CN112039689B - Network equipment performance evaluation method, device, equipment and storage medium - Google Patents

Abstract

The embodiment of the present invention relates to the field of inspection technology, and discloses a network equipment performance evaluation method, device, equipment and storage medium. In the present invention, network devices with the same hardware configuration information and the same CDN application type are divided into the same category, and then performance evaluation results for the network devices of the same category are generated based on the performance evaluation data corresponding to the network devices of the same category, thereby The problem of dealing with a single network device is converted into the problem of dealing with a type of network equipment, which greatly improves operational efficiency.

Description

Network equipment performance evaluation methods, devices, equipment and storage media

Technical field

The embodiments of the present invention relate to the field of inspection technology, and in particular to a network equipment performance evaluation method, device, equipment and storage medium.

Background technique

CDN (Content Delivery Network) is a content distribution network built on the network. It relies on edge network devices deployed in various places, specifically servers, and uses the load balancing, content distribution, scheduling and other functional modules of the central platform to enable users to obtain all their content nearby. content, reduce network congestion, and improve user access response speed and hit rate. In order to achieve this effect, a huge amount of data needs to be cached, which requires a large number of CDN servers to complete the caching, making the equipment cost account for the majority of the overall cost. Therefore, while meeting business performance requirements, it is necessary to evaluate server performance to ensure service stability and achieve better equipment cost control.

However, the current way to evaluate server performance is to configure a corresponding monitoring system for each server to be evaluated, and evaluate the corresponding servers based on the configured monitoring system. That is, the evaluation processes are independent of each other and cannot be reasonable at all. Reflects the performance of the server, and since this method requires configuring a corresponding monitoring system for each server to be evaluated, it not only increases the implementation cost, but also reduces the overall operational efficiency.

Contents of the invention

The purpose of the embodiments of the present invention is to provide a network equipment performance evaluation method, device, equipment and storage medium, so as to convert the problem of processing a single network device into the problem of processing a type of network equipment, thereby improving operational efficiency.

In order to solve the above technical problems, embodiments of the present invention provide a network device performance evaluation method, which includes the following steps:

Obtain the performance evaluation data corresponding to each network device;

Obtain the hardware configuration information and content distribution network CDN application type corresponding to each network device;

According to the hardware configuration information and the CDN application type, network devices with the same hardware configuration information and the same CDN application type are classified into the same category;

According to the performance evaluation data corresponding to each network device of the same category, a performance evaluation result for each network device of the same category is generated.

An embodiment of the present invention also provides a device, including:

at least one processor; and,

a memory communicatively connected to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can perform the network device performance evaluation method as described above.

Embodiments of the present invention also provide a computer-readable storage medium that stores a computer program. When the computer program is executed by a processor, the network device performance evaluation method as described above is implemented.

Compared with the existing technology, the embodiment of the present invention divides network devices with the same hardware configuration information and the same CDN application type into the same category, and then generates networks for the same category based on the performance evaluation data corresponding to the network devices of the same category. The performance evaluation results of the equipment can be converted from the problem of dealing with a single network device to the problem of dealing with a type of network equipment, which greatly improves operational efficiency.

In addition, the obtaining of performance evaluation data corresponding to each network device includes:

Based on the preset performance evaluation index, obtain the performance evaluation data generated by each network device in different time periods for the performance evaluation index.

In addition, based on the preset performance evaluation indicators, the performance evaluation data generated by each network device in different time periods for the performance evaluation indicators is obtained, including:

For each time period, determine the service traffic size corresponding to each network device at different time points within the time period;

According to the size of the business traffic, select the performance evaluation data generated by each network device for the performance evaluation index at a time point when the size of the business traffic is the largest.

In addition, the performance evaluation indicators include business characteristic indicators, equipment performance indicators and service quality indicators;

The performance evaluation data generated by each network device for the performance evaluation indicators at the selected time point with the largest business traffic includes:

Select the business feature data for the business feature index, the device performance data for the device performance index, and the service quality data for the service quality index generated by each network device at the time point with the largest business traffic volume, to obtain the above Performance evaluation data.

In addition, generating performance evaluation results for each network device of the same category based on the performance evaluation data corresponding to each network device of the same category includes:

Aggregate the performance evaluation data corresponding to each network device under the same category according to the preset performance evaluation indicators;

Calculate a performance evaluation value based on the aggregated performance evaluation data, and the performance evaluation value can reflect the performance changes of each network device under the same classification;

Compare the performance evaluation value with a preset performance evaluation threshold;

According to the comparison results, performance evaluation results for each network device of the same category are generated.

In addition, the performance evaluation value includes the peak value, average value and proportion corresponding to the performance evaluation indicator;

The calculation of the performance evaluation value based on the aggregated performance evaluation data includes:

Count the number of network devices that need to be inspected and obtain the total number of network devices;

Based on the data volume of the performance evaluation data corresponding to each network device under the same classification, determine the number of network devices that exceed the preset performance evaluation data volume threshold, and obtain the number of super-threshold network devices;

Calculate the proportion based on the total number of network devices and the number of super-threshold network devices;

Calculate the average value based on the aggregated performance evaluation data and the number of super-threshold network devices;

The peak value is calculated based on the aggregated performance evaluation data and the total number of network devices.

In addition, the performance evaluation thresholds include peak thresholds, mean thresholds and proportion thresholds;

Comparing the performance evaluation value with a preset performance evaluation threshold includes:

Determine whether the peak value is greater than the peak value threshold;

If the peak value is not greater than the peak value threshold, determine whether the mean value is greater than the mean value threshold;

If the mean is not greater than the mean threshold, determine whether the proportion is greater than the proportion threshold;

Wherein, the performance evaluation results for each network device of the same category are generated based on the comparison results, including:

If the peak value is not greater than the peak threshold, the mean value is not greater than the mean threshold value, and the proportion is not greater than the proportion threshold value, then generate abnormal performance for each network device of the same category. evaluation result;

Otherwise, the performance evaluation data corresponding to the performance evaluation value that is greater than the preset performance evaluation threshold is abnormally marked to obtain the abnormal performance evaluation result for each network device of the same category.

