CN106021084A - Method and apparatus for testing server performance - Google Patents

Abstract

The present invention provides a method and apparatus for testing server performance. The method comprises the steps of first obtaining current configuration information of a server, generating a corresponding tuning scheme, and outputting the current configuration information and the tuning scheme, thereby facilitating check of workers; updating a configuration file of a test tool according to the current configuration information if the workers choose not to change the current configuration environment after checking is carried out, and then using the test tool to test the server performance; and obtaining changed configuration information again if the workers choose to change the current configuration environment, and executing the same subsequent operation steps. Due to the fact that implementation processes of test environment parameters determination, test environment creation and test performing can be automatically carried out without participation of the workers, manual operation faults can be avoided. Accordingly, accuracy of server performance test results can be improved.

Description

A method and device for performing server performance testing

technical field

The invention relates to the field of computer technology, in particular to a method and device for performing server performance testing.

Background technique

The server is an important foundation of informatization construction, and the performance of the server determines the performance of the entire application system. At present, there are many indicators for measuring server performance, such as the ability of the server to perform floating-point operations.

When a server performance test is required, the staff can view the actual configuration of the server, and use a large number of calculations to determine the optimal test parameters, and perform subsequent test processes based on the test parameters.

Since it is necessary to manually determine the environmental parameters and build the test environment, and human errors are usually unavoidable, the existing implementation methods will reduce the accuracy of the server performance test results.

Contents of the invention

The invention provides a method and device for testing server performance, which can improve the accuracy of server performance testing results.

In order to achieve the above object, the present invention is achieved through the following technical solutions:

On the one hand, the present invention provides a kind of method that carries out server performance test, pre-determined test tool, also includes:

S1: Obtain the current configuration information of the server, and generate an optimization solution corresponding to the current configuration information;

S2: Outputting the current configuration information and the optimization scheme;

S3: Execute S4 when receiving the input stop execution control instruction, and execute S5 when receiving the input continue execution control instruction;

S4: Execute S1 when the input server update completion instruction is received;

S5: Update the configuration file of the test tool according to the current configuration information;

S6: According to the updated configuration file, use the test tool to test.

Further, the current configuration information includes: total memory capacity, total system threads, database type, NUMA (Non Uniform Memory Access Architecture, non-uniform memory access architecture), IO (Input/Output, input/output) of the operating system Any one or more of free ports.

Further, the current configuration information includes: total memory capacity, total system thread number, database type, NUMA, IO free ports of the operating system;

The S5 includes: calculating and obtaining the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, and the NUMA value corresponding to the NUMA according to the current configuration information. RFACTs value, the threshold value corresponding to the IO free port of the operating system; the obtained Ns value, the PXQ value, the NB value, the RFACTs value, and the threshold value are written into the test tool in the configuration file.

Further, after S5, it also includes: clearing the original log file corresponding to the server;

After S6 also includes:

Generate corresponding test results, and save the test results;

Generate a corresponding current log file based on the cleared original log file;

Judging whether there is error information in the current log file, if so, outputting the error information.

Further, the original log file includes:

The kernel log file /var/log/messages, the hardware monitoring log file /var/log/mcelog, and the test tool log file /opt/intel/mp_linpack/ia64/log/xhpl.log.

Further, the method also includes: predetermining the ./fpo_qhl.sh test script corresponding to the floating-point operation;

The S1 includes: using a terminal to run the test script, obtain current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information.

In another aspect, the present invention provides a device for performing server performance testing, including:

Determining the unit, used to determine the test tool;

The first processing unit is configured to acquire the current configuration information of the server, and generate an optimization scheme corresponding to the current configuration information; and output the current configuration information and the optimization scheme, and trigger the first communication unit;

The first communication unit is configured to trigger the second communication unit when receiving the input stop execution control instruction, and trigger the update unit when receiving the input continue execution control instruction;

The second communication unit is configured to trigger the first processing unit when receiving an input server update completion instruction;

The update unit is configured to update the configuration file of the test tool according to the current configuration information;

The testing unit is configured to use the testing tool to perform testing according to the updated configuration file.

Further, the current configuration information includes: total memory capacity, total system thread number, database type, NUMA, IO free ports of the operating system;

The update unit is specifically configured to calculate and obtain the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, and the NUMA value according to the current configuration information. Corresponding RFACTs value, the threshold value corresponding to the IO idle port of the operating system; the obtained Ns value, the PXQ value, the NB value, the RFACTs value, and the threshold value are written into the in the configuration file of the testing tool.

Further, the device for performing server performance testing also includes:

a clearing unit, configured to clear the original log file corresponding to the server;

The second processing unit is configured to generate a corresponding test result and save the test result; and generate a corresponding current log file based on the cleared original log file;

The third processing unit is configured to determine whether there is error information in the current log file, and if so, output the error information.

