthermore, historical traffic control information can also be configured by relevant personnel based on the traffic characteristics reflected by the historical network performance information and historical transmission resource information. The method for obtaining historical traffic control information is not limited here.

[0075] It should be noted that the local node's target prediction model collects information from multiple sources, including the target node's corresponding traffic bandwidth, target node status information, local node cache capacity, and local node bandwidth utilization, calculated in real time. The target prediction model analyzes this information and maps the underlying relationships between the data. Based on the analysis results, the target prediction model dynamically adjusts backpressure control information, such as the target backpressure period and its corresponding number of periods, and dynamically adjusts L2 cache configuration information, such as the L2 cache's on / off time and the target node's allocated L2 cache size. When allocating the target node's L2 cache size, the model fully considers the traffic characteristics reflected by each target node's target network performance information and target transmission resource information. This ensures that the local node can process all the target node's cached data within the target backpressure period, avoiding data backlogs and loss, thereby achieving precise control and optimized management of network traffic.

[0076] Step S3023: Switch the data cache location corresponding to the target node to the secondary cache, and cache the data of the data stream sent by the target node in the secondary cache according to the configuration information.

[0077] Specifically, when the first-level cache corresponding to one or more target nodes is about to overflow, the target prediction model provides target flow control information for each target node, wherein the target network performance information of each target node is different, so the target flow control information is different.

[0078] Step S3024: Perform flow control on the target node based on the back pressure control information.

[0079] Specifically, back pressure control is performed on the target node based on the back pressure control information to control the flow of the target node.

[0080] The flow control method provided in this embodiment utilizes a target prediction model to determine backpressure control information and secondary cache configuration information for a target node based on the flow characteristics jointly reflected by target network performance information and target transmission resource information. When the data cache location corresponding to the target node is switched to the secondary cache, the data of the data stream sent by the target node is cached in the secondary cache according to the configuration information, and flow control is performed on the target node based on the backpressure control information. Therefore, it is possible to ensure that during the backpressure control process of the target node, the local node can complete the processing of the data cached by the target node in the local node, thus avoiding data backlog and loss.

[0081] In some optional implementations, the step S3021 of obtaining target network performance information of the target node includes:

[0082] Step a1, obtain the traffic bandwidth corresponding to the target node, and send a first indication message to the target node; wherein the first indication message is used to instruct the target node to periodically insert a first idle data packet during data transmission with the local node; the first idle data packet carries the status information of the target node.

[0083] Specifically, when the first-level cache of the local node reaches the preset back pressure threshold, the local node can send a first indication message to the target node in the form of a traffic packet. After receiving the first indication message, the target node will suspend normal data transmission and send its own status information (such as load, voltage, temperature, etc.) back to the local node through the first idle data packet.

[0084] In practical applications, an initial control information can be set in advance, wherein the initial control information includes an initial back pressure period and the number of periods thereof. When the target node connects to the local node or sends data for the first time, a first indication information is sent to the target node based on the initial control information, wherein the first indication information is used to instruct the target node to periodically insert a first idle data packet during data transmission with the local node according to the initial back pressure period and the number of periods thereof. In addition, the initial control information may also include an initial back pressure level. The target node is configured with at least one correspondence between a preset back pressure level and the number of idle data packets. The first indication information is used to instruct the target node to periodically insert a number of first idle data packets corresponding to the initial back pressure level during data transmission with the local node according to the initial back pressure period, the initial back pressure level, and the number of periods corresponding to the initial back pressure period.

[0085] In addition, if the communication connection between the local node and the target node is disconnected due to power failure or other reasons, when the local node and the target node re-establish a communication connection, the local node can obtain the network performance information of the target node based on the traffic bandwidth and historical status information of the target node, and obtain the initial control information for the target node based on the obtained network performance information, the transmission resource information of the local node (such as the cache space capacity and bandwidth utilization of the local node) and the target prediction model. The local node sends a first indication message to the target node based on the initial control information, so that the target node periodically inserts a first idle data packet during the data transmission process with the local node according to the initial control information. For specific details, please refer to the relevant description of the initial control information above, and no further details will be given here.

