Network card drive optimization method and system for dynamic adaptation of interrupt number and queue number

Abstract

The invention relates to a network card drive optimization method and system for dynamic adaptation of an interrupt number and a queue number, and the method comprises the following steps: 1, after a Linux system operates, a script is used as a service to automatically operate, and the state of a network card is monitored in real time; 2, the script detects whether the interrupt number of the network card after the driver is loaded is abnormal or not; 3, setting the network card after the script detects an exception; 4, unloading the network card drive, counting the maximum queue number of all current network cards and the maximum interrupt number on the platform, calculating the corresponding relation between the queue number and the interrupt number of the network cards, and selecting the optimal configuration of the queue number and the interrupt number, so that the network cards can be normally used after the drive is loaded; according to the network card drive optimization method and system, the queue number and the interrupt number of the network card can be adjusted without manual intervention, and the error probability is reduced; the dynamic adaptation of the interrupt number and the queue number can ensure that the network card can be normally used after the driver is loaded, and the situation that the network card is used abnormally due to the fact that the queue number exceeds the interrupt number supported by the platform is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of computer networks, and more specifically, to a method and system for optimizing a network card driver with dynamic adaptation of interrupt numbers and queue numbers. Background Art

[0002] There are many network card models and brands in the market. Each type of network card has its own driver and supported number of queues. The number of queues for a single network port ranges from several to hundreds. In high-concurrency and high-load network environments, especially in the network security or server industries, a large number of network cards are used simultaneously, and the models are also diverse.

[0003] To optimize network performance, most network cards support a large number of queues by default after loading the driver. Each queue occupies an interrupt number, and the higher-end the network card, the more queues it supports. Different motherboard platforms support different numbers of interrupt numbers. When the number of queues of the network card exceeds the number of interrupts supported by the platform, it will cause abnormal use of the network card and cannot meet the usage requirements. For machines that use dozens or hundreds of network ports simultaneously, especially network cards with different brands and drivers, the supported number of queues is also different. Different motherboard platforms support different numbers of interrupt numbers. When the number of queues of the network card exceeds the number of interrupts supported by the platform, the network card will have no extra interrupts available, resulting in the network card being unusable or having to share an interrupt with other devices, resulting in the network card being unavailable or having a very low throughput and unable to meet the usage requirements. After problems occur, the speed of troubleshooting is slow, the process is complex, and after positioning the problem, it is necessary to repeatedly verify whether the settings meet the requirements, resulting in low efficiency.

[0004] The prior art usually relies on manual adjustment of the network card queue number and interrupt number allocation, which is cumbersome and error-prone. In view of the above problems, the present invention proposes a method and system for optimizing a network card driver with dynamic adaptation of interrupt numbers and queue numbers. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for optimizing a network card driver with dynamic adaptation of interrupt numbers and queue numbers in view of the above-mentioned defects of the prior art.

[0006] On the one hand, the technical solution adopted by the present invention to solve its technical problems is: a method for optimizing a network card driver with dynamic adaptation of interrupt numbers and queue numbers, which includes the following steps:

[0007] Step 1: After the Linux system runs, the script runs automatically as a service and monitors the network card status in real time.

[0008] Step 2: The script detects whether the interrupt number after the network card loads the driver is abnormal.

[0009] Step 3: After the script detects an anomaly, it configures the network card. The configuration scenarios include the following:

[0010] Configuration scenario 1: The script counts the number of interrupts supported by the platform.

[0011] Configuration scenario 2: The script counts the number of interrupts obtained for all network cards.

[0012] Configuration scenario 3: The script counts the maximum number of queues supported by the network card.

[0013] Step 4: After the script detects any one or more of the above anomaly scenarios, it uninstalls the network card driver, calculates the current number of interrupts on the platform and the number of queues of the network card. After the script obtains the data, it will count the maximum number of queues that all current network cards can support and the maximum number of interrupts that the platform can support. After the calculation is completed, the script starts to configure; calculates the correspondence between the network card queue number and the number of interrupts supported by the platform, and selects the best configuration of the queue number and the interrupt number so that the network card can be used normally after the driver is loaded.

