A method for preventing high-speed network packet loss and dynamically extending a network card receiving end cache

Abstract

The application discloses a kind of network card receiving end cache dynamic extension method for preventing high-speed network packet loss, it is related to digital information transmission technical field, including the real-time monitoring situation of ring buffer based on network card driver and obtaining ring buffer real-time use information, whether the current cache of network card receiving end needs to be expanded is judged, if it needs to be expanded, then output current cache expansion signal;Whether there is ring buffer overflow situation is judged based on the history use information of ring buffer, and if there is, then output historical expansion early warning signal;User willingness is collected and user willingness data priority result is obtained by statistics, the use habit of different types of traffic is counted and use priority is judged to obtain history use habit priority result, and data type priority result is obtained in combination with user willingness data priority result and history use habit priority result.The application has the effect of improving the efficiency of network card receiving end cache for preventing high-speed network packet loss.

Description

Technical Field

[0001] This application relates to the field of digital information transmission technology, and in particular to a method for dynamically expanding the buffer of a network interface card (NIC) receiver to prevent packet loss in high-speed networks. Background Technology

[0002] Currently, high-speed networks generally refer to network environments with data transmission rates far exceeding traditional standards. These primarily include 10 Gigabit (10Gbps), 25G, 40G, 100G, and even 400G / 800G Ethernet, as well as high-performance computing networks such as InfiniBand. In high-speed network environments, packet loss at the network interface card (NIC) receiver is a common and challenging problem. Therefore, dynamically expanding the NIC receiver's buffer to prevent packet loss in high-speed networks is crucial.

[0003] In existing technologies, a circular buffer is a circular queue allocated by the network card driver in the host memory to temporarily store data packets written by the network card via DMA (Direct Memory Access). The CPU will later retrieve the data packets from this buffer for processing. However, in existing technologies, when there is a sudden surge in traffic or the CPU cannot process the data in time, the circular buffer is quickly filled up. Newly arriving data packets will be directly discarded by the network card because there is nowhere to store them, resulting in packet loss. This leads to low buffering efficiency at the network card receiver and has shortcomings. Summary of the Invention

[0004] To improve the efficiency of network card receiver buffering in preventing packet loss in high-speed networks, this application provides a method for dynamically expanding the network card receiver buffer in order to prevent packet loss in high-speed networks.

[0005] This application provides a method for dynamically expanding the receiver buffer of a network interface card (NIC) to prevent packet loss in high-speed networks, which adopts the following technical solution: A method for dynamically expanding the receiver buffer of a network interface card (NIC) to prevent packet loss in high-speed networks includes the following steps: Step S1: Based on the real-time monitoring of the ring buffer usage by the network card driver, obtain the real-time usage information of the ring buffer. Based on the real-time usage information of the ring buffer, determine whether the network card receiver buffer needs to be expanded. If expansion is required, output the current buffer expansion signal and record the current buffer expansion information. Step S2: Obtain historical usage information of the ring buffer; based on the historical usage information of the ring buffer, determine whether there is a ring buffer overflow during the historical reception process of the network card; if so, output a historical expansion warning signal and record the historical expansion warning data information. Step S3: Collect user preferences and statistically obtain user preference data priority results; statistically analyze user terminal usage habits for different types of traffic and determine the usage priority of different types of traffic to obtain historical usage habit priority results; combine user preference data priority results and historical usage habit priority results to obtain data type priority results. Step S4: Perform traffic prediction on the ring buffer and determine whether expansion is triggered in advance. If so, output the pre-prediction expansion signal. When the current buffer expansion signal and / or historical expansion warning data signal and / or pre-prediction expansion signal are received, apply for additional space in the system memory for caching and mark it as a secondary buffer. Based on the data type priority result, classify and cache the data information received by the network card receiver into the ring buffer and the secondary buffer, and record the data caching information and cache expansion information. Step S5: Send the data cache information and cache extension information to the blockchain system for storage.