[0086] Step a2: receiving a first idle data packet sent by a target node, extracting the state information of the target node from the first idle data packet to obtain target network performance information; wherein the target network performance information includes traffic bandwidth and state information.

[0087] Specifically, after receiving the first idle data packet, the local node extracts the state information therein for use in the target prediction model to predict the flow control information without performing a buffering operation to improve information processing efficiency. That is, the local node is configured not to perform a buffering operation on the first idle data packet.

[0088] The flow control method provided in this embodiment is such that, when the amount of data in the target node's first-level cache reaches a preset back pressure threshold, the local node sends a first indication message to the target node, causing the target node to periodically insert a first idle data packet during data transmission. Since the local node does not cache the first idle data packet, the information processing efficiency of the local node can be improved. Furthermore, the first idle data packet carries the target node's status information. Therefore, upon receiving the first idle data packet, the local node can extract the target node's status information from it and, combined with the target node's traffic bandwidth calculated in real time, provide rich reference information for the target prediction model to predict target flow control information, thereby improving the accuracy of the target prediction model's prediction results.

[0089] In some optional implementations, when there are multiple target nodes, step a2 includes:

[0090] Step a21: Receive the first idle data packet sent by the target node, and extract the state information of the target node from the first idle data packet. Please refer to the relevant content of the above step a2, which will not be repeated here.

[0091] Step a22: Write the status information of the target node into a preset information table.

[0092] Specifically, both the local node and the target node are equipped with a preset information table with the same structure, which is used to store the status information of the target node.

[0093] Step a23: If the amount of data in the first-level cache corresponding to multiple target nodes reaches a preset back pressure threshold, the traffic heat of each target node is obtained.

[0094] Specifically, traffic heat is the ratio of the target node's traffic to the total bandwidth. In actual applications, the target node's data flow may not be sent for a period of time, and the target node's traffic will become zero, and the traffic heat will decrease.

[0095] Step a24, obtaining target state information from a preset information table based on traffic heat to obtain target network performance information; wherein the target network performance information includes the target state information and the traffic bandwidth of the target node corresponding to the target state information.

[0096] Specifically, the local node obtains the status information of the target node in order of traffic heat from high to low as the target status information. This method of obtaining target status information according to traffic heat allows the local node to preferentially obtain information that is crucial for back pressure control, thereby improving the pertinence and efficiency of information acquisition. Since a mapping relationship between status information and back pressure control information (such as target back pressure period) is established in the target prediction model, the local node can adjust the back pressure control information of the corresponding target node in a timely manner according to the target status information currently obtained, thereby avoiding the problem of reduced bandwidth utilization due to long-term back pressure. Through this management method, network traffic management can be made more flexible and efficient, thereby ensuring the stable operation of the switching network.

[0097] The flow control method provided in this embodiment obtains target state information from the state information of multiple target nodes stored in a preset information table based on the traffic popularity of the target node, and combines this information with the traffic bandwidth of the corresponding target node to obtain target network performance information. Therefore, the local node can prioritize using the target prediction model to predict the target flow control information of target nodes with high traffic popularity, and promptly adjust the flow control for the corresponding target node, avoiding the reduction of local node bandwidth utilization due to prolonged flow control, thereby ensuring the stable operation of the switching network.

[0098] In some optional embodiments, the configuration information includes a valid time range for the secondary cache to provide cache services to the target node. Caching the data of the data stream sent by the target node in the secondary cache according to the configuration information in step S3023 includes: caching the data of the data stream sent by the target node in the secondary cache within the valid time range.

[0099] Among them, the effective time range is represented by the opening time and closing time of the secondary cache to provide cache services for the target node. The opening time here refers to the use of the secondary cache for the target node's data. It can be understood that since the secondary cache is shared, it may be occupied by data from other nodes. Therefore, when the target node's data needs to open the secondary cache, the data processing needs to be completed within the allocated opening time and closing time. If not completed, the data will be discarded. Then, the target prediction model will record the data processing results of the target node this time, and will subsequently delay the closing time of the target node. In actual applications, users can also manually adjust the weight ratio of the target node in the target prediction model based on the importance of the target node.