In addition, the performance evaluation data corresponding to the performance evaluation value that is greater than the preset performance evaluation threshold is abnormally marked to obtain the abnormal performance evaluation results for each network device of the same category, including:

When the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the service characteristic data corresponding to the service characteristic indicator, the obtained performance evaluation results indicating that each network device of the same category has abnormalities includes any of the following: Or several: the volatility of http requests increases, the volatility of https requests increases, the number of read and write operations per second IOPS volatility increases;

When the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the device performance data corresponding to the device performance indicator, the obtained performance evaluation results indicating abnormality in each network device of the same category include any of the following: Or several: increased volatility + a high proportion of days that are too high, an increase in volatility + a proportion of days that are too high, 0, an increase in volatility + a proportion of days that are too low, 0, a decrease in volatility + a proportion of days that are too high, 0 , Volatility reduction + too low days ratio 0, too low days ratio;

When the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the service quality data corresponding to the service quality indicator, the obtained abnormal performance evaluation results for each network device of the same category include any of the following Or a few things: the number of timeout machines fluctuates and increases, the average bandwidth fluctuates and decreases, and the average number of requests fluctuates and decreases.

In addition, after generating performance evaluation results for each network device of the same category based on the comparison results, the method further includes:

According to the performance evaluation results, a network device performance adjustment strategy for each network device of the same category is generated.

In addition, generating a network device performance adjustment strategy for each network device of the same category based on the performance evaluation results includes:

When the performance evaluation results are for the business characteristic indicators and include any one or more of the following: http request volatility increase, https request volatility rate increase, IOPS volatility rate increase, generate the same classification The network equipment performance adjustment strategies for each network equipment are: adjusting the business planning plan;

The performance evaluation results are based on equipment performance indicators, and include any one or more of the following: increased volatility + proportion of too high days 0, increased volatility + proportion of too low days 0, decreased volatility + The proportion of days that are too high is 0, the volatility is reduced + the number of days that are too low is 0, and the network device performance adjustment strategy generated for each network device of the same category is: continue to pay attention, and do not process it for the time being;

When the performance evaluation results are for equipment performance indicators, and include: increased volatility + high proportion of excessively high days, the network device performance adjustment strategy generated for each network device of the same category is any one or more of the following: Category: Improve corresponding hardware configuration, reduce rated equipment capabilities, optimize software performance, and adjust business planning solutions;

When the performance evaluation results are for device performance indicators and include: the proportion of days with too low values, the network device performance adjustment strategy generated for each network device of the same category is any one or more of the following: Improve the corresponding hardware configuration , reduce the rated equipment capacity;

When the performance evaluation results are based on service quality indicators and include any one or more of the following: increased fluctuations in the number of timeout machines, decreased fluctuations in average bandwidth, and decreased fluctuations in the average number of requests, each of the performance evaluation results for the same category is generated. The network equipment performance adjustment strategy for network equipment is any one or more of the following: improving the corresponding hardware configuration and reducing the rated equipment capacity.

Description of the drawings

One or more embodiments are exemplified by the pictures in the corresponding drawings. These exemplary illustrations do not constitute limitations to the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements. Unless otherwise stated, the figures in the drawings are not intended to be limited to scale.

Figure 1 is a specific flow chart of a network device performance evaluation method according to the first embodiment of the present invention;

Figure 2 is a schematic diagram of the source of performance evaluation data in the network equipment performance evaluation method according to the first embodiment of the present invention;

Figure 3 is a schematic diagram of the source of performance evaluation values in the network equipment performance evaluation method according to the first embodiment of the present invention;

Figure 4 is a schematic diagram of the configuration of performance evaluation thresholds in the network device performance evaluation method according to the first embodiment of the present invention;

Figure 5 is a specific flow chart of a network device performance evaluation method according to the second embodiment of the present invention;

Figure 6 is a schematic structural diagram of a network equipment performance evaluation device according to a third embodiment of the present invention;

Figure 7 is a schematic structural diagram of equipment according to the fourth embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, each embodiment of the present invention will be described in detail below with reference to the accompanying drawings. However, those of ordinary skill in the art will understand that in various embodiments of the present invention, many technical details are provided to enable readers to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solution claimed in this application can also be implemented.

The following divisions of the various embodiments are for convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined with each other and quoted from each other on the premise that there is no contradiction.

The first embodiment of the present invention relates to a network device performance evaluation method, which obtains performance evaluation data corresponding to each network device and device classification information corresponding to each network device; classifies each network device based on the device classification information; according to the same The performance evaluation data corresponding to each classified network device generates performance evaluation results for each network device of the same classification, thereby converting the problem of processing a single network device into the problem of processing a class of network devices, greatly improving operational efficiency. .

The implementation details of the network device performance evaluation method of this embodiment are described below. The following content is only provided for the convenience of understanding and is not necessary for implementation of this solution.

The network device performance evaluation method of this embodiment is specifically applied to servers, that is, the network device performance evaluation method is aimed at performance evaluation of servers.

The specific process of this implementation is shown in Figure 1, which specifically includes the following steps:

Step 101: Obtain performance evaluation data corresponding to each network device.

Specifically, in order to be able to reflect the performance of network equipment from multiple dimensions, the performance evaluation data corresponding to each network equipment obtained in this embodiment is specifically data for various different preset indicators.

Taking the network device as a CDN server as an example, the preset indicators in this implementation include business characteristic indicators, server performance indicators, and service quality indicators.

Correspondingly, the obtained performance evaluation data corresponding to each network device is the data corresponding to each preset indicator. For convenience of explanation, description will be made below with reference to Figure 2 .

As shown in Figure 2, in actual applications, the data corresponding to business characteristic indicators mainly include the number of http/https requests and IOPS (Input/Output Operations Per Second, the number of read and write operations per second) that can represent business characteristics; server The data corresponding to the performance indicators mainly include CPU (Central Processing Unit, central processing unit) usage, memory usage, disk usage (including system disk usage and single data disk usage), IO (Input/ Output, input/output) performance (specifically IOwait is obtained in this implementation); the data corresponding to the service quality indicator mainly includes the first packet response time that can represent the service quality.

In addition, regarding the IOwait mentioned above, it specifically refers to the process wait caused by the system IO.

It should be understood that the above is only an example and does not constitute any limitation to the technical solution of this embodiment. In practical applications, those skilled in the art can set preset indicators as needed, and the values corresponding to each preset indicator that need to be obtained. The data can be used as performance evaluation data. For example, the channel bandwidth representing the size of the service traffic can also be obtained. This implementation method does not place a limit on this.