[0014] In the network card driver optimization method of the present invention, in step 2, the script monitors the status of the network card after the driver is loaded, and the detected anomaly situations include the following:

[0015] Anomaly situation 1: The network card fails to obtain an interrupt number, resulting in the network card being unable to be used.

[0016] Anomaly situation 2: The multi-queue requirement of the network card cannot be met, and the interrupt number needs to be shared, resulting in packet loss or communication interruption.

[0017] In the network card driver optimization method of the present invention, the script has a logging function to record the change history of the network card configuration and system events for problem tracking and system maintenance.

[0018] In the network card driver optimization method of the present invention, the steps of the script for adjusting the queue configuration are as follows:

[0019] Determine the maximum number of interrupt numbers supported by the motherboard platform;

[0020] Calculate the number of queues for each network card according to the queue requirements of the network card and the interrupt number limit of the platform;

[0021] If the calculated number of queues exceeds the number of interrupts supported by the platform, reduce the number of queues according to a preset priority or strategy;

[0022] Apply the new queue configuration and monitor the network performance to ensure the effectiveness of the configuration;

[0023] If the new queue configuration causes a decline in network performance, readjust the queue configuration until the optimal configuration is found.

[0024] The network card driver optimization method according to the present invention, wherein calculating the correspondence between the network card queue number and the interrupt number includes:

[0025] Statistical total queue number requirements for all network cards in the current system, that is, the sum of the maximum queue numbers supported by all network cards;

[0026] Compare the total queue number requirements with the maximum interrupt number supported by the platform:

[0027] If the total queue number requirements are less than or equal to the maximum interrupt number supported by the platform, allocate the maximum queue number supported by each network card;

[0028] If the total queue number requirements are greater than the maximum interrupt number supported by the platform, allocate interrupt numbers according to priority or weight to ensure that high-priority network cards or critical network cards obtain sufficient queue numbers while reducing the queue numbers of low-priority network cards.

[0029] The network card driver optimization method according to the present invention, wherein the allocation rules for the priority or weight include several types:

[0030] Determine the priority according to the use of the network card;

[0031] Determine the weight according to the brand and model of the network card, and give high-performance network cards priority to allocate more queue numbers;

[0032] Determine the priority and weight according to user-defined configuration.

[0033] The network card driver optimization method according to the present invention, wherein calculating the correspondence between the network card queue number and the interrupt number further includes the following steps:

[0034] When the total queue number requirements are greater than the maximum interrupt number supported by the platform, adopt dynamic adjustment steps, specifically including:

[0035] Reduce the queue numbers of non-critical network cards to ensure that critical network cards obtain sufficient interrupt numbers;

[0036] For network cards that support multiple queues, dynamically adjust their queue numbers according to actual needs to avoid excessive occupation of interrupt numbers;

[0037] For low-priority network cards, adopt the method of sharing interrupt numbers to reduce interrupt number occupation.

[0038] The network card driver optimization method according to the present invention, wherein in step four, the script includes the following two setting options:

[0039] The first setting: Set the network card queue for the specified chip;

[0040] The second setting: Recalculate and set the network card queues for all current network cards.

[0041] In the network card driver optimization method of the present invention, various network card queue setting methods are set in the script for multiple network cards to use simultaneously;

[0042] On the other hand, the present invention also provides a network card driver optimization system for dynamically adapting the interrupt number and the number of queues, which includes the following modules:

[0043] Script monitoring module: used to run the script under the Linux system and monitor the network card status in real time;

[0044] Abnormal detection module: the script detects whether the interrupt number is abnormal after the network card loads the driver; after the script detects an abnormality, it sets the network card; the setting situations include the following:

[0045] Setting situation 1: the script counts the number of interrupts supported by the platform;

[0046] Setting situation 2: the script counts the number of interrupts obtained by all network cards;

[0047] Setting situation 3: the script counts the maximum number of queues supported by the network card;

[0048] Queue setting module: used to unload and reload the network card driver and set the number of network card queues;

[0049] Data statistics module: calculates the current number of interrupts of the platform and the number of queues of the network card. After the script obtains the data, it will count the maximum number of queues supported by all current network cards and the maximum number of interrupts supported by the platform. After the calculation is completed, the script starts to set; calculates the corresponding relationship between the number of network card queues and the number of interrupts supported by the platform, and selects the best configuration of the number of queues and interrupts.