[0100] The flow control method provided in this embodiment, when the amount of data in the first-level cache corresponding to the target node reaches a preset back pressure threshold, uses the target prediction model to give an effective time range for the second-level cache to provide cache services to the target node, and caches the data of the target node to the second-level cache within the effective time range. Therefore, it can prevent the target node from occupying the resources of the shared second-level cache for a long time, so as to centrally allocate the resources of the second-level cache and ensure the high efficiency of network data processing.

[0101] In some optional embodiments, the flow control method of the present disclosure further includes: discarding the data of the target node in the secondary cache when the valid time range is exceeded, and recording this event. Based on this event, the parameters of the target prediction model are updated to obtain an updated target prediction model. The updated target prediction model delays the shutdown time of the target node when predicting the valid time range within which the secondary cache provides cache services for the target node.

[0102] In some optional implementations, the configuration information further includes a cache size allocated to the target node in the secondary cache. The above-mentioned caching of the data stream sent by the target node into the secondary cache within the effective time range includes:

[0103] Step b1: Allocate a corresponding cache area for the target node in the secondary cache according to the cache size.

[0104] Step b2: caching the data of the data stream sent by the target node into the cache area within the valid time range.

[0105] The flow control method provided in this embodiment is based on the shared secondary cache and different flow characteristics of each target node. Therefore, when the amount of data in the primary cache corresponding to the target node reaches a preset back pressure threshold, the target prediction model provides the corresponding cache size allocated to each target node in the secondary cache, thereby ensuring that the local node can complete the processing of the target node's cache data during the target node's back pressure control process.

[0106] In some optional embodiments, the backpressure control information includes a target backpressure period and a corresponding number of periods. Step S3024 includes: sending second indication information to the target node based on the backpressure control information to perform flow control on the target node; wherein the second indication information is used to instruct the target node to periodically insert a second idle data packet during data transmission with the local node based on the target backpressure period and the number of periods.

[0107] It should be noted that the second idle data packet carries the target node's state information. Upon receiving the second idle data packet, the local node extracts this state information for use in the target prediction model to predict the next round of flow control information. This information is not cached, improving information processing efficiency. In other words, the local node does not cache the second idle data packet.

[0108] The flow control method provided in this embodiment uses a target prediction model to provide a target back pressure period and its corresponding number of periods when the amount of data in the first-level cache corresponding to the target node reaches a preset back pressure threshold. Therefore, the back pressure period and number of periods of each target node can be dynamically adjusted according to the traffic characteristics of each target node, thereby avoiding a reduction in bandwidth utilization due to long-term back pressure and improving network performance.

[0109] In some optional embodiments, the back pressure control information also includes a target back pressure level; the target node is configured with a correspondence between at least one preset back pressure level and the number of idle data packets; the preset back pressure level is positively correlated with the number of idle data packets; the second indication information is used to instruct the target node to periodically insert a number of second idle data packets corresponding to the target back pressure level during data transmission with the local node based on the target back pressure period, the number of periods and the target back pressure level.

[0110] It should be noted that the number of second idle data packets sent in each target back pressure cycle will increase as the back pressure level increases. In practical applications, the target prediction model can predict the increasing number of idle data packets for each preset back pressure level based on the mapping relationship between network performance information, transmission resource information and preset back pressure levels, so as to obtain a correspondence between at least one preset back pressure level and the number of idle data packets. Then, the correspondence is sent to the target node for storage, so that when the target node receives the target back pressure level, it determines the number of second idle data packets inserted in each target back pressure cycle corresponding to the target back pressure level based on the correspondence. In addition, the correspondence between at least one preset back pressure level and the number of idle data packets can also be dynamically adjusted according to actual conditions.

[0111] Not only that, the local node is also configured with at least one preset correspondence between the pressure sharing level and the number of idle data packets. After obtaining the target back pressure level, the local node can determine the number of second idle data packets inserted in each target back pressure period corresponding to the target back pressure level based on the target back pressure level and the locally stored correspondence, and send the number to the target node so that the target node periodically inserts the number of second idle data packets corresponding to the target back pressure level into the data stream sent.