In addition, it is worth mentioning that, in order to make the performance evaluation results generated based on the obtained performance evaluation data better reflect the performance of the network equipment and make the performance evaluation results more reasonable, this implementation method adds When performance evaluation data is obtained, the performance evaluation data of each network device in different time periods is obtained.

That is, in this embodiment, obtaining the performance evaluation data corresponding to each network device in step 101 specifically includes: based on the preset performance evaluation indicators, obtaining the performance evaluation data generated by each network device in different time periods. Performance evaluation data for indicators.

Furthermore, while ensuring the rationality of the performance evaluation results, in order to reduce the amount of subsequent data processing as much as possible, that is, to reduce the amount of data for performance evaluation data as much as possible to improve the processing speed, this implementation method is based on the preset performance evaluation indicators. , when obtaining the performance evaluation data for the performance evaluation indicators generated by each network device in different time periods, it is specifically obtained in the following ways:

Specifically, for each time period, the performance evaluation data of each network device at different time points within the time period is obtained, and the performance evaluation data at one of the time points is selected as the performance evaluation data of the time period.

That is to say, the performance evaluation data of each time period is represented by the performance evaluation data of one point in time within the time period.

In addition, since the size of the performance evaluation data will reflect the performance consumption of the network device, in order to make the subsequently generated performance evaluation results more accurately reflect the performance of the network device, this implementation method determines the performance evaluation data for each time period. Specifically, the performance evaluation data representing the time points of the corresponding time period is selected based on the size of the business traffic corresponding to different time points, that is, for each time period, the performance evaluation data corresponding to each network device at different time points within the time period is determined. The size of the business traffic, and then according to the size of the business traffic, select the performance evaluation data generated by each network device for the performance evaluation index at a time point when the size of the business traffic is the largest.

In order to facilitate understanding, the following is explained with examples:

For example, if the performance evaluation data of each network device needs to be obtained within a week, you can use 24 hours as a time period, that is, the performance evaluation data corresponding to each network device for seven days from Monday to Sunday is obtained.

Correspondingly, when obtaining the performance evaluation data corresponding to each day of the seven days from Monday to Sunday, one hour can be used as a time point, that is, each day will correspond to the performance evaluation data of 24 time points, and then based on these 24 times The business traffic size of the performance evaluation data at each point (for ease of explanation, hereafter referred to as: data volume size), from the performance evaluation data corresponding to 24 time points every day, select the performance evaluation data at one time point as the performance evaluation for that day. data.

Because usually, the greater the amount of performance evaluation data, the greater the corresponding performance consumption. Therefore, in order to know the overall performance of the network equipment when the performance consumption is maximum, when selecting the performance evaluation data at a point in time as the performance evaluation data of the day, specifically the performance evaluation corresponding to the time point when the size of the selected data is the maximum value. indicator data.

It should be understood that the above is only a specific implementation method. In actual applications, the collection granularity of performance evaluation data can be further refined, such as taking every minute as a collection cycle, and then selecting several consecutive The maximum value of the performance evaluation data corresponding to the collection period is used as the performance evaluation data for this time period. In this way, the performance evaluation data within a large time range is finally determined.

In addition, as can be seen from the above description, in this embodiment, the performance evaluation indicators include business characteristic indicators, equipment performance indicators and service quality indicators.

Therefore, when selecting the performance evaluation data generated by each network device for the performance evaluation index at the time point when the business traffic is the largest, specifically, selecting the performance evaluation data generated by each network device for the business characteristics at the time point when the business traffic is the largest. The performance evaluation data is obtained from the business characteristic data of the indicator, the equipment performance data for the equipment performance indicator, and the service quality data for the service quality indicator. That is, the obtained performance evaluation data includes business characteristic data, equipment performance data and service quality data.

Step 102: Obtain the hardware configuration information and content distribution network CDN application type corresponding to each network device.

Specifically, since the hardware configuration information and CDN application types obtained above are used to classify network devices, in practical applications, the above-mentioned hardware configuration information and CDN application types can be collectively referred to as network devices. Classified information.

Regarding the above-mentioned hardware configuration information, it specifically includes CPU model/number of cores, memory capacity, disk configuration (type, number), etc., which will not be listed one by one here, and this embodiment does not impose specific restrictions on this.

Step 103: Classify network devices with the same hardware configuration information and the same CDN application type into the same category according to the hardware configuration information and the CDN application type.

Specifically, in actual applications, the performance of network devices with different hardware configuration information will be very different. At the same time, in actual applications, there will be many types of CDN applications, and different CDN application types will correspond to different business characteristics. and business traffic. Therefore, in order to convert the problem of processing a single network device into the problem of processing a class of network devices, it is necessary to classify network devices with the same hardware configuration information and the same CDN application type into the same category, so as to avoid the subsequent confusion of different business characteristics. Interference between performance evaluation data.

In order to facilitate understanding, the following is explained with examples:

Assume that there are 10 network devices that need to be evaluated for performance. Among them, 5 network devices have CPU model a1, core number a2, memory capacity a3, disk configuration (type a4, number a5), and the other 5 The CPU model of the network device is b1, the number of cores is b2, the memory capacity is b3, and the disk configuration (type is b4, number is b5).

First, based on the first division based on hardware configuration information, the five network devices with CPU model a1, core number a2, memory capacity a3, and disk configuration (type a4, number a5) can be divided into one category , for the convenience of explanation, five network devices with CPU model b1, core number b2, memory capacity b3, and disk configuration (type b4, number b5) can be divided into one category. For convenience of explanation, it is referred to as category B below.

Assume that among Class A network devices, there are 3 network devices corresponding to the CDN application type CDN1, and 2 network devices corresponding to the CDN application type CDN2; among Class B network devices, there is 1 network device corresponding to the CDN application type CDN3, there are 4 network devices corresponding to the CDN application type CDN4.

Next, a second division is performed based on the CDN application type, and the three network devices corresponding to CDN1 in the Class A network equipment can be further divided into one category. For ease of explanation, they are hereinafter referred to as Class A1, and the two networks corresponding to CDN2 are The equipment is divided into one category. For ease of explanation, it is hereafter referred to as category A2.

Correspondingly, one network device corresponding to CDN3 among Class B network devices can be divided into one category. For ease of explanation, it is hereinafter referred to as category B1. Four network devices corresponding to CDN4 can be divided into one category. For ease of explanation, the following Called Category B2.

It should be understood that the above are only examples and do not constitute any limitation to the specific solutions of this embodiment. In practical applications, those skilled in the art can set device classification information and classification standards as needed. This embodiment does not Make restrictions.

