System parameter adjusting method

Abstract

The invention discloses a system parameter adjustment method. The method comprises the following steps: screening to-be-adjusted system parameters according to a preset strategy and sorting the system parameters; according to the default values of the system parameters, adjusting ranges corresponding to the system parameters are generated, and adjusting strategies corresponding to the types of the system parameters are obtained; selecting system parameters one by one according to the sorting result, and performing adjustment and speed measurement according to an adjustment range and an adjustment strategy corresponding to the system parameters so as to obtain an optimal adjustment value corresponding to the optimal speed measurement result; performing adjustment and speed measurement on the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter; and after the optimal adjustment values of the system parameters are obtained, combining the optimal adjustment values corresponding to the system parameters into a system parameter optimal adjustment set. By applying the technical scheme of the invention, each system parameter in the speed measurement system can be reasonably and efficiently adjusted, the optimal performance of the test system is realized, and the optimal speed measurement result is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of speed measurement systems, and in particular to a method for adjusting system parameters. The present invention also relates to a speed measurement system with adjustable parameters. Background Art

[0002] In the development of modern science and technology, speed measurement systems are widely used in many fields, including but not limited to traffic management, industrial manufacturing and scientific research. Due to the complexity of the network environment, the speed measurement system is an important tool for evaluating network performance, and its performance is directly related to its effectiveness and reliability in various application scenarios.

[0003] During the R&D and testing phase of network equipment, parameters can only be optimized and adjusted for the LAN or existing WAN environment, but the network environments in user scenarios vary greatly, resulting in the speed test system under preset parameters not being able to perform at its optimal performance. In order to achieve the network speed test indicators expected by users, engineers need to be frequently dispatched to the site to adjust system parameters for different users, which adds a lot of unnecessary debugging complexity.

[0004] Therefore, how to reasonably adjust system parameters in different network environments to achieve optimal performance has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0005] The present invention provides a system parameter adjustment method, which screens and orderly adjusts various system parameters in a speed measurement system to bring into play the best performance of the test system, thereby obtaining the best speed measurement result.

[0006] In order to achieve the above object, the present invention provides a method for adjusting system parameters, which comprises:

[0007] Filter and sort the system parameters to be adjusted according to the preset strategy;

[0008] generating an adjustment range corresponding to each of the system parameters according to the default values ​​of the system parameters, and acquiring an adjustment strategy corresponding to the type of each of the system parameters;

[0009] Selecting the system parameters one by one according to the ranking results, and adjusting and measuring the speed according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result;

[0010] Adjusting the next system parameter based on the optimal adjustment value and measuring the speed to obtain the optimal adjustment value corresponding to the next system parameter;

[0011] After obtaining the optimal adjustment values of each of the system parameters, merge the optimal adjustment values corresponding to each of the system parameters into an optimal adjustment set of system parameters.

[0012] Preferably, screen the system parameters to be adjusted according to a preset strategy and sort them, specifically:

[0013] Analyze the architecture of the system to obtain the influence of each of the system parameters on the performance of the system, and screen and sort the system parameters in descending order of influence;

[0014] Or, conduct a performance benchmark test on the system, and screen and sort the system parameters according to the results of the performance benchmark test.

[0015] Preferably, obtain an adjustment strategy corresponding to the type of each of the system parameters, specifically:

[0016] The type of system parameter corresponding to the default value being a boolean value is a boolean type, and the adjustment strategy is to sequentially adjust the system parameter based on the adjustment range;

[0017] The type of system parameter corresponding to the default value being a specific numerical value is a numerical type, and the adjustment strategy is to sequentially adjust the system parameter according to the intervals after dividing the adjustment range into multiple intervals based on a preset number of adjustment times.

[0018] Preferably, divide the adjustment range into multiple intervals based on a preset number of adjustment times, specifically:

[0019] If the adjustment range is divisible by the preset number of adjustment times, set the range sizes of the multiple intervals to be the same in an evenly divided manner, and use the mid-value of the range of each interval or an adjustable value close to the mid-value as an adjustment parameter;

[0020] If the adjustment range is not divisible by the preset number of adjustment times, determine the ranges of the multiple intervals based on the rounding rule, so that the range sizes of the remaining intervals except the last one are the same, and use the mid-value of the range of each interval or an adjustable value close to the mid-value as an adjustment parameter.

[0021] Preferably, select the system parameters one by one according to the sorting result, and adjust and measure the speed according to the adjustment range and adjustment strategy corresponding to the system parameter to obtain the optimal adjustment value corresponding to the optimal speed measurement result, specifically:

[0022] Obtain a data difference based on the speed measurement results of each time;

[0023] If the data difference exceeds a preset threshold, select the adjustment value with the highest speed measurement result value as the corresponding optimal adjustment value;

[0024] If the data difference does not exceed a preset threshold, select the adjustment value with the minimum corresponding power consumption as the corresponding optimal adjustment value.

[0025] Preferably, the system parameters include but are not limited to the central processing unit (CPU) frequency, memory frequency, hardware synthesizer status, and graphics processing unit (GPU) frequency.