[0112] It should be noted that the lowest preset back pressure level causes the destination node to only periodically insert the second idle packet into the transmitted data stream (or data flow), which has little impact on normal data transmission. The highest preset back pressure level requires the remote node to directly stop sending the data stream and only send the second idle packet to quickly relieve the pressure on the local node.

[0113] It can be understood that this embodiment can control the data sending of the target node in a hierarchical manner through the target back pressure period, number of periods, target back pressure level and status information of the requested target node, and can achieve from lightly inserting the second idle data packet to completely stopping the sending of the data stream (that is, only sending the second idle data packet). The target node returns the status information of the target node according to the target back pressure level, and the local node optimizes the flow control again accordingly, accurately adjusts the flow, and prevents congestion.

[0114] The flow control method provided in this embodiment, when the amount of data in the first-level cache corresponding to the target node reaches a preset back pressure threshold, the target prediction model gives a target back pressure period, the number of periods and the target back pressure level, and hierarchically controls the number of second idle data packets sent by the target node in each target back pressure period. Therefore, it is possible to accurately adjust the flow of each target node according to its traffic characteristics to prevent congestion.

[0115] In summary, the flow control method disclosed herein has the following main features:

[0116] First, a two-level cache collaborative architecture: The disclosed traffic control method utilizes a collaborative architecture that integrates a primary cache and a shared secondary cache. The primary cache independently processes data for each target node, avoiding interference. The secondary cache activates when the amount of data in the primary cache reaches a preset backpressure threshold, storing overflow data. This improves cache utilization on local nodes and reduces data loss and latency.

[0117] Second, hierarchical pressure distribution and information exchange: Data transmission to the target node is controlled according to the target pressure level predicted by the target prediction model. This can range from lightly inserting idle packets to completely stopping data transmission to the target node. The target node transmits status information based on the target pressure level. The local node further optimizes flow control based on this returned status information, precisely adjusting traffic flow and preventing congestion.

[0118] Third, the target prediction module of multi-source data: The target prediction model integrates multi-source data such as the traffic bandwidth of the target node, the status information of the target node, the cache space capacity of the local node and the bandwidth utilization of the local node. After analysis, it dynamically adjusts the target back pressure cycle, the effective time range and cache size of the secondary cache to provide cache services for the target node, so as to accurately predict the traffic of the target node and realize intelligent traffic management.

[0119] Fourth, intelligent management of status information: The local node and the target node set up a preset information table with the same structure. The local node obtains the status information of the target node based on the traffic heat, so that the target prediction model can establish a mapping relationship between the status information and the back pressure cycle, so that the local node can adjust the back pressure strategy in time to avoid reducing bandwidth utilization due to long-term back pressure, thereby improving network performance.

[0120] It can be understood that the flow control method disclosed in the present invention effectively reduces data loss and delay through the collaborative work of the independent first-level cache and the shared second-level cache in the two-level cache collaborative architecture. Through hierarchical back pressure control and state information interaction, the flow is accurately adjusted to avoid congestion. With the help of the target prediction model of multi-source data, the target prediction model dynamically optimizes the back pressure and cache strategy to achieve intelligent control of the flow. Through intelligent state information management, state information is obtained and a mapping relationship is established based on the flow heat, which avoids bandwidth waste while ensuring the flow control effect. The flow control method disclosed in the present invention significantly improves the utilization, stability and transmission efficiency of network resources, and provides strong support for the stable and efficient operation of the switching network.

[0121] In this embodiment, a flow control device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments. Details already described will not be repeated here. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0122] This embodiment provides a flow control device, such as Figure 4 Shown, including:

[0123] The first processing module 401 is configured to configure the first-level cache of the target node in response to data transmission of the target node, and configure the data cache location corresponding to the target node as the first-level cache;

[0124] The second processing module 402 is used to switch the data cache location corresponding to the target node to the second-level cache and perform flow control on the target node when the amount of data in the first-level cache reaches a preset back pressure threshold; wherein the second-level cache is configured as a shared cache for all nodes transmitting data to the local node.