Step 104: Generate performance evaluation results for each network device of the same category based on the performance evaluation data corresponding to each network device of the same category.

Specifically, since the obtained performance evaluation data is data corresponding to multiple preset indicators (business characteristic data corresponding to business characteristic indicators, equipment performance data corresponding to equipment performance indicators, and service quality data corresponding to service quality indicators), therefore When implementing the operation of step 104 above, specifically, the performance evaluation data corresponding to each network device under the same category can be aggregated according to the preset performance evaluation indicators, that is, the business characteristic data corresponding to the service characteristic indicators of each network device can be aggregated. As one category, the device performance data corresponding to the device performance indicators are aggregated into one category, the service quality data corresponding to the service quality indicators are aggregated into one category, and then the aggregated data calculations based on different preset indicators can reflect this type of network The performance evaluation value of the device performance change is then compared with the calculated performance evaluation value with the preset performance evaluation threshold, and finally based on the comparison result, a performance evaluation result for each network device of the same category is generated.

Regarding the above-mentioned performance evaluation values, in this embodiment, it mainly includes the peak value, average value and proportion corresponding to the performance evaluation index.

Correspondingly, the above-mentioned operation of calculating the performance evaluation value based on the aggregated performance evaluation data is as follows:

First, count the number of network devices that need to be inspected to obtain the total number of network devices;

Then, based on the data volume of the performance evaluation data corresponding to each network device under the same classification, determine the number of network devices that exceed the preset performance evaluation data volume threshold, and obtain the number of super-threshold network devices;

Next, calculate the proportion based on the total number of network devices and the number of super-threshold network devices;

Next, calculate the average value based on the aggregated performance evaluation data and the number of super-threshold network devices;

Next, the peak value is calculated based on the aggregated performance evaluation data and the total number of network devices.

In addition, it should be understood that since the performance evaluation indicators in this embodiment include business characteristic indicators, equipment performance indicators and service quality indicators, the peak values, average values and proportions mentioned above are essentially the peak values and average values corresponding to the business characteristic indicators. and proportion, the peak value, average value and proportion corresponding to the equipment performance indicators, and the peak value, average value and proportion corresponding to the service quality indicator.

In addition, since in practical applications, the above three performance evaluation indicators can be divided into multiple specific small indicators according to the situation, the above-mentioned peak value, average value and proportion actually correspond to the specific classification under each performance evaluation indicator. small indicators.

For ease of explanation, the several preset indicators and corresponding data listed in Figure 2 are still used as an example. The calculated performance evaluation value that can reflect the performance changes of the network equipment is roughly shown in Figure 3.

As shown in Figure 3, the performance evaluation values calculated based on the data corresponding to the business characteristic indicators mainly include the peak number of http/https requests, IOPS_r (number of read operations per second) peak and IOPS_w (number of read and write operations per second) ) peak value; the performance evaluation values calculated based on data corresponding to server performance indicators mainly include peak CPU usage, peak memory usage, peak system disk usage, peak data disk usage, and peak IOwait; based on data corresponding to service quality indicators The calculated performance evaluation values mainly include the proportion of timeout machines, average bandwidth and average number of requests.

Regarding the calculation of the peak value, average value and proportion mentioned above, it can be obtained by combining formula (1) to formula (4), as follows:

m＝|w%*n| (4)

In the formula, max _x is the _peak value calculated from the data corresponding _to each indicator, avg _x is the average value calculated from the data corresponding to each indicator, per Performance evaluation data of the device (which preset indicator is calculated is the data corresponding to the preset indicator), n is the total number of network devices, m is the number of machines whose performance indicator data exceeds the threshold, and w is the required "peak w value" .

Regarding the determination of m, it is specifically determined based on the selected "peak w value".

For example, _{by sorting} X _i from large to small, such as: _{X 1 ≥} The average value of the first 20% of the data from the largest to the smallest is then substituted into formula (4) based on the determined w and the current total number of network devices n, and the corresponding m can be obtained.

Then, after the above operations are completed, each calculated performance evaluation value that can reflect the performance change of the network device can be compared with the corresponding performance evaluation threshold, and then the performance of each network device of the same category is generated based on the comparison result. evaluation result.

It should be understood that in this implementation, the preset indicators are divided into business characteristic indicators, server performance indicators, and service quality indicators, and each type of indicator corresponds to different indicator data. Therefore, when configuring the performance evaluation threshold, you can configure the performance evaluation threshold according to the peak value, average value, and proportion divided in Figure 3.

Regarding the performance evaluation thresholds mentioned above, in this embodiment, they include: peak threshold, average threshold and proportion threshold.

In order to facilitate understanding, the performance evaluation threshold classifications divided in this embodiment are specifically shown in Figure 4.

As shown in Figure 4, for business characteristic indicators, you can configure the traffic coefficient range, http request number fluctuation threshold, https request number fluctuation threshold, IOPS_r fluctuation threshold, IOPS_w fluctuation threshold; for server performance indicators, you can configure the CPU fluctuation threshold and CPU peak value. Threshold range, memory usage fluctuation threshold, memory usage peak threshold range, system disk usage fluctuation threshold, system disk usage threshold range, data disk usage threshold range, IOwait peak threshold range, timeout first packet response time threshold, timeout The fluctuation threshold of the number of machines, the first packet response time range threshold, the average bandwidth fluctuation change threshold, and the average request data fluctuation change threshold.

Correspondingly, see Table 1 for details on the values of the above performance evaluation thresholds and the explanations of the corresponding indicators.

Table 1 Performance evaluation threshold configuration table

Correspondingly, when comparing each calculated performance evaluation value with the performance evaluation threshold, it is specifically determined whether each calculated performance evaluation value exceeds the corresponding performance evaluation threshold during the inspection period. The specific comparison logic is as follows:

Determine whether the peak value is greater than the peak value threshold;

If the peak value is not greater than the peak value threshold, determine whether the mean value is greater than the mean value threshold;

If the mean is not greater than the mean threshold, it is determined whether the proportion is greater than the proportion threshold.

In addition, it is worth mentioning that during the specific comparison process, the comparison between the performance evaluation value and the preset performance evaluation threshold is to remove and sum the corresponding maximum and minimum values in each indicator as The fluctuation value of the corresponding indicator is then compared with the corresponding threshold.