[0026] Preferably, adjust and measure the speed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically:

[0027] When the system parameter is the CPU frequency, the default value is a specific value, the parameter type is a numerical type, the adjustment range is the CPU overclocking range and the CPU overclocking range does not exceed a preset overclocking percentage of the base frequency;

[0028] Corresponding to three adjustment times, divide the adjustment range into three intervals: the first interval, the second interval, and the third interval, and use the middle value of the three intervals or an adjustable value close to the middle value as the adjustment value;

[0029] Based on each of the adjustment values, adjust the CPU frequency in sequence and measure the speed.

[0030] Preferably, adjust and measure the speed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically:

[0031] When the system parameter is the memory frequency, the default value is a specific value, the parameter type is a numerical type, the corresponding adjustment range is the memory overclocking range and the memory overclocking range does not exceed a preset overclocking percentage of the base frequency;

[0032] Corresponding to three adjustment times, divide the adjustment range into three equal intervals, and use the middle value of the three intervals or an adjustable value close to the middle value as the adjustment value;

[0033] Based on each of the adjustment values, adjust the memory frequency in sequence and measure the speed.

[0034] Preferably, adjust and measure the speed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically:

[0035] When the system parameter is the hardware synthesizer status, the default value is a Boolean value, the parameter type is a Boolean type, and the corresponding adjustment range is the on and off states;

[0036] Based on the on and off states, adjust the hardware synthesizer status in sequence and measure the speed.

[0037] On the other hand, the present invention also provides a speed measurement system with adjustable parameters, including:

[0038] A first determination module, configured to screen the system parameters to be adjusted according to a preset strategy and sort them;

[0039] A second determination module, configured to generate an adjustment range corresponding to each of the system parameters according to the default value of the system parameters, and obtain an adjustment strategy corresponding to the type of each of the system parameters;

[0040] A third determination module, configured to sequentially select the system parameters according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain an optimal adjustment value corresponding to the optimal speed measurement result;

[0041] A fourth determination module, configured to perform adjustment and speed measurement on the next system parameter based on the optimal adjustment value to obtain an optimal adjustment value corresponding to the next system parameter;

[0042] A storage module, configured to, after obtaining the optimal adjustment values of all the system parameters, merge the optimal adjustment values corresponding to all the system parameters into an optimal adjustment set of system parameters.

[0043] The present application discloses a method for adjusting system parameters. The method screens the system parameters to be adjusted according to a preset strategy and sorts them; generates an adjustment range corresponding to each system parameter according to the default value of the system parameters, and obtains an adjustment strategy corresponding to the type of each system parameter; sequentially selects system parameters according to the sorting result, and performs adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain an optimal adjustment value corresponding to the optimal speed measurement result; performs adjustment and speed measurement on the next system parameter based on the optimal adjustment value to obtain an optimal adjustment value corresponding to the next system parameter; after obtaining the optimal adjustment values of all the system parameters, merges the optimal adjustment values corresponding to all the system parameters into an optimal adjustment set of system parameters. By applying the technical solution of the present application, it is possible to reasonably and efficiently adjust each system parameter in the speed measurement system, realize the best performance of the test system, and obtain the optimal speed measurement result. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0045] Figure 1 It is a schematic flowchart of a method for adjusting system parameters;

[0046] Figure 2 This is a schematic diagram of the process of a specific embodiment of the present application;

[0047] Figure 3 This is a schematic diagram of the structure of a speed measurement system with adjustable parameters. Specific Embodiment

[0048] As described in the background art, due to the great differences in network environments, the preset parameters of the speed measurement system cannot exert their best performance, resulting in the network speed measurement indicators not meeting the expectations of users.

[0049] To solve the above problems, the embodiments of the present application propose a system parameter adjustment method. This solution screens and orderly adjusts each system parameter in the speed measurement system to exert the best performance of the test system, thereby obtaining the optimal speed measurement result.

[0050] As Figure 1 shown, this is a schematic diagram of the process of the above system parameter adjustment method. Before introducing the specific solution, the concepts involved in the following text are introduced:

[0051] CPU (Central Processing Unit, central processing unit): As the operation and control core of a computer system, it is the final execution unit for information processing and program operation.

[0052] GPU (graphics processing unit, graphics processing unit): Also known as the display core, visual processor, display chip, it is a microprocessor that specializes in performing image and graphics-related operations on personal computers, workstations, game consoles, and some mobile devices (such as tablets, smartphones, etc.).

[0053] Android: It is a free and open-source mobile operating system based on the Linux kernel.

[0054] Linux: Based on the design concept and architecture of the Unix operating system, it is a multi-user network operating system with stable performance.

[0055] HWC (Hardware Composer, hardware composer): Its function is to accelerate the interface rendering and graphics presentation in the Android system by using hardware acceleration technology, providing higher graphics performance and lower power consumption.

[0056] IRQ (Interrupt ReQuest, interrupt request): It is a signal used to notify the processor that a device needs to be processed. When a device triggers an IRQ signal, the processor will stop the current operation and turn to handle this interrupt request.

[0057] RPS (Receive Packet Steering): It is a network optimization technology in the Linux kernel, aiming to improve the load balancing and parallel processing capabilities of network traffic in a multi-processor system.