[0050] The beneficial effects of the present invention are as follows: The network card driver optimization method and system are ingeniously designed. Through the automatic operation and monitoring of the script, the number of network card queues and the interrupt number can be adjusted without manual intervention, reducing the complexity and error probability of manual operations; by optimizing the corresponding relationship between the network card queue and the interrupt number, dynamically adapting the interrupt number and the number of queues can ensure that the network card can be used normally after loading the driver, avoiding abnormal network card use caused by the number of queues exceeding the number of interrupts supported by the platform, ensuring that each network card can reasonably use the interrupt number, avoiding resource waste, improving the use efficiency of the interrupt number, and being applicable to complex usage environments. Description of the Drawings

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will further illustrate the present invention in conjunction with the drawings and embodiments. The drawings in the following description are only partial embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts:

[0052] Figure 1 is the flowchart of the network card driver optimization method for dynamic adaptation of interrupt numbers and queue numbers in the first embodiment of the present invention;

[0053] Figure 2 is the schematic diagram of the network card driver optimization system in the second embodiment of the present invention. Specific embodiments

[0054] The terms "first", "second", "third", "fourth", etc. in the description and claims of the present invention and the accompanying drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0055] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0056] "Plurality" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0057] Moreover, the terms indicating directions, such as "up, down, front, back, left, right, upper end, lower end, longitudinal", etc., are all referenced based on the posture position of the device or equipment described in this solution during normal use.

[0058] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0059] A network card driver optimization method for dynamic adaptation of interrupt numbers and queue numbers in a preferred embodiment of the present invention, as Figure 1As shown, it includes the following steps:

[0060] Step 1: After the Linux system runs, the script runs automatically as a service and monitors the network card status in real time; reducing manual intervention and improving the automation level, the system can automatically adjust and optimize the network card configuration without the need for administrator intervention.

[0061] Step 2: The script detects whether the network card has any abnormalities; in this embodiment, it mainly monitors and sets the network card hardware, network card driver, network card queue, network card interrupt, etc.; through real-time monitoring, the script can immediately detect any abnormalities that may occur after the network card loads the driver, and thus take actions quickly.

[0062] Among them, the script monitors the status of the network card after loading the driver, and the detected abnormal situations include the following:

[0063] Abnormal situation 1: The network card fails to obtain an interrupt number, resulting in the inability to use the network card. Detecting abnormal situation 1 can ensure that the network card can obtain the necessary interrupt number, prevent the network card from being unable to use, and thus avoid network service interruption.

[0064] Abnormal situation 2: The multi-queue requirement of the network card cannot be met, and the interrupt number needs to be shared, resulting in packet loss or communication interruption. Detecting abnormal situation 2 helps to ensure that the multi-queue requirement of the network card is met, avoid packet loss or communication interruption caused by sharing the interrupt number, and thus guarantee the stability of network communication and service quality.

[0065] Step 3: After the script detects an abnormality, it sets the network card, and the setting situations include the following:

[0066] Setting situation 1: The script counts the number of interrupts supported by the platform; knowing the number of interrupts supported by the platform can help the script ensure that the network card configuration does not exceed the hardware limitations, avoid configuration errors, and ensure compatibility.

[0067] Setting situation 2: The script counts the number of interrupts obtained by all network cards; knowing the number of interrupts currently used by each network card helps the script for resource management and optimization; through statistics, the script can balance the interrupt usage between different network cards, avoiding overloading of some network cards while other network cards have idle resources.

[0068] Setting situation 3: The script counts the maximum number of queues supported by the network card; knowing the maximum number of queues supported by the network card, the script can adjust the number of queues according to the actual network load to optimize performance.

[0069] Through the above three setting situations, the following benefits are obtained: providing the necessary information for the script to be able to intelligently adjust and optimize the network card configuration, thus ensuring the efficient and stable operation of the network.