Correspondingly, the performance evaluation results for each network device of the same category are generated based on the comparison results, specifically as follows:

If the peak value is not greater than the peak threshold, the mean value is not greater than the mean threshold value, and the proportion is not greater than the proportion threshold value, then generate abnormal performance for each network device of the same category. evaluation result;

Otherwise, the performance evaluation data corresponding to the performance evaluation value that is greater than the preset performance evaluation threshold is abnormally marked to obtain the abnormal performance evaluation result for each network device of the same category.

For ease of understanding, the following provides a specific method for abnormally labeling performance evaluation data corresponding to performance evaluation values greater than the preset performance evaluation threshold, and obtaining the abnormal performance evaluation results for each network device of the same category. Condition:

For example, when the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the service characteristic data corresponding to the service characteristic indicator, the obtained abnormal performance evaluation results for each network device of the same category include the following: Any one or several items: http request volatility increases, https request volatility increases, IOPS volatility increases in the number of read and write operations per second.

For example, when the performance evaluation data corresponding to a performance evaluation value greater than the preset performance evaluation threshold is device performance data corresponding to a device performance indicator, the obtained performance evaluation results for abnormal network devices in the same category include the following: Any one or several items: increased volatility + a high proportion of days that are too high, increased volatility + a proportion of days that are too high, 0, increased volatility + a proportion of days that are too low, 0, decreased volatility + the number of days that are too high Proportion of 0, volatility decrease + too low number of days, proportion of 0, too low number of days.

For another example, when the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the service quality data corresponding to the service quality indicator, the obtained performance evaluation results for each network device of the same category that have abnormalities include: Any one or more of the following: an increase in the number of timeout machines, a decrease in the average bandwidth, and a decrease in the average number of requests.

In addition, regarding the generated performance evaluation results, in the specific implementation, in addition to displaying the above information, the CDN application type, hardware configuration information, and current performance evaluation data of network devices that exceed any of the above thresholds can also be displayed.

Taking the comparison of the fluctuation value of the number of http requests and the fluctuation threshold of the number of http requests as an example, when comparing the two, it is first necessary to determine the fluctuation value _ΔQPS of the number of http requests based on the peak value obtained by the above calculation.

Regarding the calculation of Δ _QPS , the details are as follows:

In the formula, N is the time to remove the maximum value and the minimum value. Taking one week as an example, N here is 5; The sum of the maximum and minimum values removed from the inspection end time (the end time of the obtained performance evaluation data) and the first N+1, that is, the 6-day max _QPS ;/> The sum of the maximum and minimum values is removed from the inspection start time (the time when the obtained performance evaluation data starts) and the previous N+1, that is, the max _QPS of 6 days.

It should be noted that in actual applications, the calculation of the fluctuation values corresponding to other indicators is roughly the same as the method of calculating the fluctuation value of the number of http requests given above, and will not be described again here.

Regarding the calculation of the usage fluctuation range corresponding to each indicator, the details are as follows:

In the formula, pre _max is the proportion of days with excessive usage, pre _min is the proportion of days with low usage, m is the number of days in the inspection period that are greater than the corresponding maximum usage, and i is less than the minimum corresponding usage. In the world, n is the total number of days in the inspection period.

Furthermore, the generated performance evaluation results can also be prioritized according to the urgency of processing, so that managers can prioritize network devices with serious problems after viewing the performance evaluation results.

It should be understood that the above are only examples and do not constitute any limitation on the technical solution of this embodiment. In practical applications, those skilled in the art can set the performance evaluation threshold as needed, and this embodiment does not impose any limitation on this.

It is easy to find from the above description that the network device performance evaluation method provided in this embodiment divides network devices with the same hardware configuration information and the same CDN application type into the same category, and then based on the performance evaluation data corresponding to the network devices of the same category To generate performance evaluation results for network equipment of the same category, thereby converting the problem of processing a single network device into the problem of processing a class of network equipment, greatly improving operational efficiency.

The second embodiment of the present invention relates to a network device performance evaluation method. The second embodiment is further improved on the basis of the first embodiment. The main improvement is that based on the performance evaluation data corresponding to each network device of the same classification, a performance evaluation data for each network device of the same classification is generated. After the performance evaluation results are obtained, a network device performance adjustment strategy for each network device of the same category is also generated based on the performance evaluation results.

For ease of understanding, detailed description is given below in conjunction with Figure 5:

Step 501: Obtain performance evaluation data corresponding to each network device.

Step 502: Obtain the hardware configuration information and content distribution network CDN application type corresponding to each network device.

Step 503: Classify network devices with the same hardware configuration information and the same CDN application type into the same category according to the hardware configuration information and the CDN application type.

Step 504: Generate performance evaluation results for each network device of the same category based on the performance evaluation data corresponding to each network device of the same category.

It is easy to find that steps 501 to 504 in this embodiment are substantially the same as steps 101 to 104 in the first embodiment, and will not be described again here.

Step 505: Based on the performance evaluation results, generate a network device performance adjustment policy for each network device of the same category.

For ease of explanation, this implementation provides several possible performance evaluation results for business characteristic indicators, server performance indicators and service quality indicators, as follows:

The performance evaluation results for business characteristic indicators can include any one or more of the following: http request fluctuation data, https request fluctuation data, IOPS fluctuation data; the performance evaluation results for server performance indicators can include the following: Any one or several items: CPU usage fluctuation data, proportion of days when CPU exceeds the threshold, memory usage fluctuation data, proportion of days when memory exceeds the threshold, disk usage fluctuation data, statistical data of disk usage exceeding the threshold, IOwait Statistical data on the number of days exceeding the threshold; performance evaluation results for service quality indicators can include any one or more of the following: fluctuation data in the proportion of timeout machines, average bandwidth fluctuation data, and average number of requests fluctuation data.

In other words, the final performance evaluation result generated is a combination of the performance evaluation results corresponding to each indicator.

In addition, it should be noted that in actual applications, as long as one of the above indicator items is abnormal, it will be displayed in the final performance evaluation results.

Correspondingly, the network device performance adjustment strategy for each network device of the same category that is finally generated based on the performance evaluation results can also be analyzed based on the performance evaluation results corresponding to each preset indicator.

Specifically, based on the above performance evaluation results included in the server performance indicators, the analysis results obtained can be roughly divided into the following six situations: ① Increased volatility + high proportion of excessively high days; ② Increased volatility + excessively high days The proportion is 0; ③ The volatility increases + the number of days that are too low accounts for 0; ④ The volatility decreases + the number of days that are too high accounts for 0; ⑤ The volatility decreases + the number of days that are too low accounts for 0; ⑥ The number of days that are too low accounts for 0.