[0058] RFS (Receive Flow Steering): It is a technology in the Linux kernel network protocol stack, aiming to evenly distribute the network data traffic received by the network card to multiple CPUs, thereby improving network processing performance and parallel performance.

[0059] Specifically, the present invention aims to achieve the best performance of the speed measurement system, obtain the optimal speed measurement result, and reach the expected network speed measurement index. Therefore, before implementing the steps of this solution, it is necessary to select an optimal speed measurement node.

[0060] It should be noted that there are many methods to select the speed measurement node. Any known scheme can be used to select the optimal speed measurement node, and no further explanation will be given here.

[0061] Specifically, on this basis, the method includes the following steps:

[0062] Step S101: Screen the system parameters to be adjusted according to a preset strategy and sort them.

[0063] The present invention aims to obtain the optimal test result by adjusting different system parameters for different network environments. Therefore, in order to facilitate the orderly adjustment of different system parameters and improve the efficiency and accuracy of adjustment, it is necessary to screen the system parameters to be adjusted and sort them.

[0064] As described above, first obtain and screen the system parameters in the speed measurement system that will affect the network speed measurement result through a preset strategy, and then sort the screened system parameters and output the sorting result, which is convenient for the subsequent parameter adjustment of each system parameter to be carried out in order and improves the adjustment efficiency of the system.

[0065] It should be noted that sorting the system parameters to be adjusted is to improve the adjustment efficiency of the system and ensure that the optimal test result can be obtained relatively quickly. Randomly adjusting each system parameter without sorting and finally achieving the optimization of the speed measurement result also falls within the protection scope of this application.

[0066] Step S102: Generate an adjustment range corresponding to each of the system parameters according to the default value of the system parameters, and obtain an adjustment strategy corresponding to the type of each of the system parameters.

[0067] The present invention aims to adjust various system parameters on the premise of ensuring system stability, so as to achieve more excellent performance and obtain better test results. Therefore, in order to ensure the stability of the system, this step generates corresponding adjustment ranges for each system parameter through the default values of each system parameter. When adjusting each system parameter category, according to the adjustment strategy of the parameter type of each system parameter and its adjustment range, it is ensured that the change range of the adjusted parameter does not exceed a reasonable range, avoiding overly extreme parameter settings and ensuring the stability of the system.

[0068] Step S103: Select the system parameters one by one according to the sorting result, and adjust and measure the speed according to the corresponding adjustment range and adjustment strategy of the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0069] The present invention aims to find the optimal adjustment value of each system parameter, optimize the overall performance of the system, and ensure the best speed measurement result. Therefore, in this step, each system parameter is adjusted and the speed is measured in sequence according to its adjustment range and adjustment strategy, the speed measurement results after each adjustment of each system parameter are obtained and compared, and the parameter adjustment value corresponding to the optimal speed measurement result is selected as the optimal adjustment value of the system parameter category.

[0070] Step S104: Based on the optimal adjustment value, adjust and measure the speed of the next system parameter to obtain the optimal adjustment value corresponding to the next system parameter.

[0071] The present invention aims to gradually optimize each system parameter through step-by-step iteration to obtain the global optimal adjustment value. Therefore, in this step, based on the optimal adjustment values of each system parameter obtained, the next system parameter is adjusted and the speed is measured in sequence until the optimal adjustment values corresponding to all the screened system parameters are obtained, ensuring that the optimal configuration of the entire system can be accurately found, realizing the overall performance optimization of the system, and thus obtaining the best speed measurement result.

[0072] It should be noted that only adjusting some system parameter categories to the optimal system parameters and realizing partial optimization of the system, so that the speed measurement result reaches the expected index, also belongs to the protection scope of this application.

[0073] S105: After obtaining the optimal adjustment values of each system parameter, merge the optimal adjustment values corresponding to each system parameter into an optimal adjustment set of system parameters.

[0074] The present invention aims to determine the optimal system settings corresponding to the network environment, so as to achieve the purpose of not needing to repeat the adjustment test when measuring the speed of the network environment later. Therefore, this step merges the optimal adjustment values corresponding to each system parameter to form an optimal adjustment set of system parameters, and this set represents the optimal configuration of the system corresponding to the network environment.

[0075] As described above, the optimal adjustment values corresponding to each system parameter are aggregated to form an optimal adjustment set for the system in this network environment. In the later stage, when measuring the network speed in this network environment again, there is no need to repeat the adjustment test. Just calling this optimal adjustment set can make the entire system reach the best performance.

[0076] It should be noted that each optimal adjustment set obtained by the system for different network environments belongs to the protection scope of this solution, whether it is stored at a local storage location of the device for the system to call later, or stored in the cloud for other devices to call when facing the same network environment.

[0077] To make the adjustment process of system parameters more organized and efficient, and to ensure that the adjustment order of system parameters conforms to the optimized logic, the system parameters to be adjusted are screened and sorted according to a preset strategy. Specifically, there are the following two solutions:

[0078] (1) Analyze the architecture of the system to obtain the influence of each of the system parameters on the performance of the system, and screen and sort the system parameters in descending order of influence.