Further, the determination unit is also used to determine the ./fpo_qhl.sh test script corresponding to the floating-point operation;

The first processing unit is specifically configured to use a terminal to run the test script, obtain current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information.

The present invention provides a method and device for server performance testing. First, obtain the current configuration information of the server, generate a corresponding optimization plan, and output the current configuration information and the optimization plan so that the staff can check it; If the personnel choose not to change the current configuration environment after viewing, they will update the configuration file of the test tool according to the current configuration information, and then use the test tool for testing; if the staff chooses to change the current configuration environment, they will reacquire the changed configuration information, and Perform the same subsequent steps. Since the implementation process of determining the test environment parameters, building the test environment and conducting the test can be carried out automatically without the participation of staff, thereby avoiding human error. Therefore, the present invention can improve the accuracy of server performance test results.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are For some embodiments of the present invention, those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of a method for performing server performance testing provided by an embodiment of the present invention;

FIG. 2 is a flow chart of another method for performing server performance testing provided by an embodiment of the present invention;

Fig. 3 is a schematic diagram of a device for performing a server performance test provided by an embodiment of the present invention;

Fig. 4 is a schematic diagram of another device for performing a server performance test provided by an embodiment of the present invention.

detailed description

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the protection of the present invention. scope.

As shown in Figure 1, the embodiment of the present invention provides a method for performing server performance testing, which may include the following steps:

Step 101: Predetermine the test tool.

Step 102: Obtain current configuration information of the server, and generate an optimization solution corresponding to the current configuration information.

Step 103: Output the current configuration information and the optimization scheme.

Step 104: Execute step 105 when receiving the input stop execution control instruction, and execute step 106 when receiving the input continuation execution control instruction.

Step 105: Execute step 102 when the input server update completion instruction is received.

Step 106: Update the configuration file of the test tool according to the current configuration information.

Step 107: According to the updated configuration file, use the test tool to test.

The embodiment of the present invention provides a method for performing server performance testing. Firstly, the current configuration information of the server is obtained, a corresponding tuning plan is generated, and the current configuration information and tuning plan are output, so that the staff can check them; If the personnel choose not to change the current configuration environment after viewing, they will update the configuration file of the test tool according to the current configuration information, and then use the test tool for testing; if the staff chooses to change the current configuration environment, they will reacquire the changed configuration information, and Perform the same subsequent steps. Since the implementation process of determining the test environment parameters, building the test environment and conducting the test can be carried out automatically without the participation of staff, thereby avoiding human error. Therefore, the present invention can improve the accuracy of server performance test results.

In an embodiment of the present invention, the current configuration information may include: any one or more of total memory capacity, total system thread count, database type, NUMA, and IO free ports of the operating system.

In an embodiment of the present invention, in order to illustrate the updating method of a configuration file of a testing tool, so,

The current configuration information includes: the total memory capacity, the total number of threads in the system, the database type, NUMA, the IO idle ports of the operating system;

The step 106 includes: according to the current configuration information, calculate and obtain the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, and the NUMA corresponding The RFACTs value, the threshold value corresponding to the IO idle port of the operating system; the obtained Ns value, the PXQ value, the NB value, the RFACTs value, and the threshold value are written into the test in the tool's configuration file.

In one embodiment of the present invention, in order to illustrate the subsequent implementation process of testing with testing tools, therefore,

After step 106, further comprising: clearing the original log file corresponding to the server;

After step 107, further comprising: generating corresponding test results, and saving the test results; based on the cleared original log file, generating a corresponding current log file; judging whether there is error information in the current log file, If yes, output the error message.

In one embodiment of the present invention, the original log file includes: kernel log file /var/log/messages, hardware monitoring log file /var/log/mcelog and test tool log file /opt/intel/mp_linpack/ia64/log /xhpl.log.

In an embodiment of the present invention, the method may further include: predetermining the ./fpo_qhl.sh test script corresponding to the floating-point operation;

The step 102 includes: using a terminal to run the test script, obtain current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information.

As shown in Figure 2, an embodiment of the present invention provides a method for performing server performance testing, which may include the following steps:

Step 201: Determine the test tool, and determine the test script corresponding to the floating point operation.

Since a very important indicator for measuring server performance is the ability of the server to perform floating-point calculations, in this embodiment, the method for performing server performance testing proposed in this solution can be used to evaluate the floating-point calculation performance of the server. test.

Before testing, the corresponding test tools and test scripts can be determined first. Wherein, the test script may be the ./fpo_qhl.sh test script, and this test process may be performed based on a shell script under the Linux system.

Step 202: use the terminal to run the test script, obtain the current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information.

In detail, the current configuration information may include any one or more of the total memory capacity, the total number of system threads, database type, NUMA, and IO free ports of the operating system.

For example, in this embodiment, the terminal is used to run the ./fpo_qhl.sh test script to obtain the current total memory capacity of the server and the total number of system threads, that is, the hyperthreading is enabled and the memory is fully configured.