Correspondingly, for the above situation ①, the generated network equipment performance adjustment strategies can be divided into the following three types: (1) improve the corresponding hardware configuration or reduce the rated equipment capacity; (2) optimize software performance; (3) adjust the business planning plan .

In response to the above situations ② to ⑤, the generated network device performance adjustment strategy may specifically be: continue to pay attention and do not process it for the time being.

In view of the above situation ⑥, the generated network device performance adjustment strategy can be: reducing the corresponding hardware configuration or increasing the rated device capability.

In addition, based on the above performance evaluation results included in the business characteristic indicators, the analysis results can be roughly divided into the following three situations: ① The fluctuation rate of the number of http requests increases; ② The fluctuation rate of the number of https requests increases; ③ The fluctuation rate of IOPS increases.

Correspondingly, for the above three situations, the generated network device performance adjustment strategy may specifically include: adjusting the service planning solution.

In addition, based on the above performance evaluation results included in the service quality indicators, the analysis results can be roughly divided into the following three situations: ① The proportion of timeout machines fluctuates and increases; ② The average bandwidth fluctuates and decreases; ③ The average number of requests fluctuates and decreases.

Correspondingly, for the above three situations, the generated network device performance adjustment strategy may be: improving the corresponding hardware configuration or reducing the rated device capability.

It is easy to find from the above description that in this embodiment, the network device performance adjustment strategy for each network device of the same category generated based on the performance evaluation results is specifically based on the "hardware configuration adjustment" and "rated device capability". The five directions are considered: "Adjustment", "Business planning adjustment", "Optimizing software performance adjustment", and "Continuous attention, not processing for now".

In addition, it should be understood that what is given above is only a network device performance adjustment strategy generated based on the performance evaluation results corresponding to each specific preset indicator. In practical applications, if the network equipment has problems corresponding to each of the above preset indicators, the generated network equipment performance adjustment strategy includes the network equipment performance adjustment strategy corresponding to each preset indicator.

Furthermore, in order to facilitate management personnel to view, in this embodiment, the generated performance evaluation results and network device performance adjustment strategies can be displayed in the form of charts. Specific formats can be set by those skilled in the art as needed. This embodiment No restrictions.

Therefore, the network device performance evaluation method provided in this embodiment generates performance evaluation results for each network device of the same classification based on the performance evaluation data corresponding to each network device of the same classification. As a result, network device performance adjustment strategies are automatically generated, so that network device performance can be adjusted without manual intervention, which not only reduces labor costs, but also further improves efficiency.

The steps of the various methods above are divided just for the purpose of clear description. During implementation, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent. ; Adding insignificant modifications or introducing insignificant designs to the algorithm or process without changing the core design of the algorithm and process are within the scope of protection of this patent.

The third embodiment of the present invention relates to a network equipment performance evaluation device, as shown in Figure 6, including: an acquisition module 601, a classification module 602 and an evaluation module 603.

Among them, the acquisition module 601 is used to obtain the performance evaluation data corresponding to each network device, as well as the hardware configuration information and content distribution network CDN application type corresponding to each network device; the classification module 602 is used to obtain the performance evaluation data corresponding to the hardware configuration information and the content distribution network CDN application type. The CDN application type divides network devices with the same hardware configuration information and the same CDN application type into the same category; the evaluation module 603 is used to generate each network device for the same category based on the performance evaluation data corresponding to each network device of the same category. Performance evaluation results of network equipment.

Furthermore, in another example, the acquisition module 601 is specifically configured to acquire, based on a preset performance evaluation index, performance evaluation data generated by each network device in different time periods for the performance evaluation index.

In addition, in another example, the acquisition module 601 is specifically configured to determine, for each time period, the business traffic size corresponding to each network device at different time points within the time period; and select the business traffic volume according to the business traffic size. Performance evaluation data for the performance evaluation index generated by each network device at the largest time point.

Furthermore, in another example, the performance evaluation indicators include business characteristic indicators, equipment performance indicators and service quality indicators.

Correspondingly, the acquisition module 601 is specifically used to select the business characteristic data for the service characteristic index, the device performance data for the device performance index and the service quality generated by each network device at the time point when the business traffic is the largest. The service quality data of the indicators are used to obtain the performance evaluation data.

In addition, in another example, the evaluation module 603 is specifically configured to aggregate the performance evaluation data corresponding to each network device under the same category according to preset performance evaluation indicators; and calculate a performance evaluation value based on the aggregated performance evaluation data. , the performance evaluation value can reflect the performance changes of each network device under the same category; compare the performance evaluation value with a preset performance evaluation threshold; generate a performance evaluation for each network device in the same category according to the comparison result result.

Furthermore, in another example, the performance evaluation value includes the peak value, average value and proportion corresponding to the performance evaluation index.

Accordingly, the evaluation module 603 is specifically used to count the number of network devices that need to be inspected to obtain the total number of network devices; based on the data size of the performance evaluation data corresponding to each network device under the same category, determine the amount of performance evaluation data that exceeds the preset The number of network devices at the threshold is used to obtain the number of super-threshold network devices; the proportion is calculated based on the total number of network devices and the number of super-threshold network devices; based on the aggregated performance evaluation data and the super-threshold network devices quantity, calculate the average value; calculate the peak value based on the aggregated performance evaluation data and the total number of network devices.

Furthermore, in another example, the performance evaluation threshold includes a peak threshold, an average threshold, and a proportion threshold.

Correspondingly, the evaluation module 603 is specifically configured to determine whether the peak value is greater than the peak threshold value; if the peak value is not greater than the peak threshold value, then determine whether the mean value is greater than the mean value threshold; if the mean value is not greater than the mean value threshold. If the mean threshold is determined, it is determined whether the proportion is greater than the proportion threshold.

When the peak value is not greater than the peak threshold, the mean value is not greater than the mean threshold, and the proportion is not greater than the proportion threshold, the evaluation module 603 is also configured to target each network device of the same category. Generate a performance evaluation result without abnormality; otherwise, the evaluation module 603 is also used to annotate the performance evaluation data corresponding to the performance evaluation value that is greater than the preset performance evaluation threshold to obtain the network devices of the same category. There are unusual performance evaluation results.