[0079] Specifically, when analyzing the architecture of the system, first understand and evaluate each parameter of the system, consider the influence of each system parameter on the overall system performance, and evaluate their relative importance in affecting the system performance. For example, in some application scenarios, the clock frequency of the CPU may have a greater impact on the system performance, while in other scenarios, the processing power of the GPU may be more critical. According to these analysis results, the system parameters that can affect the speed measurement result are screened out, and the screened system parameters are sorted in descending order of their influence degree.

[0080] (2) Conduct a performance benchmark test on the system, and screen and sort the system parameters based on the results of the performance benchmark test.

[0081] Specifically, when conducting the performance benchmark test, each parameter of the system will be affected to varying degrees, thus affecting the performance of the entire system. Based on the results of the performance benchmark test, the actual influence degree of each parameter on the system performance can be intuitively understood, providing objective data support for the screening and sorting of system parameters. By simulating real workloads or application scenarios, measuring the performance of the system under different parameter settings, and based on the actual performance data, the system parameters that can affect the speed measurement result are screened out, and the screened system parameters are sorted in descending order of their influence degree.

[0082] It should be noted that the above two sorting methods can be used alone or in combination according to specific circumstances. For example, a preliminary sorting can be performed based on system architecture analysis first, and then the sorting results can be verified and fine-tuned through performance benchmark tests. Such a comprehensive sorting strategy can more comprehensively consider the impact of system parameters.

[0083] In order to adjust the parameters corresponding to each system parameter category on the premise of ensuring system stability, so as to achieve more excellent performance and obtain better test results, and obtain adjustment strategies corresponding to the types of each of the system parameters. Specifically, there are the following two solutions:

[0084] (1) The system parameter type corresponding to the default value being a boolean value is the boolean type, and the adjustment strategy is to sequentially adjust the system parameters based on the adjustment range.

[0085] As described above, system parameters of the boolean type usually represent switch states and have only two possible values, namely true or false. In this case, the adjustment range is the switching of boolean values, and the corresponding adjustment strategy is to change from true to false or from false to true.

[0086] (2) The system parameter type corresponding to the default value being a specific numerical value is the numerical type, and the adjustment strategy is to sequentially adjust the system parameters according to the intervals after dividing the adjustment range into multiple intervals based on a preset number of adjustment times.

[0087] As described above, system parameters of the numerical type may have a continuous value range. According to the hardware specifications and design limits of each system parameter of the numerical type, physical limits are clarified, such as the maximum frequency of the CPU, the maximum capacity of the memory, etc. Then, combined with the stability requirements of the system, a reasonable parameter adjustment range is determined. This ensures that the adjustment process is carried out within the limits of the system hardware and stability. In this case, in order to adjust the parameters more meticulously, the corresponding adjustment strategy is to divide the adjustment range into multiple intervals and perform successive adjustments within each interval.

[0088] The introduction of the preset number of adjustment times can efficiently search for the optimal solution in the parameter space, divide the parameter adjustment range into multiple intervals, and sequentially adjust each system parameter category according to the preset number of times, which helps the system gradually approach the optimal configuration and avoids blind search of the entire parameter space. Specifically, first, according to the preset value of the number of adjustment times, the adjustment range of each system parameter is reasonably divided, and then the system parameters are orderly adjusted and speed-tested according to the divided multiple intervals. In this way, a more comprehensive search can be achieved in a limited number of adjustments, accelerating the process of finding the optimal configuration.

[0089] It should be noted that the separate implementation or combined use of the above two solutions both fall within the protection scope of this application.

[0090] In order to achieve more accurate and balanced parameter adjustment for each system parameter category, the adjustment range is divided into multiple intervals based on a preset number of adjustment times, specifically as follows:

[0091] If the parameter adjustment range is divisible by the preset number of adjustment times, the range sizes of the multiple intervals are set to be the same in an evenly divided manner, and the adjustable value at the middle of the range of each interval or an adjustable value close to the middle value is used as the adjustment parameter;

[0092] If the parameter adjustment range is not divisible by the preset number of adjustment times, the ranges of the multiple intervals are determined based on the rounding rule, so that the range sizes of the remaining intervals except the last one are the same, and the adjustable value at the middle of the range of each interval or an adjustable value close to the middle value is used as the adjustment parameter.

[0093] It should be noted that when dividing the parameter adjustment range into multiple intervals, various methods can be used for the evenly divided manner, not limited to the divisible case. In addition to evenly dividing the range size, factors such as the sensitivity of the parameter and the degree of influence on performance can also be considered for dynamic allocation, so that the key areas can be more targeted during the parameter adjustment process. For example, for parameters with a relatively simple linear relationship, the parameter adjustment range can be divided by an equal division method; for parameters with a non-linear relationship, when dividing the intervals, the interval sizes can be dynamically allocated according to the non-linear characteristics of the parameter, ensuring that more adjustment times are invested in the area where the optimal solution may exist, which is conducive to more comprehensively searching the entire parameter space. At the same time, the rounding processing method can be based on different rules, such as using the nearest even number method, which is conducive to maintaining the fairness and consistency of parameter adjustment.