[0070] Step 4: After the script detects any one or more of the above abnormal situations, it uninstalls the network card driver, calculates the current number of interrupts on the platform and the number of queues of the network card. After the script obtains the data, it will count the maximum number of queues that all current network cards can support and the maximum number of interrupts that the platform can support. After the calculation is completed, the script starts to set; calculate the correspondence between the network card queue number and the number of interrupts supported by the platform, and select the best configuration of the queue number and the number of interrupts, so that the network card can be used normally after loading the driver.

[0071] Optionally, the steps of the script for adjusting the queue configuration are as follows:

[0072] 1. Determine the maximum number of interrupt numbers supported by the motherboard platform;

[0073] This can avoid configuring a queue number that exceeds the hardware capabilities, ensuring hardware compatibility and stability.

[0074] 2. Calculate the number of queues for each network card according to the queue requirements of the network card and the interrupt number limit of the platform;

[0075] If the calculated number of queues exceeds the number of interrupts supported by the platform, the number of queues is reduced according to a preset priority or strategy; this can prevent interrupt number resource conflicts, ensure that the queue configuration of each network card is optimal, and thus improve the network processing ability.

[0076] 3. Apply the new queue configuration and monitor the network performance to ensure the effectiveness of the configuration;

[0077] If the new queue configuration causes a decline in network performance, readjust the queue configuration until the optimal configuration is found. Ensure that the network service is not affected by the configuration change.

[0078] This dynamic optimization process helps to find the optimal configuration and improve network performance.

[0079] Furthermore, calculating the correspondence between the network card queue number and the interrupt number includes:

[0080] Statistical total queue number requirements for all network cards in the current system, that is, the sum of the maximum number of queues supported by all network cards; compare the total queue number requirements with the maximum number of interrupts supported by the platform: this can ensure the most optimized allocation of interrupt number resources and avoid resource waste.

[0081] If the total queue number requirements are less than or equal to the maximum number of interrupts supported by the platform, allocate the maximum number of queues supported by each network card; this can maximize the network throughput.

[0082] If the total queue number requirements are greater than the maximum number of interrupts supported by the platform, allocate interrupt numbers according to the priority or weight to ensure that high-priority network cards or key network cards obtain sufficient queue numbers, while reducing the queue numbers of low-priority network cards to improve the overall network performance.

[0083] Furthermore, the rules for allocating the priorities or weights include several types:

[0084] Determine the priority according to the usage of the network card;

[0085] Determine the weight according to the brand and model of the network card, and preferentially allocate more queue numbers to high-performance network cards;

[0086] Determine the priority and weight according to the user-defined configuration.

[0087] The following benefits can be achieved through the above allocation rules:

[0088] 1. By ensuring that the resource requirements of critical services and high-performance network cards are met, the overall network performance and stability are improved.

[0089] 2. Allocate resources according to actual needs and hardware capabilities, avoiding resource waste.

[0090] 3. By defining clear rules, the complexity of network configuration and management is simplified.

[0091] 4. User-defined configuration allows network services to better meet the needs of specific users or services.

[0092] 5. By reasonably allocating resources, the possibility of network failures and performance bottlenecks is reduced.

[0093] Optionally, the steps for calculating the correspondence between the queue numbers and the interrupt numbers further include the following steps:

[0094] When the total queue number requirement is greater than the maximum number of interrupts supported by the platform, adopt dynamic adjustment steps, specifically including:

[0095] Reduce the queue numbers of non-critical network cards to ensure that critical network cards obtain sufficient interrupt numbers; preferentially allocate the limited interrupt number resources to critical services to improve the efficiency of the overall network service;

[0096] For network cards that support multiple queues, dynamically adjust their queue numbers according to actual needs to avoid excessive occupation of interrupt numbers; dynamically adjust the queue numbers according to the actual network load and requirements to make the network configuration more flexible and efficient.

[0097] For low-priority network cards, adopt the method of sharing interrupt numbers to reduce the occupation of interrupt numbers; save resources, and by sharing interrupt numbers, the total occupation of interrupt numbers is reduced, enabling more network cards to operate with limited resources.