In addition, in another example, the evaluation module 603 is also configured to obtain the obtained performance evaluation data for the same category when the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the business characteristic data corresponding to the business characteristic indicator. Abnormal performance evaluation results of various network devices include any one or more of the following: increased http request volatility, increased https request volatility, increased IOPS volatility in the number of read and write operations per second; when the value is greater than the preset When the performance evaluation data corresponding to the performance evaluation value of the performance evaluation threshold is the device performance data corresponding to the device performance indicator, the obtained abnormal performance evaluation results for each network device of the same category include any one or more of the following: Increased volatility + a high proportion of days that are too high, increased volatility + a proportion of days that are too high, 0, increased volatility + a proportion of days that are too low, 0, decreased volatility + a proportion of days that are too high, 0, volatility decreases + The proportion of days that are too low is 0, the proportion of days that is too low; when the performance evaluation data corresponding to the performance evaluation value greater than the preset performance evaluation threshold is the service quality data corresponding to the service quality indicator, the obtained for the same category The abnormal performance evaluation results of each network device include any one or more of the following: increased fluctuations in the number of timeout machines, decreased fluctuations in average bandwidth, and decreased fluctuations in the average number of requests.

Furthermore, in another example, the network device performance evaluation apparatus further includes a policy generation module.

Correspondingly, the policy generation module is configured to generate a network device performance adjustment policy for each network device of the same category according to the performance evaluation result.

In addition, in another example, the policy generation module is specifically configured to perform the performance evaluation when the performance evaluation result is for the business characteristic indicator and includes any one or more of the following: increased http request volatility, https request volatility rate and IOPS fluctuation rate increase, the network device performance adjustment strategy generated for each network device of the same category is: adjusting the business planning plan; the performance evaluation results are based on device performance indicators, and include any of the following Item or items: increased volatility + 0 proportion of days that are too high, 0 increase in volatility + 0 proportion of days that are too low, 0 decrease in volatility + 0 proportion of days that are too high, 0 decrease in volatility + 0 proportion of days that are too low , the network device performance adjustment strategy generated for each network device of the same category is: continue to pay attention, and do not process it for the time being; the performance evaluation results are for device performance indicators, and include: increased volatility + excessively high number of days Compared with Gao, generating a network device performance adjustment strategy for each network device of the same category is any one or more of the following: improving corresponding hardware configuration, reducing rated device capabilities, optimizing software performance, and adjusting business planning solutions; in the The performance evaluation results are based on device performance indicators, and include: the proportion of days that are too low, and the network device performance adjustment strategy generated for each network device of the same category is any one or more of the following: improving the corresponding hardware configuration, reducing the rating Equipment capabilities; when the performance evaluation results are based on service quality indicators, and include any one or more of the following: an increase in the number of timeout machines, a decrease in the average bandwidth, and a decrease in the average number of requests, generate a report for the same The network device performance adjustment strategy for each classified network device is any one or more of the following: improving the corresponding hardware configuration and reducing the rated device capability.

It is not difficult to find that this embodiment is a device implementation corresponding to the first or second embodiment, and this embodiment can be implemented in cooperation with the first or second embodiment. The relevant technical details mentioned in the first or second embodiment are still valid in this embodiment, and will not be described again in order to reduce duplication. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the first or second embodiment.

It is worth mentioning that each module involved in this implementation is a logical module. In practical applications, a logical unit can be a physical unit, or a part of a physical unit, or it can be multiple physical units. The combination of units is realized. In addition, in order to highlight the innovative part of the present invention, units that are not closely related to solving the technical problems raised by the present invention are not introduced in this embodiment, but this does not mean that other units do not exist in this embodiment.

The fourth embodiment of the present invention relates to a device, as shown in Figure 7, including at least one processor 701; and a memory 702 communicatively connected to the at least one processor 701; wherein the memory 702 stores information that can be Instructions executed by the at least one processor 701, the instructions are executed by the at least one processor 701, so that the at least one processor 701 can execute the network device described in the first embodiment or the second embodiment. Performance evaluation methods.

The memory 702 and the processor 701 are connected using a bus. The bus may include any number of interconnected buses and bridges. The bus links various circuits of one or more processors 701 and the memory 702 together. The bus can also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are all well known in the art and therefore will not be described further herein. The bus interface provides the interface between the bus and the transceiver. A transceiver may be one element or may be multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. The data processed by the processor 701 is transmitted on the wireless medium through the antenna. Further, the antenna also receives the data and transmits the data to the processor 701 .

Processor 701 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory 702 may be used to store data used by the processor when performing operations.

Those skilled in the art can understand that all or part of the steps in the method of implementing the above embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium and includes several instructions to make a device (which can be a microcontroller) , chip, etc.) or processor (processor) executes all or part of the steps of the method described in each embodiment of the application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code.

Those of ordinary skill in the art can understand that the above-mentioned embodiments are specific implementations of the present invention, and in practical applications, various changes can be made in form and details without departing from the spirit and spirit of the present invention. scope.

Claims (11)

1. A method for evaluating performance of a network device, comprising:

acquiring performance evaluation data corresponding to each network device;

acquiring hardware configuration information and a content delivery network CDN application type corresponding to each network device;

dividing network devices with the same hardware configuration information and the same CDN application type into the same classification according to the hardware configuration information and the CDN application type;

generating performance evaluation results aiming at all network devices in the same category according to the performance evaluation data corresponding to all network devices in the same category;

the generating the performance evaluation result of each network device aiming at the same classification comprises the following steps:

when the performance evaluation data corresponding to the performance evaluation value larger than the preset performance evaluation threshold value is the service characteristic data corresponding to the service characteristic index, the obtained performance evaluation result of abnormality of each network device in the same category comprises any one or more of the following items: the http request fluctuation rate is increased, the https request fluctuation rate is increased, and the number of read-write operations per second IOPS fluctuation rate is increased;

When the performance evaluation data corresponding to the performance evaluation value larger than the preset performance evaluation threshold value is the equipment performance data corresponding to the equipment performance index, the obtained performance evaluation result of abnormality of each network equipment in the same category comprises any one or more of the following items: increasing the fluctuation rate and increasing the ratio of the number of days too high, increasing the fluctuation rate and increasing the ratio of the number of days too high to 0, increasing the fluctuation rate and increasing the ratio of the number of days too low to 0, decreasing the fluctuation rate and increasing the ratio of the number of days too high to 0, decreasing the fluctuation rate and increasing the ratio of the number of days too low to 0, and increasing the ratio of the number of days too low to 0;

when the performance evaluation data corresponding to the performance evaluation value larger than the preset performance evaluation threshold value is the service quality data corresponding to the service quality index, the obtained performance evaluation result of abnormality of each network device in the same class comprises any one or more of the following items: the time-out machine number duty ratio fluctuation is increased, the average bandwidth fluctuation is reduced, and the average request number fluctuation is reduced;

the performance evaluation value is calculated by the performance evaluation data corresponding to the network devices in the same class according to the data collected by the preset performance evaluation indexes and is used for reflecting the performance change of the network devices in the same class;

The performance evaluation index includes: business characteristic index, equipment performance index and service quality index.