[0094] In order to find the configuration with the best system performance and improve the speed measurement accuracy and efficiency of the system, the system parameters are selected one by one according to the sorting result, and adjusted and speed measured according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result, specifically as follows:

[0095] Data differences are obtained based on the speed measurement results of each time;

[0096] If the data difference exceeds the preset threshold, the adjustment value with the highest speed measurement result value is selected as the corresponding optimal adjustment value;

[0097] If the data difference does not exceed the preset threshold, the adjustment value corresponding to the minimum power consumption is selected as the corresponding optimal adjustment value.

[0098] As described above, the data difference is obtained by analyzing and comparing the results of each speed measurement. By calculating the differences between these speed measurement results, we can more comprehensively understand the impact of parameter adjustments of each system parameter on system performance. If the data difference exceeds the preset threshold, it can be understood that the adjustment of the system parameters has caused an obvious performance change. In this case, the adjustment value with the highest corresponding speed measurement result value is selected as the optimal adjustment value; if the data difference does not exceed the preset threshold, it indicates that the performance difference of the system is small. At this time, the adjustment value with the smallest corresponding power consumption is selected as the optimal adjustment value to reduce the system power consumption and improve the energy efficiency, meeting the requirements of energy conservation and resource utilization efficiency.

[0099] To better meet the requirements of system performance adjustment in different scenarios, the system parameters to be adjusted are screened and sorted according to a preset strategy, specifically:

[0100] The system parameters to be screened include but are not limited to CPU frequency, memory frequency, hardware synthesizer status, and GPU frequency.

[0101] It should be noted that when sorting the system parameters, their correlation needs to be considered to ensure that the sorting order meets the actual system performance optimization requirements. For example, if a certain task depends on the performance of both the CPU and the GPU, these two system parameters can be given priority to ensure that they have a higher priority in the sorting.

[0102] To improve the speed measurement accuracy and efficiency of the system and obtain better test results, adjustments and speed measurements are performed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically:

[0103] When the system parameter is the CPU frequency, the default value is a specific value, the parameter type is a numerical parameter, the adjustment range is the CPU overclocking range and the CPU overclocking range does not exceed the preset overclocking percentage of the base frequency;

[0104] Corresponding to three adjustment times, the adjustment range is divided into three intervals: the first interval, the second interval, and the third interval. The intermediate value of the three intervals or an adjustable value close to the intermediate value is used as the adjustment value;

[0105] Based on each of the adjustment values, the CPU frequency is adjusted and speed measured in sequence.

[0106] In a specific implementation scenario of the present invention, when adjusting the CPU frequency multiplier, based on factors such as its own model, motherboard support, power supply capacity, and system stability, it is determined that the overclocking range does not exceed 20% of the base frequency. Combining with the preset adjustment times three times, the three adjustment intervals of the CPU are divided into a first interval of 7%, a second interval of 7%, and a third interval of 6%. The frequency of the CPU is sequentially set to the intermediate value of the three intervals or an adjustable value close to the intermediate value and the speed is measured. According to the three speed measurement results, the optimal set frequency multiplier of the CPU is selected accordingly.

[0107] To improve the speed measurement accuracy and efficiency of the system and obtain better test results, adjustments and speed measurements are performed according to the adjustment range and adjustment strategy corresponding to the system parameters. Specifically:

[0108] When the system parameter is the memory frequency, the default value is a specific value, the parameter type is a numerical type, and the corresponding adjustment range is the memory overclocking range and the memory overclocking range does not exceed the preset overclocking percentage of the base frequency;

[0109] Corresponding to the three adjustment times, the adjustment range is evenly divided into three intervals, and the intermediate value of the three intervals or an adjustable value close to the intermediate value is used as the adjustment value;

[0110] Based on each of the adjustment values, the memory frequency is sequentially adjusted and the speed is measured.

[0111] In a specific implementation scenario of the present invention, when adjusting the memory frequency, based on factors such as the memory chip type and motherboard support ability, it is determined that the overclocking range does not exceed 15% of the base frequency. Combining with the preset adjustment times three times, the three adjustment intervals of the memory are equally divided into three intervals of 5%. The frequency of the memory is sequentially set to the intermediate value of the three intervals or an adjustable value close to the intermediate value and the speed is measured. According to the three speed measurement results, the optimal set frequency of the memory is selected accordingly.

[0112] To improve the speed measurement accuracy and efficiency of the system and obtain better test results, adjustments and speed measurements are performed according to the adjustment range and adjustment strategy corresponding to the system parameters. Specifically:

[0113] When the system parameter is the hardware synthesizer state, the default value is a boolean value, the parameter type is a boolean type, and the corresponding adjustment range is the on and off states;

[0114] Based on the on and off states, the hardware synthesizer state is sequentially adjusted and the speed is measured.

[0115] In a specific implementation scenario of the present invention, the adjustment range for determining the state of the hardware synthesizer is two states: on and off. Set them to the on state and the off state respectively and measure the speed. Select the optimal setting of the hardware synthesizer state according to the two speed measurement results.