[0098] Furthermore, in step four, the script includes the following two setting options:

[0099] The first setting: Set the network card queues of the specified chip;

[0100] The second setting: Recalculate and set all current network card queues.

[0101] This network card driver optimization method is ingeniously designed. Through the automatic operation and monitoring of scripts, the number of network card queues and interrupt numbers can be adjusted without manual intervention, reducing the complexity and error probability of manual operations; by optimizing the correspondence between network card queues and interrupt numbers, dynamically adapting the interrupt number and queue number can ensure that the network card can be used normally after loading the driver, avoiding abnormal use of the network card caused by the queue number exceeding the interrupt number supported by the platform, ensuring that each network card can reasonably use the interrupt number, avoiding resource waste, improving the utilization efficiency of the interrupt number, and being applicable to complex usage environments.

[0102] Furthermore, the script has a logging function to record the change history of network card configurations and system events for problem tracking and system maintenance.

[0103] Furthermore, multiple network card queue setting methods are set in the script for multiple network cards to use simultaneously; it supports single or simultaneous use of network cards of multiple brands, including intel, realtek, NetQin, Moochuang, etc., and has good compatibility.

[0104] Embodiment 2:

[0105] The present invention also provides a network card driver optimization system for dynamically adapting the interrupt number and queue number, as Figure 2 described, including the following modules:

[0106] Script monitoring module 10: Used to run the script under the Linux system and monitor the network card status in real time;

[0107] Abnormality detection module 20: The script detects whether the interrupt number is abnormal after the network card loads the driver; after detecting an abnormality, the script sets the network card; the setting situations include the following:

[0108] Situation 1: The script counts the number of interrupts supported by the platform;

[0109] Situation 2: The script counts the interrupt numbers obtained for all network cards;

[0110] Situation 3: The script counts the maximum number of queues supported by the network card;

[0111] Queue setting module 30: Used to unload and reload the network card driver and set the number of network card queues;

[0112] Data statistics module 40: Calculate the current number of interrupts on the platform and the number of queues of the network card. After the script obtains the data, it will count the maximum number of queues that all current network cards can support and the maximum number of interrupts that the platform can support. After the calculation is completed, the script starts to set; calculate the corresponding relationship between the network card queue number and the number of interrupts supported by the platform, and select the best configuration of the queue number and the interrupt number.

[0113] The network card driver optimization method and system are ingeniously designed. Through the automatic operation and monitoring of the script, the network card queue number and interrupt number can be adjusted without manual intervention, reducing the complexity and error probability of manual operations; by optimizing the corresponding relationship between the network card queue and the interrupt number, dynamically adapting the interrupt number and the queue number can ensure that the network card can be used normally after the driver is loaded, avoiding abnormal use of the network card caused by the queue number exceeding the number of interrupts supported by the platform, ensuring that each network card can reasonably use the interrupt number, avoiding resource waste, improving the use efficiency of the interrupt number, and being applicable to complex usage environments.

[0114] It should be understood that those of ordinary skill in the art can make improvements or changes according to the above description, and all such improvements and changes should fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for optimizing a network card driver with dynamic adaptation of interrupt numbers and queue numbers, characterized in that, It includes the following steps: Step 1: After the Linux system runs, the script runs automatically as a service and monitors the network card status in real time; Step 2: The script detects whether the interrupt number is abnormal after the network card loads the driver; Step 3: After the script detects an abnormality, it makes settings for the network card. The setting situations include the following: Setting situation 1: The script counts the number of interrupts supported by the platform; Setting situation 2: The script counts the number of interrupts obtained for all network cards; Setting situation 3: The script counts the maximum number of queues supported by the network card; Step 4: After the script detects any one or more of the above abnormal situations, it uninstalls the network card driver, calculates the current interrupt number of the platform and the queue number of the network card. After the script obtains the data, it will count the maximum number of queues that all current network cards can support and the maximum number of interrupts that the platform can support. After the calculation is completed, the script starts to set; calculates the corresponding relationship between the network card queue number and the interrupt number supported by the platform, selects the best configuration of the queue number and the interrupt number, so that the network card can be used normally after loading the driver.