[0125] In some optional implementations, the second processing module 402 includes:

[0126] An information acquisition unit, configured to acquire target network performance information of a target node and target transmission resource information of a local node;

[0127] a model prediction unit, configured to obtain target flow control information for a target node based on target network performance information, target transmission resource information, and a target prediction model; wherein the target prediction model is obtained by training an initial prediction model based on associated historical network performance information, historical transmission resource information, and historical flow control information; and the target flow control information includes back pressure control information and L2 cache configuration information;

[0128] A cache management unit, configured to switch the data cache location corresponding to the target node to the secondary cache, and cache the data of the data stream sent by the target node in the secondary cache according to the configuration information;

[0129] The flow control unit is used to perform flow control on the target node based on the back pressure control information.

[0130] In some optional implementations, the information acquisition unit includes:

[0131] a first acquisition subunit, configured to acquire a traffic bandwidth corresponding to a target node and send first indication information to the target node; wherein the first indication information is configured to instruct the target node to periodically insert a first idle data packet during data transmission with the local node; the first idle data packet carries status information of the target node;

[0132] The second acquisition subunit is used to receive a first idle data packet sent by the target node, and extract the status information of the target node from the first idle data packet to obtain target network performance information; wherein the target network performance information includes traffic bandwidth and status information.

[0133] In some optional embodiments, the second acquisition sub-unit is specifically used to: receive a first idle data packet sent by the target node, extract the status information of the target node from the first idle data packet; write the status information of the target node into a preset information table; in the case of multiple target nodes, if the amount of data in the first-level cache corresponding to multiple target nodes reaches a preset back pressure threshold, then obtain the traffic heat of each target node; obtain the target status information from the preset information table based on the traffic heat to obtain the target network performance information; wherein, the target network performance information includes the target status information and the traffic bandwidth of the target node corresponding to the target status information.

[0134] In some optional implementations, the configuration information includes a valid time range for the secondary cache to provide cache services to the target node. The cache management unit includes:

[0135] The data cache subunit is used to cache the data of the data stream sent by the target node into the secondary cache within a valid time range.

[0136] In some optional embodiments, the configuration information further includes a cache size allocated to the target node in the secondary cache. The data cache subunit is specifically configured to: allocate a corresponding cache area in the secondary cache for the target node based on the cache size; and cache data of the data stream sent by the target node in the cache area within a valid time range.

[0137] In some optional implementations, the back pressure control information includes a target back pressure period and a corresponding number of periods. The flow control unit includes:

[0138] A flow control subunit is used to send a second indication information to the target node based on the back pressure control information to perform flow control on the target node; wherein the second indication information is used to instruct the target node to periodically insert a second idle data packet during data transmission with the local node according to the target back pressure period and the number of periods.

[0139] In some optional embodiments, the back pressure control information also includes a target back pressure level; the target node is configured with a correspondence between at least one preset back pressure level and the number of idle data packets; the preset back pressure level is positively correlated with the number of idle data packets; the second indication information is used to instruct the target node to periodically insert a number of second idle data packets corresponding to the target back pressure level during data transmission with the local node based on the target back pressure period, the number of periods and the target back pressure level.

[0140] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0141] The flow control device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0142] The present disclosure also provides an electronic device. Figure 5 , Figure 5 This is a structural block diagram of an electronic device provided by an optional embodiment of the present disclosure, such as Figure 5As shown, the electronic device includes: one or more processors 501, a memory 502, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. The various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the electronic device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 501 is taken as an example.

[0143] Processor 501 may be a central processing unit, a network processor, or a combination thereof. Processor 501 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0144] The memory 502 stores instructions that can be executed by at least one processor 501, so as to enable the at least one processor 501 to implement the flow control method shown in the above embodiment.

[0145] The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device, etc. In addition, the memory 502 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 502 may optionally include a memory remotely located relative to the processor 501, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0146] The memory 502 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid state drive; the memory 502 may also include a combination of the above types of memory.

[0147] The electronic device further includes a communication interface 503 for the electronic device to communicate with other devices or a communication network.