[0116] Compared with the prior art, this method screens the system parameters to be adjusted according to a preset strategy and sorts them; generates an adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtains an adjustment strategy corresponding to each system parameter type; selects the system parameters one by one according to the sorting result, and adjusts and measures the speed according to the adjustment range and adjustment strategy corresponding to the system parameter to obtain the optimal adjustment value corresponding to the optimal speed measurement result; adjusts and measures the speed of the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter; when the optimal adjustment values of all system parameters are obtained, merge the optimal adjustment values corresponding to each system parameter into an optimal adjustment set of system parameters. By applying the technical solution of this application, it is possible to reasonably and efficiently adjust each system parameter in the speed measurement system, achieve the best performance of the test system, and obtain the optimal speed measurement result.

[0117] Next, the technical solutions in this application will be described clearly and completely with reference to the accompanying drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0118] The main steps in this specific embodiment are as Figure 2 shown, which is a schematic flowchart of a specific embodiment of this application.

[0119] In this specific embodiment, an optimal speed measurement node is selected automatically or manually in the current network environment, and then the selected speed measurement node is measured for speed. If the speed measurement result does not reach the expected network speed measurement index, it proves that the speed measurement system does not reach the best performance. Therefore, it is necessary to adjust the parameters of the speed measurement system to obtain the optimal speed measurement result.

[0120] Based on the above settings, the system parameter setting scheme mainly includes the following steps:

[0121] Step S201: Screen the system parameters to be adjusted according to a preset strategy and sort them.

[0122] Obtain and filter the system parameters that will affect the network speed measurement results in the system, and sort the filtered system parameters according to the degree of influence from large to small. Finally, obtain the sorting result of "CPU, DDR memory, HWC hardware synthesizer, GPU, interrupt signal affinity, network packet flow direction, and network socket flow direction".

[0123] Step S301: Generate an adjustment range corresponding to each of the system parameters according to the default values of the system parameters, and obtain an adjustment strategy corresponding to the type of each of the system parameters.

[0124] Determine the adjustment range of the CPU frequency within the system allowable range, divide the frequency adjustment range of the CPU based on the number of adjustments, and adjust the frequency of the CPU and measure the speed in sequence according to the divided results.

[0125] Step S302: Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0126] Obtain and compare the speed measurement data after each adjustment of the CPU frequency. If the data difference exceeds the preset threshold, select the frequency corresponding to the highest speed measurement result value as the optimal CPU frequency of the system; if the data difference does not exceed the preset threshold, select the frequency corresponding to the minimum power consumption as the optimal CPU frequency of the system.

[0127] Step S303: Based on the optimal adjustment value, perform adjustment and speed measurement on the next system parameter to obtain the optimal adjustment value corresponding to the next system parameter.

[0128] After setting the system to the optimal CPU frequency, perform adjustment tests on the DDR memory frequency in sequence.

[0129] Step S311: Generate an adjustment range corresponding to each of the system parameters according to the default values of the system parameters, and obtain an adjustment strategy corresponding to the type of each of the system parameters.

[0130] Determine the frequency adjustment range of the DDR memory within the system allowable range, divide the frequency adjustment range of the DDR memory based on the number of adjustments, and adjust the frequency of the DDR memory and measure the speed in sequence according to the divided results.

[0131] Step S312: Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0132] Obtain and compare the speed measurement data after each adjustment of the DDR memory frequency. If the data difference exceeds the preset threshold, select the frequency corresponding to the highest speed measurement result value as the optimal DDR memory frequency of the system; if the data difference does not exceed the preset threshold, select the frequency corresponding to the minimum power consumption as the optimal DDR memory frequency of the system.

[0133] Step S313: Adjust and measure the speed of the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter.

[0134] After setting the system to the optimal DDR memory frequency, adjust and test the HWC hardware synthesizer in sequence.

[0135] Step S321: Generate the adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtain the adjustment strategy corresponding to the type of each system parameter.

[0136] Determine that the parameter adjustment range for the HWC hardware synthesizer status is two states: on and off. Measure the speed of the HWC hardware synthesizer in sequence in the on state and the off state respectively.

[0137] Step S322: Select each system parameter one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameter to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0138] Obtain and compare the speed measurement data after two adjustments of the HWC hardware synthesizer status. If the data difference exceeds the preset threshold, select the status corresponding to the higher speed measurement result value as the optimal HWC hardware synthesizer status of the system; if the data difference does not exceed the preset threshold, select the status corresponding to the minimum power consumption as the optimal HWC hardware synthesizer status of the system.

[0139] Step S323: Adjust and measure the speed of the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter.

[0140] After setting the system to the optimal HWC hardware synthesizer status, adjust and test the GPU in sequence.

[0141] Step S331: Generate the adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtain the adjustment strategy corresponding to the type of each system parameter.

[0142] Determine the frequency adjustment range of the GPU within the system allowable range, divide the frequency adjustment range of the GPU based on the number of adjustments, and adjust and measure the speed of the GPU according to the divided results in sequence.

[0143] Step S332: Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0144] Obtain and compare the speed measurement data after each adjustment of the GPU frequency. If the data difference exceeds the preset threshold, select the frequency corresponding to the highest speed measurement result value as the optimal GPU frequency of the system; if the data difference does not exceed the preset threshold, select the frequency corresponding to the minimum power consumption as the optimal GPU frequency of the system.