[0006] Preferably, the real-time usage information of the ring buffer is obtained by real-time monitoring of the ring buffer based on the network card driver. The real-time usage information of the ring buffer includes buffer size information, current data volume information, and current read / write speed information. Based on the buffer size information and the current data volume information, the real-time utilization rate information of the circular buffer is obtained through calculation and processing. The real-time utilization rate information of the circular buffer is compared with the preset buffer utilization rate threshold to obtain the current buffer utilization rate comparison result. If the real-time utilization rate information of the circular buffer is greater than or equal to the preset buffer utilization rate threshold in the current buffer utilization rate comparison result, the current real-time expansion signal is output. Upon receiving the current real-time expansion signal, the difference between the buffer usage threshold and the current usage information of the circular buffer is calculated to obtain the current overflow information.

[0007] Preferably, the current read / write speed information is compared with a preset read / write speed threshold to obtain the current read / write speed comparison result. If the real-time utilization information of the circular buffer is less than the preset buffer utilization threshold in the current buffer utilization comparison result, and the current read / write speed information is lower than the preset read / write speed threshold in the current read / write speed comparison result, then the current expansion warning signal is output. Upon receiving the current extended warning signal, calculate the difference between the read / write speed threshold and the current read / write speed information to obtain the current read / write speed difference information. The current real-time extended signal and the current extended early warning signal are combined to form the current cached extended signal; The current overflow information and the current read / write speed difference information are combined to form the current cache extension information.

[0008] Preferably, the historical usage information of the circular buffer is obtained, wherein the historical usage information of the circular buffer includes historical data volume information and historical usage timestamp information; Based on the size information of the circular buffer and the historical data volume information, the historical usage rate information of the circular buffer is obtained by calculating the usage rate of the circular buffer at each historical moment. The historical usage information of the circular buffer is compared with the preset buffer usage threshold to obtain the historical buffer usage comparison result; Based on the historical buffer usage comparison results, it is determined whether there is a ring buffer overflow in the network card during the historical reception process. If the historical usage information of the ring buffer is greater than or equal to the preset buffer usage threshold, there is a ring buffer overflow, and a historical extension warning signal is output.

[0009] Preferably, upon receiving a historical expansion warning signal, the difference between the buffer usage rate threshold and the historical usage rate information of the circular buffer is calculated to obtain the historical usage overflow information. Based on historical usage timestamp information, the time information corresponding to the occurrence of circular buffer overflow is marked as historical overflow mark time information, and the duration of the overflow is recorded to obtain historical overflow duration information. The historical usage overflow information and the historical overflow duration information are combined to form historical extended early warning data information.

[0010] Preferably, a user preference data priority level questionnaire is sent to the user's terminal, and the user fills out the questionnaire and provides feedback. Based on the user preference data priority level questionnaire, users' preferences for caching priorities of different types of traffic are collected, and user preference data priority results are obtained; Historical traffic usage information is obtained by statistically analyzing the historical traffic usage of users. The historical traffic usage information includes historical traffic type information, historical traffic frequency information for each type of traffic, and historical traffic duration information for each type of traffic. Based on historical traffic frequency information, the priority of usage frequency for each type of traffic is determined to obtain the historical traffic frequency priority. Based on historical traffic duration information, the priority of usage duration for each type of traffic is determined to obtain historical traffic duration priority. By combining historical traffic frequency priority and historical traffic duration priority, the historical usage habit priority results for each type of traffic are determined; By combining the priority results of user willingness data and the priority results of historical usage habits, the priority of each type of traffic is determined to obtain the data type priority result.