2. The network card driver optimization method according to claim 1, wherein In step 2, the script monitors the status of the network card after loading the driver, and the detected abnormal situations include the following: Abnormal situation 1: The network card fails to obtain an interrupt number, resulting in the inability to use the network card; Abnormal situation 2: The multi-queue requirement of the network card cannot be met, and the interrupt number needs to be shared, resulting in packet loss or communication interruption.

3. The network card driver optimization method according to claim 1 or 2, characterized in that The script has a logging function to record the change history of network card configuration and system events for problem tracking and system maintenance.

4. The network card driver optimization method according to claim 3, wherein The steps for the script to adjust the queue configuration are as follows: Determine the maximum number of interrupt numbers supported by the motherboard platform; Calculate the queue number of each network card according to the queue requirement of the network card and the interrupt number limit of the platform; If the calculated queue number exceeds the interrupt number supported by the platform, reduce the queue number according to the preset priority or strategy; Apply the new queue configuration and monitor the network performance to ensure the effectiveness of the configuration; If the new queue configuration causes a decline in network performance, readjust the queue configuration until the optimal configuration is found.

5. The network card driver optimization method according to claim 1 or 4, characterized in that Calculating the corresponding relationship between the network card queue number and the interrupt number includes: Statistical total queue number requirements for all network cards in the current system, that is, the sum of the maximum queue numbers supported by all network cards; Compare the total queue number requirement with the maximum number of interrupts supported by the platform: If the total queue number requirement is less than or equal to the maximum number of interrupts supported by the platform, allocate the maximum queue number supported by each network card; If the total queue number requirement is greater than the maximum number of interrupts supported by the platform, allocate interrupt numbers according to the priority or weight to ensure that high-priority network cards or key network cards obtain sufficient queue numbers, while reducing the queue numbers of low-priority network cards.

6. The network card driver optimization method according to claim 5, characterized in that The allocation rules of the priority or weight include several types: Determine the priority according to the use of the network card; Determine the weight according to the brand and model of the network card, and high-performance network cards are preferentially allocated more queue numbers; Determine the priority and weight according to the user-defined configuration.

7. The network card driver optimization method according to claim 5, wherein Calculating the corresponding relationship between the network card queue number and the interrupt number also includes the following steps: When the total queue number requirement is greater than the maximum number of interrupts supported by the platform, adopt dynamic adjustment steps, specifically including: Reduce the queue number of non-critical network cards to ensure that critical network cards obtain sufficient interrupt numbers; For network cards that support multiple queues, dynamically adjust the number of queues according to actual needs to avoid excessive occupation of interrupt numbers; For low-priority network cards, use the method of sharing interrupt numbers to reduce the occupation of interrupt numbers.

8. The network card driver optimization method according to claim 1, characterized in that In step 4, the script includes the following two setting options: The first setting: Set the network card queues of the specified chip; The second setting: Recalculate and set all current network card queues.

9. The network card driver optimization method according to any one of claims 1-6, characterized in that Multiple network card queue setting methods are set in the script for multiple network cards to use simultaneously.

10. A network card driver optimization system with dynamic adaptation of interrupt numbers and queue numbers, characterized in that, It includes the following modules: Script monitoring module: Used to run the script under the Linux system and monitor the network card status in real time; Abnormality detection module: The script detects whether there is an abnormality in the interrupt number after the network card loads the driver; After the script detects an abnormality, it sets the network card; The setting situations include the following: Setting situation 1: The script counts the number of interrupts supported by the platform; Setting situation 2: The script counts the number of interrupts obtained for all network cards; Setting situation 3: The script counts the maximum number of queues supported by the network card; Queue setting module: Used to unload and reload the network card driver and set the number of network card queues; Data statistics module: Calculate the current interrupt number of the platform and the number of network card queues. After the script obtains the data, it will count the maximum number of queues that all current network cards can support and the maximum number of interrupts that the platform can support. After the calculation is completed, the script starts to set; Calculate the corresponding relationship between the network card queue number and the interrupt number supported by the platform, and select the best configuration of the queue number and interrupt number.

Images