[0145] Step S333: Based on the optimal adjustment value, adjust and measure the speed of the next system parameter to obtain the optimal adjustment value corresponding to the next system parameter.

[0146] After setting the system to the optimal GPU frequency, adjust and test the interrupt signal affinity in sequence.

[0147] Step S341: Generate the adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtain the adjustment strategy corresponding to the type of each system parameter.

[0148] Adjust the network card hardware interrupt affinity (irq smp_affinity), bind the interrupt signals to different CPUs one by one and perform speed measurement.

[0149] Step S342: Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0150] Obtain and compare the speed measurement data after each adjustment of the interrupt signal affinity, and select the parameter with the best effect as the optimal interrupt signal setting of the system.

[0151] Step S343: Based on the optimal adjustment value, adjust and measure the speed of the next system parameter to obtain the optimal adjustment value corresponding to the next system parameter.

[0152] After setting the system to the optimal interrupt signal setting, adjust and test the flow direction of network data packets in sequence.

[0153] Step S351: Generate the adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtain the adjustment strategy corresponding to the type of each system parameter.

[0154] Adjust the flow direction of network data packets (rps receive packet steering) so that network data packets are distributed to different CPUs for processing and perform speed measurement.

[0155] Step S352: Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0156] Output the test results under various permutations and combinations one by one, and select the combination with the best effect as the optimal network data packet flow setting of the system.

[0157] Step S353: Adjust and measure the speed of the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter.

[0158] After setting the system to the optimal network data packet flow setting, adjust and test the network socket flow in sequence.

[0159] Step S361: Generate an adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtain an adjustment strategy corresponding to the type of each system parameter.

[0160] Adjust the network socket flow (rfs receive flow steering) so that the data stream of each socket is processed by the same CPU as much as possible to improve the cache hit rate. Gradually increase the number of receive queues (rps_sock_flow_entries) until the speed measurement result does not increase significantly.

[0161] Step S362: Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result.

[0162] Select the number of receive queues corresponding to the highest speed measurement result value as the optimal network socket flow setting of the system.

[0163] Step S401: After obtaining the optimal adjustment values of all the system parameters, merge the optimal adjustment values corresponding to each system parameter into an optimal adjustment set of system parameters.

[0164] After several rounds of speed measurement in the front, the optimal adjustment values of each optimal system parameter selected are the optimal adjustment set of the system in the current environment, and store the finally obtained optimal adjustment set in a specific location in the non-volatile memory of the system.

[0165] It should be noted that the above-mentioned test selection process of parameters only needs to be run once in a device under the same network environment. After obtaining the optimal adjustment set, it can be automatically synchronized to the cloud, and then pushed by the server to all devices in this network environment for storage and use.

[0166] Preferably, the present invention further includes a system service process for detecting whether the system is connected to the network. When the network is connected, this service automatically sets the optimal adjustment set previously written to a specific location into the system to improve the system network performance. When the network is disconnected, this service restores each system parameter to its default value to reduce the possible increase in system power consumption caused by the previously set optimal adjustment set.

[0167] Based on the same inventive concept as the above method, an embodiment of the present application also proposes a speed measurement system with adjustable parameters, as Figure 3 shown in the structural schematic diagram of a speed measurement system with adjustable parameters. This system includes:

[0168] A first determination module for screening the system parameters to be adjusted according to a preset strategy and sorting them;

[0169] A second determination module for generating an adjustment range corresponding to each of the system parameters according to the default values of the system parameters, and obtaining an adjustment strategy corresponding to the type of each of the system parameters;

[0170] A third determination module for sequentially selecting the system parameters according to the sorting result, and adjusting and measuring the speed according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result;

[0171] A fourth determination module for adjusting and measuring the speed of the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter;

[0172] A storage module for, after obtaining the optimal adjustment values of all the system parameters, combining the optimal adjustment values corresponding to each of the system parameters into an optimal adjustment set of system parameters.

[0173] Compared with the prior art, a system parameter adjustment method proposed in an embodiment of the present application screens system parameters to be adjusted according to a preset policy and sorts them; generates an adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtains an adjustment policy corresponding to each system parameter type; selects system parameters one by one according to the sorting result, and adjusts and measures the speed according to the adjustment range and adjustment policy corresponding to the system parameter to obtain an optimal adjustment value corresponding to the optimal speed measurement result; adjusts and measures the speed of the next system parameter based on the optimal adjustment value to obtain an optimal adjustment value corresponding to the next system parameter; when the optimal adjustment values of all system parameters are obtained, the optimal adjustment values corresponding to each system parameter are merged into an optimal adjustment set of system parameters. By applying the technical solution of the present application, it is possible to reasonably and efficiently adjust each system parameter in the speed measurement system, achieve the best performance of the test system, and obtain the optimal speed measurement result.

[0174] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present invention.

[0175] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present invention.

[0176] Those skilled in the art can understand that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed to be located in one or more devices different from the present implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0177] The above serial numbers of the present invention are only for description and do not represent the advantages or disadvantages of the implementation scenarios.