[0011] Preferably, based on the historical usage information of the ring buffer, a time series model is used to predict the peak usage rate in real time within a set time period in the future to obtain the predicted peak information. The CPU soft interrupt utilization rate and NAPI polling cycle are collected simultaneously. When the predicted peak value is greater than or equal to the buffer utilization rate threshold, or when the CPU soft interrupt utilization rate is saturated, the ring buffer expansion is triggered in advance, and a pre-prediction expansion signal is output. Upon receiving the current cache expansion signal and / or the historical expansion warning data signal and / or the pre-prediction expansion signal, request additional space from the system memory for caching and mark it as a secondary cache area; Based on the current overflow information and / or the current read / write speed difference information and / or historical expansion warning data information, the size of the additional space for caching is determined to obtain the secondary cache size information; Based on the data type priority result, the data packets with higher priority among the data information received by the network card receiver are first put into the circular buffer. After the circular buffer overflows, they are then put into the secondary buffer in turn. Record data caching information, which includes cache data type information, data cache address information, and cache time information; The cache extension information is recorded, including cache extension address information, cache extension size information, and cache extension time information.

[0012] Preferably, the blockchain system is acquired and a signal connection link is established between the network card and the blockchain system; Retrieve data cache information and cache extension information and update them in real time; Data caching information and cache extension information are sent to the blockchain system for storage in real time.

[0013] In summary, this application includes at least one of the following beneficial technical effects: The network interface card (NIC) driver monitors the ring buffer usage in real time to obtain real-time ring buffer usage information and determines whether the NIC receiver buffer needs to be expanded. If expansion is needed, a current buffer expansion signal is output. Based on the historical ring buffer usage information, it is determined whether the NIC has experienced ring buffer overflow during historical reception. If so, a historical expansion warning signal is output. The need for NIC receiver buffer expansion is considered from multiple dimensions based on the current and historical buffers. According to user preferences and usage patterns, the priority of data types is analyzed to obtain data type priority results. When a current buffer expansion signal or a historical expansion warning data signal is received, additional space is requested from system memory for caching and marked as a secondary buffer. Based on the data type priority results, the data information received by the NIC receiver is classified and cached in the ring buffer and secondary buffer, temporarily accommodating data packets at risk of overflow. This reduces the risk of packet loss due to insufficient caching and improves the NIC receiver caching efficiency in preventing packet loss in high-speed networks. Attached Figure Description

[0014] Figure 1 This embodiment is a flowchart illustrating a method for dynamically expanding the network card receiver cache to prevent packet loss in high-speed networks. Detailed Implementation

[0015] The present application will be further described in detail below with reference to the accompanying drawings.

[0016] This application discloses a method for dynamically expanding the network card receiver cache to prevent packet loss in high-speed networks.

[0017] A method for dynamically expanding the receiver buffer of a network interface card (NIC) to prevent packet loss in high-speed networks includes the following steps: Reference Figure 1 Step S1 involves obtaining real-time ring buffer usage information based on real-time monitoring of the ring buffer by the network card driver. Based on this information, it is determined whether the network card receiver buffer needs expansion. If expansion is required, a current buffer expansion signal is output, and the current buffer expansion information is recorded. Step S1 specifically includes the following sub-steps: Step A1: Based on the network card driver, monitor the ring buffer usage in real time to obtain the ring buffer real-time usage information, which includes buffer size information, current data volume information, and current read / write speed information.

[0018] Step A2: Based on the buffer size information and the current data volume information, the real-time utilization information of the circular buffer is obtained through calculation and processing.

[0019] Specifically, the difference between the buffer size information and the current data volume information is used to calculate the amount of data read. The ratio of the amount of data read to the buffer size information is then calculated to obtain the real-time utilization information of the circular buffer.

[0020] Step A3: Compare the real-time utilization information of the circular buffer with the preset buffer utilization threshold to obtain the current buffer utilization comparison result. If the real-time utilization information of the circular buffer is greater than or equal to the preset buffer utilization threshold in the current buffer utilization comparison result, then output the current real-time expansion signal.

[0021] Step A4: After receiving the current real-time expansion signal, calculate the difference between the buffer usage rate threshold and the current usage rate information of the circular buffer to obtain the current usage overflow information.