[0148] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0149] A portion of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present disclosure through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes but is not limited to a source file, an executable file, an installation package file, etc. Accordingly, the way in which the computer program instruction is executed by the computer includes but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium that can be accessed by the computer.

[0150] Although the embodiments of the present disclosure have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A flow control method, characterized in that: Applicable to a node in a switching network, the method includes: In response to data transmission from a target node, configuring a first-level cache of the target node, and configuring a data cache location corresponding to the target node as the first-level cache; When the amount of data in the first-level cache reaches a preset back pressure threshold, the data cache location corresponding to the target node is switched to the second-level cache, and flow control is performed on the target node; wherein the second-level cache is configured as a shared cache for all nodes transmitting data to the local node.

2. The flow control method according to claim 1, characterized in that: The step of switching the data cache location corresponding to the target node to the secondary cache and performing flow control on the target node includes: Acquire target network performance information of the target node and target transmission resource information of the local node; Based on the target network performance information, the target transmission resource information, and a target prediction model, target flow control information for the target node is obtained; wherein the target prediction model is obtained by training an initial prediction model based on associated historical network performance information, historical transmission resource information, and historical flow control information; the target flow control information includes back pressure control information and configuration information of the secondary cache; Switching the data cache location corresponding to the target node to the secondary cache, and caching the data of the data stream sent by the target node in the secondary cache according to the configuration information; Flow control is performed on the target node based on the back pressure control information.

3. The flow control method according to claim 2, characterized in that: The acquiring target network performance information of the target node includes: Obtaining a traffic bandwidth corresponding to the target node and sending first indication information to the target node; wherein the first indication information is used to instruct the target node to periodically insert a first idle data packet during data transmission with the local node; the first idle data packet carries status information of the target node; Receive the first idle data packet sent by the target node, and extract the status information of the target node from the first idle data packet to obtain the target network performance information; wherein the target network performance information includes the traffic bandwidth and the status information.

4. The flow control method according to claim 3, characterized in that: In the case that there are multiple target nodes, receiving the first idle data packet sent by the target node, and extracting the state information of the target node from the first idle data packet to obtain the target network performance information includes: receiving the first idle data packet sent by the target node, and extracting the status information of the target node from the first idle data packet; Writing the status information of the target node into a preset information table; If the amount of data in the first-level cache corresponding to multiple target nodes reaches the preset back pressure threshold, the traffic heat of each target node is obtained; Target state information is obtained from the preset information table based on the traffic heat to obtain the target network performance information; wherein the target network performance information includes the target state information and the traffic bandwidth of the target node corresponding to the target state information.

5. The flow control method according to claim 2, characterized in that: The configuration information includes a valid time range for the secondary cache to provide a cache service for the target node; and caching data of the data stream sent by the target node in the secondary cache according to the configuration information includes: Within the valid time range, the data of the data stream sent by the target node is cached in the secondary cache.

6. The flow control method according to claim 5, characterized in that: The configuration information also includes a cache size allocated to the target node in the secondary cache; The step of caching the data of the data stream sent by the target node into the secondary cache within the valid time range includes: Allocate a corresponding cache area for the target node in the secondary cache according to the cache size; Within the valid time range, the data of the data stream sent by the target node is cached in the cache area.

7. The flow control method according to claim 2, characterized in that: The back pressure control information includes a target back pressure period and a corresponding number of periods; and the performing flow control on the target node based on the back pressure control information includes: Based on the back pressure control information, second indication information is sent to the target node to perform flow control on the target node; wherein the second indication information is used to instruct the target node to periodically insert a second idle data packet during data transmission with the local node according to the target back pressure period and the number of periods.

8. The flow control method according to claim 7, characterized in that: The back pressure control information also includes a target back pressure level; the target node is configured with a correspondence between at least one preset back pressure level and the number of idle data packets; the preset back pressure level is positively correlated with the number of idle data packets; the second indication information is used to instruct the target node to periodically insert a number of second idle data packets corresponding to the target back pressure level during data transmission with the local node based on the target back pressure period, the number of periods and the target back pressure level.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the flow control method according to any one of claims 1 to 8 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the flow control method according to any one of claims 1 to 8.

Images