2. The method for evaluating performance of a network device according to claim 1, wherein the obtaining performance evaluation data corresponding to each network device includes:

and acquiring performance evaluation data aiming at the performance evaluation indexes, which are generated by each network device in different time periods, based on the preset performance evaluation indexes.

3. The method for evaluating performance of network devices according to claim 2, wherein the obtaining performance evaluation data for the performance evaluation index generated by each network device in different time periods based on a preset performance evaluation index includes:

for each time period, determining the service flow sizes corresponding to different time points of each network device in the time period;

and selecting performance evaluation data aiming at the performance evaluation index and generated by each network device at the time point with the maximum service flow according to the service flow.

4. The network device performance evaluation method of claim 3, wherein the performance evaluation index comprises a traffic characteristic index, a device performance index, and a quality of service index;

And selecting performance evaluation data aiming at the performance evaluation index and generated by each network device at the time point with the maximum service flow, wherein the performance evaluation data comprises the following components:

and selecting service characteristic data aiming at the service characteristic index, equipment performance data aiming at the equipment performance index and service quality data aiming at the service quality index, which are generated by each network equipment at the time point with the maximum service flow, to obtain the performance evaluation data.

5. The method for evaluating performance of network devices according to claim 4, wherein generating the performance evaluation result for each network device of the same class according to the performance evaluation data corresponding to each network device of the same class comprises:

the performance evaluation data corresponding to each network device are converged according to a preset performance evaluation index under the same classification;

calculating a performance evaluation value based on the converged performance evaluation data, wherein the performance evaluation value can reflect the performance change of each network device under the same classification;

comparing the performance evaluation value with a preset performance evaluation threshold value;

and generating performance evaluation results of the network devices aiming at the same class according to the comparison results.

6. The network device performance evaluation method according to claim 5, wherein the performance evaluation value includes a peak value, a mean value, and a duty ratio corresponding to the performance evaluation index;

the calculating a performance evaluation value based on the aggregated performance evaluation data includes:

counting the number of network devices needing to be inspected to obtain the total number of the network devices;

determining the number of network devices exceeding a preset performance evaluation data quantity threshold based on the service flow of the performance evaluation data corresponding to each network device under the same classification to obtain the number of super-threshold network devices;

calculating the duty cycle based on the total number of network devices and the super-threshold number of network devices;

calculating the mean value based on the converged performance evaluation data and the super-threshold network equipment number;

and calculating the peak value based on the aggregated performance evaluation data and the total number of network devices.

7. The network device performance evaluation method of claim 6, wherein the performance evaluation threshold comprises a peak threshold, a mean threshold, and a duty cycle threshold;

the comparing the performance evaluation value with a preset performance evaluation threshold value comprises the following steps:

Judging whether the peak value is larger than the peak value threshold value or not;

if the peak value is not greater than the peak value threshold value, judging whether the average value is greater than the average value threshold value or not;

if the average value is not greater than the average value threshold value, judging whether the duty ratio is greater than the duty ratio threshold value or not;

wherein the generating, according to the comparison result, a performance evaluation result for each network device of the same class includes:

if the peak value is not greater than the peak value threshold value, the average value is not greater than the average value threshold value, and the duty ratio is not greater than the duty ratio threshold value, generating a performance evaluation result without abnormality for each network device in the same classification;

otherwise, performing anomaly labeling on the performance evaluation data corresponding to the performance evaluation value larger than the preset performance evaluation threshold to obtain the performance evaluation result of the anomaly of each network device in the same category.

8. The network device performance evaluation method according to claim 1, wherein after the generating of the performance evaluation result for each network device of the same class based on the comparison result, the method further comprises:

and generating a network device performance adjustment strategy aiming at the network devices of the same class according to the performance evaluation result.

9. The network device performance evaluation method according to claim 8, wherein the generating a network device performance adjustment policy for each network device of the same class according to the performance evaluation result comprises:

and when the performance evaluation result is specific to the service characteristic index, any one or more of the following items are included: the http request fluctuation rate is increased, the https request fluctuation rate is increased, the IOPS fluctuation rate is increased, and the network device performance adjustment strategy for each network device of the same class is generated as follows: adjusting a service planning scheme;

and the performance evaluation result is specific to the equipment performance index and comprises any one or more of the following: the network device performance adjustment policies for the network devices of the same class are generated by increasing the fluctuation rate, occupying the ratio of the excessive days to 0, decreasing the fluctuation rate, occupying the ratio of the excessive days to 0: continuing to pay attention to, and temporarily not processing;

when the performance evaluation result is specific to the equipment performance index, the method comprises the following steps: the ratio of the fluctuation rate to the number of days which is too high is high, and the network equipment performance adjustment strategy of each network equipment aiming at the same class is generated as any one or more of the following: corresponding hardware configuration is improved, rated equipment capacity is reduced, software performance is optimized, and a service planning scheme is adjusted;

When the performance evaluation result is specific to the equipment performance index, the method comprises the following steps: generating a network device performance adjustment policy for each network device of the same class for the excessively low number of days of occupancy, the network device performance adjustment policy being any one or more of: the corresponding hardware configuration is improved, and the rated equipment capacity is reduced;

and when the performance evaluation result is specific to the service quality index, any one or more of the following items are included: the overtime machine number duty ratio fluctuation is increased, the average bandwidth fluctuation is reduced, the average request number fluctuation is reduced, and the network equipment performance adjustment strategy of each network equipment aiming at the same class is generated as any one or more of the following: and the corresponding hardware configuration is improved, and the rated equipment capacity is reduced.

10. A network device performance evaluation device, comprising:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the network device performance assessment method of any one of claims 1 to 9.

11. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the network device performance evaluation method of any one of claims 1 to 9.