[0022] Step S1 further includes the following sub-steps: Step B1: Compare the current read / write speed information with the preset read / write speed threshold to obtain the current read / write speed comparison result. If the real-time utilization information of the ring buffer is less than the preset buffer utilization threshold in the current buffer utilization comparison result, and the current read / write speed information is lower than the preset read / write speed threshold in the current read / write speed comparison result, then output the current expansion warning signal.

[0023] Step B2: After receiving the current extended warning signal, calculate the difference between the read / write speed threshold and the current read / write speed information to obtain the current read / write speed difference information.

[0024] Step B3: The current real-time extended signal and the current extended warning signal are combined to form the current cached extended signal.

[0025] Step B4: The current overflow information and the current read / write speed difference information are combined to form the current cache extension information.

[0026] Reference Figure 1 Step S2 involves obtaining historical usage information of the ring buffer. Based on this information, it is determined whether the network card experienced a ring buffer overflow during historical reception. If so, a historical expansion warning signal is output, and the historical expansion warning data is recorded. Step S2 specifically includes the following sub-steps: Step C1: Obtain historical usage information of the circular buffer, which includes historical data volume information and historical usage timestamp information.

[0027] Step C2: Based on the size information and historical data volume information of the circular buffer, calculate the historical usage rate of the circular buffer at each historical moment to obtain the historical usage rate information of the circular buffer. Specifically, calculate the difference between the buffer size information and the historical data volume information to obtain the historical data read information, and calculate the ratio of the historical data read information to the buffer size information to obtain the historical usage rate information of the circular buffer at each moment.

[0028] Step C3: Compare the historical usage information of the circular buffer with the preset buffer usage threshold to obtain the historical buffer usage comparison result.

[0029] Step C4: Based on the historical buffer usage comparison results, determine whether there is a ring buffer overflow in the network card during the historical reception process. If the historical usage information of the ring buffer is greater than or equal to the preset buffer usage threshold, then there is a ring buffer overflow, and a historical extension warning signal is output.

[0030] Step S2 further includes the following sub-steps: Step D1: After receiving the historical expansion warning signal, calculate the difference between the buffer usage rate threshold and the historical usage rate information of the circular buffer to obtain the historical usage overflow information.

[0031] Step D2: Based on historical usage timestamp information, mark the time information corresponding to the existence of a circular buffer overflow as historical overflow mark time information, and record the duration of the overflow to obtain historical overflow duration information.

[0032] Step D3: Historical usage overflow information and historical overflow duration information are combined to form historical extended early warning data information.

[0033] Reference Figure 1 Step S3 involves collecting user preferences and statistically analyzing the user preference data priority results. It also involves analyzing user usage habits for different types of traffic and determining the usage priority of different traffic types to obtain historical usage habit priority results. Finally, it combines the user preference data priority results and the historical usage habit priority results to obtain the data type priority results. Step S3 specifically includes the following sub-steps: Step S31: Send a user preference data priority survey questionnaire to the user's terminal. The user fills out the user preference data priority survey questionnaire and provides feedback.

[0034] Step S32: Collect users' preferences for caching priorities of different types of traffic based on the user preference data priority survey questionnaire, and obtain the user preference data priority results.

[0035] Step S33: Statistically analyze the historical traffic usage of the user terminal to obtain historical traffic usage information, which includes historical traffic type information, historical traffic frequency information for each type of traffic, and historical traffic duration information for each type of traffic.

[0036] Step S34: Based on historical traffic frequency information, determine the priority of usage frequency for each type of traffic to obtain historical traffic frequency priority. The higher the historical traffic frequency information, the higher the historical traffic frequency priority.

[0037] Step S35: Based on historical traffic duration information, determine the priority of usage duration for each type of traffic to obtain historical traffic duration priority. The longer the historical traffic duration information, the higher the historical traffic duration priority.

[0038] Step S36: Combine historical traffic frequency priority and historical traffic duration priority to determine the historical usage habit priority results for each type of traffic.

[0039] Step S37: Combine the priority results of user willingness data and the priority results of historical usage habits to determine the priority of each type of traffic and obtain the data type priority result.