[0178] The above discloses only several specific implementation scenarios of the present invention. However, the present invention is not limited thereto, and any change that can be thought of by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for adjusting system parameters, characterized in that Including: Screen the system parameters to be adjusted according to a preset strategy and sort them; Generate an adjustment range corresponding to each of the system parameters according to the default value of the system parameters, and obtain an adjustment strategy corresponding to the type of each of the system parameters; Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result; Adjust and measure the speed of the next system parameter based on the optimal adjustment value to obtain the optimal adjustment value corresponding to the next system parameter; After obtaining the optimal adjustment values of all the system parameters, merge the optimal adjustment values corresponding to all the system parameters into an optimal adjustment set of system parameters.

2. The method according to claim 1, wherein Screen the system parameters to be adjusted according to a preset strategy and sort them, specifically: Analyze the architecture of the system to obtain the influence of each of the system parameters on the performance of the system, and screen and sort the system parameters in descending order of influence; Or, perform a performance benchmark test on the system, and screen and sort the system parameters according to the performance benchmark test result.

3. The method according to claim 2, wherein Obtain the adjustment strategy corresponding to the type of each of the system parameters, specifically: The system parameter type corresponding to the default value being a boolean value is a boolean type, and the adjustment strategy is to sequentially adjust the system parameter based on the adjustment range; The system parameter type corresponding to the default value being a specific numerical value is a numerical type, and the adjustment strategy is to sequentially adjust the system parameter according to the interval after dividing the adjustment range into multiple intervals based on a preset number of adjustment times.

4. The method according to claim 3, wherein Divide the adjustment range into multiple intervals based on a preset number of adjustment times, specifically: If the adjustment range is divisible by the preset number of adjustment times, set the range sizes of the multiple intervals to be the same in an evenly divided manner, and use the range intermediate value or an adjustable value close to the intermediate value of each interval as the adjustment parameter; If the adjustment range is not divisible by the preset number of adjustment times, determine the ranges of the multiple intervals based on the rounding rule, so that the range sizes of the remaining intervals except the last interval are the same, and use the range intermediate value or an adjustable value close to the intermediate value of each interval as the adjustment parameter.

5. The method according to claim 4, wherein Select the system parameters one by one according to the sorting result, and perform adjustment and speed measurement according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain the optimal adjustment value corresponding to the optimal speed measurement result, specifically: Obtain a data difference based on the speed measurement results of each time; If the data difference exceeds a preset threshold, select the adjustment value with the highest speed measurement result value as the corresponding optimal adjustment value; If the data difference does not exceed the preset threshold, select the adjustment value with the minimum corresponding power consumption as the corresponding optimal adjustment value.

6. The method according to claim 5, wherein: The system parameters include but are not limited to the central processing unit (CPU) frequency, memory frequency, hardware synthesizer status, and graphics processing unit (GPU) frequency.

7. The method according to claim 6, characterized in that, Adjust and measure speed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically as follows: When the system parameter is the CPU frequency, the default value is a specific value, the parameter type is a numerical type, the adjustment range is the CPU overclocking range and the CPU overclocking range does not exceed the preset overclocking percentage of the base frequency; Corresponding to three adjustment times, divide the adjustment range into three intervals: the first interval, the second interval, and the third interval, and use the intermediate value of the three intervals or an adjustable value close to the intermediate value as the adjustment value; Based on each of the adjustment values, adjust the CPU frequency in sequence and measure the speed.

8. The method according to claim 6, wherein Adjust and measure speed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically as follows: When the system parameter is the memory frequency, the default value is a specific value, the parameter type is a numerical type, the corresponding adjustment range is the memory overclocking range and the memory overclocking range does not exceed the preset overclocking percentage of the base frequency; Corresponding to three adjustment times, divide the adjustment range evenly into three intervals, and use the intermediate value of the three intervals or an adjustable value close to the intermediate value as the adjustment value; Based on each of the adjustment values, adjust the memory frequency in sequence and measure the speed.

9. The method according to claim 6, wherein Adjust and measure speed according to the adjustment range and adjustment strategy corresponding to the system parameters, specifically as follows: When the system parameter is the hardware synthesizer state, the default value is a boolean value, the parameter type is a boolean type, and the corresponding adjustment range is the on and off states; Based on the on and off states, adjust the hardware synthesizer state in sequence and measure the speed.

10. A speed measurement system with adjustable parameters, characterized in that, Including: A first determination module, configured to screen system parameters to be adjusted according to a preset strategy and sort them; A second determination module, configured to generate an adjustment range corresponding to each system parameter according to the default value of the system parameter, and obtain an adjustment strategy corresponding to the type of each system parameter; A third determination module, configured to sequentially select the system parameters according to the sorting result, and adjust and measure the speed according to the adjustment range and adjustment strategy corresponding to the system parameters, so as to obtain an optimal adjustment value corresponding to the optimal speed measurement result; A fourth determination module, configured to adjust and measure the speed of the next system parameter based on the optimal adjustment value to obtain an optimal adjustment value corresponding to the next system parameter; A storage module, configured to, after obtaining the optimal adjustment values of all the system parameters, merge the optimal adjustment values corresponding to each system parameter into a system parameter optimal adjustment set.

Images