[0040] Reference Figure 1 Step S4: Upon receiving a current cache expansion signal or a historical expansion warning data signal, additional space is requested from system memory for caching and marked as a secondary cache. Based on data type priority, the data information received by the network card receiver is classified and cached in the circular cache and the secondary cache, and the data caching information and cache expansion information are recorded. Step S4 specifically includes the following sub-steps: Step S41: Based on the historical usage information of the ring buffer, the peak usage rate within a set time period in the future is predicted in real time using a time-series model to obtain the predicted peak information. The CPU soft interrupt (softirq) occupancy rate and NAPI polling cycle are collected simultaneously. When the predicted peak is greater than or equal to the buffer usage rate threshold, or when the CPU soft interrupt occupancy rate is saturated, the ring buffer expansion is triggered in advance, and the pre-prediction expansion signal is output. It should be noted that the timing model in the embodiments of this application refers to LSTM / ARIMA / EWMA; In practical applications, the expansion detection of the ring buffer in this embodiment includes two cases: expansion after overflow and expansion before overflow. The decision dimension is upgraded from single-dimensional utilization rate to a three-dimensional perspective of utilization rate, historical prediction and CPU load prediction. At the same time, the CPU softirq linkage can distinguish between real traffic bursts and CPU not being able to receive them in time, which improves the detection accuracy of dynamic expansion of the network card receiver buffer. Step S42: Upon receiving the current cache expansion signal and / or the historical expansion warning data signal and / or the pre-prediction expansion signal, request additional space from the system memory for caching and mark it as a secondary cache area.

[0041] Step S43: Based on the current overflow information and / or the current read / write speed difference information and / or the historical extended warning data information, determine the size of the additional space to be cached to obtain the secondary cache size information.

[0042] Specifically, the larger the current overflow information, the larger the secondary cache size information is, and it is sufficient to cover the current overflow information; the lower the current read / write speed difference information, the larger the secondary cache size information is; and the larger the historical overflow information or the longer the historical overflow duration information, the larger the secondary cache size information is.

[0043] Step S44: Based on the data type priority result, the data packets with higher priority among the data information received by the network card receiver are placed into the circular buffer first. After the circular buffer overflows, they are then placed into the secondary buffer in sequence.

[0044] Step S45: Record the data cache information, which includes cache data type information, data cache address information, and cache time information.

[0045] Step S46: Record the cache extension information, which includes cache extension address information, cache extension size information, and cache extension time information.

[0046] Reference Figure 1 Step S5 involves sending the data cache information and cache extension information to the blockchain system for storage. Step S5 specifically includes the following sub-steps: Step S51: Obtain the blockchain system and establish a signal connection link between the network card and the blockchain system.

[0047] Step S52: Obtain data cache information and cache extension information and update them in real time.

[0048] Step S53: Send the data cache information and cache extension information to the blockchain system for storage in real time.

[0049] Those skilled in the art will understand that embodiments of the present invention can be provided as methods or systems. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0050] This invention is described with reference to flowchart illustrations and structural diagrams of methods and systems according to embodiments of the invention. It should be understood that the combination of each process and module in the flowchart and structural diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the process. Figure 1 One or more processes and structures Figure 1 A device for a function specified in one or more modules.

[0051] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and structures Figure 1 The function specified in one or more modules.

[0052] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and structures Figure 1 The steps of a specified function in one or more modules.

[0053] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0054] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dynamically expanding the receiver buffer of a network interface card (NIC) to prevent packet loss in high-speed networks, characterized in that, Includes the following steps: The ring buffer usage information is obtained by real-time monitoring of the ring buffer based on the network card driver. Based on the ring buffer usage information, it is determined whether the network card receiver buffer needs to be expanded. If expansion is required, the current buffer expansion signal is output and the current buffer expansion information is recorded. Obtain historical usage information of the ring buffer, and determine whether the network card has a ring buffer overflow during the historical reception process based on the historical usage information of the ring buffer. If it does, output a historical expansion warning signal and record the historical expansion warning data information. Collect user preferences and statistically analyze the user preference data priority results. Analyze user usage habits for different types of traffic and determine the usage priority of different types of traffic to obtain historical usage habit priority results. Combine the user preference data priority results and historical usage habit priority results to obtain data type priority results. Traffic prediction is performed on the ring buffer to determine whether expansion is triggered in advance. If so, a pre-prediction expansion signal is output. When the current buffer expansion signal and / or historical expansion warning data signal and / or pre-prediction expansion signal are received, additional space is requested from the system memory for caching and marked as a secondary buffer. Based on the data type priority result, the data information received by the network card receiver is classified and cached in the ring buffer and the secondary buffer, and the data caching information and cache expansion information are recorded. Data cache information and cache extension information are sent to the blockchain system for storage.

2. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 1, characterized in that, The step of obtaining real-time usage information of the ring buffer based on real-time monitoring of the ring buffer using the network card driver, determining whether the network card receiver buffer needs to be expanded based on the real-time usage information of the ring buffer, and outputting a current buffer expansion signal and recording the current buffer expansion information, specifically includes the following sub-steps: The real-time usage information of the ring buffer is obtained by monitoring the ring buffer usage in real time based on the network card driver. The real-time usage information of the ring buffer includes buffer size information, current data volume information, and current read / write speed information. Based on the buffer size information and the current data volume information, the real-time utilization rate information of the circular buffer is obtained through calculation and processing. The real-time utilization rate information of the circular buffer is compared with the preset buffer utilization rate threshold to obtain the current buffer utilization rate comparison result. If the real-time utilization rate information of the circular buffer is greater than or equal to the preset buffer utilization rate threshold in the current buffer utilization rate comparison result, the current real-time expansion signal is output. Upon receiving the current real-time expansion signal, the difference between the buffer usage threshold and the current usage information of the circular buffer is calculated to obtain the current overflow information.

3. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 2, characterized in that, The step of obtaining real-time usage information of the ring buffer based on real-time monitoring of the ring buffer using the network card driver, determining whether the network card receiver buffer needs to be expanded based on the real-time usage information of the ring buffer, and outputting a current buffer expansion signal and recording the current buffer expansion information, specifically includes the following sub-steps: The current read / write speed information is compared with the preset read / write speed threshold to obtain the current read / write speed comparison result. If the real-time utilization information of the circular buffer is less than the preset buffer utilization threshold in the current buffer utilization comparison result, and the current read / write speed information is lower than the preset read / write speed threshold in the current read / write speed comparison result, then the current expansion warning signal is output. Upon receiving the current extended warning signal, calculate the difference between the read / write speed threshold and the current read / write speed information to obtain the current read / write speed difference information. The current real-time extended signal and the current extended early warning signal are combined to form the current cached extended signal; The current overflow information and the current read / write speed difference information are combined to form the current cache extension information.

4. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 3, characterized in that, The step of obtaining historical usage information of the ring buffer, determining whether the network card experienced a ring buffer overflow during historical reception based on the historical usage information, and outputting a historical expansion warning signal and recording the historical expansion warning data information specifically includes the following sub-steps: Obtain historical usage information of the circular buffer, which includes historical data volume information and historical usage timestamp information; Based on the size information of the circular buffer and the historical data volume information, the historical usage rate information of the circular buffer is obtained by calculating the usage rate of the circular buffer at each historical moment. The historical usage information of the circular buffer is compared with the preset buffer usage threshold to obtain the historical buffer usage comparison result; Based on the historical buffer usage comparison results, it is determined whether there is a ring buffer overflow in the network card during the historical reception process. If the historical usage information of the ring buffer is greater than or equal to the preset buffer usage threshold, there is a ring buffer overflow, and a historical extension warning signal is output.

5. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 4, characterized in that, The step of obtaining historical usage information of the ring buffer, determining whether the network card experienced a ring buffer overflow during historical reception based on the historical usage information, and outputting a historical expansion warning signal and recording the historical expansion warning data information, further includes the following sub-steps: Upon receiving a historical expansion warning signal, the difference between the buffer usage rate threshold and the historical usage rate information of the circular buffer is calculated to obtain the historical usage overflow information. Based on historical usage timestamp information, the time information corresponding to the occurrence of circular buffer overflow is marked as historical overflow mark time information, and the duration of the overflow is recorded to obtain historical overflow duration information. The historical usage overflow information and the historical overflow duration information are combined to form historical extended early warning data information.

6. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 5, characterized in that, The steps of collecting user preferences and statistically analyzing user preference data priority results, statistically analyzing user usage habits for different types of traffic and determining the usage priority of different types of traffic to obtain historical usage habit priority results, and combining user preference data priority results and historical usage habit priority results to obtain data type priority results, specifically include the following sub-steps: Send a user preference data priority level questionnaire to the user's terminal, and the user fills out the questionnaire and provides feedback; Based on the user preference data priority level questionnaire, users' preferences for caching priorities of different types of traffic are collected, and user preference data priority results are obtained; Historical traffic usage information is obtained by statistically analyzing the historical traffic usage of users. The historical traffic usage information includes historical traffic type information, historical traffic frequency information for each type of traffic, and historical traffic duration information for each type of traffic. Based on historical traffic frequency information, the priority of usage frequency for each type of traffic is determined to obtain the historical traffic frequency priority. Based on historical traffic duration information, the priority of usage duration for each type of traffic is determined to obtain historical traffic duration priority. By combining historical traffic frequency priority and historical traffic duration priority, the historical usage habit priority results for each type of traffic are determined; By combining the priority results of user willingness data and the priority results of historical usage habits, the priority of each type of traffic is determined to obtain the data type priority result.

7. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 6, characterized in that, The steps of predicting traffic in the ring buffer, determining whether to trigger expansion in advance, and outputting a pre-prediction expansion signal if so, and requesting additional space from system memory for caching and marking it as a secondary buffer after receiving the current buffer expansion signal and / or historical expansion warning data signal and / or pre-prediction expansion signal, based on data type priority, classifying and caching the data information received by the network card receiver into the ring buffer and the secondary buffer, and recording the data caching information and buffer expansion information, specifically include the following sub-steps: Based on the historical usage information of the ring buffer, a time series model is used to predict the peak usage rate in real time within a set time period in the future to obtain the predicted peak information. The CPU soft interrupt utilization rate and NAPI polling cycle are collected simultaneously. When the predicted peak value is greater than or equal to the buffer utilization rate threshold, or when the CPU soft interrupt utilization rate is saturated, the ring buffer expansion is triggered in advance, and a pre-prediction expansion signal is output. Upon receiving the current cache expansion signal and / or the historical expansion warning data signal and / or the pre-prediction expansion signal, request additional space from the system memory for caching and mark it as a secondary cache area; Based on the current overflow information and / or the current read / write speed difference information and / or historical expansion warning data information, the size of the additional space for caching is determined to obtain the secondary cache size information; Based on the data type priority result, the data packets with higher priority among the data information received by the network card receiver are first put into the circular buffer. After the circular buffer overflows, they are then put into the secondary buffer in turn. Record data caching information, which includes cache data type information, data cache address information, and cache time information; The cache extension information is recorded, including cache extension address information, cache extension size information, and cache extension time information.

8. The method for dynamically expanding the network card receiver buffer to prevent packet loss in high-speed networks according to claim 7, characterized in that, The step of sending the data cache information and cache extension information to the blockchain system for storage specifically includes the following sub-steps: Acquire the blockchain system and establish a signal connection link between the network card and the blockchain system; Retrieve data cache information and cache extension information and update them in real time; Data caching information and cache extension information are sent to the blockchain system for storage in real time.

Images