Based on this current configuration information, it can be determined through calculation that the optimization scheme corresponding to the current configuration information of the server is hyper-threading off and the memory is 1/2 configuration.

Run the test script through the terminal, the terminal can have 3 prompts, namely:

A. The current configuration of the server is as follows;

B. The tuning configuration scheme is as follows;

C. Whether to continue testing according to the current configuration.

Corresponding to the above prompt A, in one embodiment of the present invention, the main execution code corresponding to this prompt may be as follows:

Corresponding to the above prompt B and prompt C, in one embodiment of the present invention, the main execution code corresponding to this prompt may be as follows:

NODE number: 1

Sockets number: 1

memory policy: 1/3DPS

BIOS setup:

HT: disable

EIST: enable

Turbo: enable

T/C state: disable

Energy performance: performance

Do you want to continue using this configuration(Y/N)

Step 203: output current configuration information and an optimization solution.

When the current configuration information of the server and the corresponding optimization scheme are determined, they can be output, for example, displayed on a display interface, so that the staff can analyze and view them.

In an embodiment of the present invention, a group of function buttons, such as YES/NO, may be provided on the display interface at the same time, to prompt the user to choose whether to perform the test according to the current configuration environment. Assuming that the staff chooses to rebuild the server configuration environment at this stage, they can click the NO button.

Step 204: Execute step 205 when receiving the input stop execution control instruction, and execute step 206 when receiving the input continuation execution control instruction.

When the worker clicks the NO button to input the stop execution control command, the system can automatically shut down the server. According to the above tuning scheme, the staff can turn off hyperthreading and change the memory configuration to 1/2 configuration to complete the rebuilding of the test environment.

In this embodiment, the system can recommend a corresponding optimization solution according to the current configuration information of the server, so that the staff can choose according to actual needs. This implementation method does not require staff to manually obtain server configuration information, and does not need to manually calculate the optimal test environment parameters, thereby saving human resources and time investment, and avoiding errors that may be caused by human operations, so it can be improved. The accuracy and authority of floating-point performance test results.

Step 205: Execute step 202 when the input server update completion instruction is received.

After rebuilding the test environment, the staff can start the server, that is, the system can obtain the server update completion instruction input by the staff, so that step 202 can be executed again, that is, the current configuration information of the server can be obtained again.

According to the above description, it can be seen that the current configuration information obtained again can be hyper-threading off and memory 1/2 configuration, and the corresponding tuning solution is the optimal current configuration environment. Therefore, the worker can click the YES button on the display interface to input the control instruction to continue execution, so that step 206 can be executed. In a possible implementation manner, if the worker clicks the NO button at this time, the test script ends.

Step 206: Calculate and obtain corresponding test parameter values according to the current configuration information.

Specifically, the test parameter value may include the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, the RFACTs value corresponding to NUMA, and the threshold value corresponding to the IO idle port of the operating system. Any one or more.

Wherein, for RFACTs value, 0=left, 1=Crout, 2=Right.

In this embodiment, according to the total memory capacity and the total number of system threads in the acquired current configuration information of the server, the corresponding Ns value and PXQ value can be calculated and obtained respectively. When the test tool executes the test, it can perform the test according to the Ns value and the PXQ value.

Step 207: Write the test parameter value into the configuration file of the test tool, and clear the original log file corresponding to the server.

When the calculation of test parameter values is completed, the HPL configuration file of the test tool can be updated according to the calculated Ns and PXQ values. Specifically, the Ns value and the PXQ value can be automatically written into the configuration file of the test tool for corresponding update and replacement.

In one embodiment of the present invention, some original alarm information can be recorded in the original log file corresponding to the server. Since the original alarm information may not match the current server configuration environment, the original log can be cleared before testing. file to avoid false alarm conditions.

In detail, the original log file can be:

The kernel log file /var/log/messages, the hardware monitoring log file /var/log/mcelog, and the test tool log file /opt/intel/mp_linpack/ia64/log/xhpl.log.

In the embodiment of the present invention, by running the script, the relevant parameters of the test environment can be calculated according to the server configuration information, the configuration file can be updated according to the relevant parameters, and the log information can be cleared. In detail, the main execution code corresponding to this implementation process can be as follows:

Step 208: According to the updated configuration file, use the test tool to test, and save the generated test result.

When the configuration file of the testing tool is updated, the testing tool can be used to perform automatic testing to test the floating-point computing performance of the server. And when the test is completed, the generated test results can be saved. Of course, the system can also display the generated test results.

In this embodiment, the test is started according to the configuration file, and the test results are automatically collected. The main execution code corresponding to this implementation process can be as follows:

echo OMP_NUM_THREADS=$NUMPROC numactl

--physcpubind=${cpuset[NUMSOCK]} --interleave=all ./xhpl_intel64|tee-a$runlog

OMP_NUM_THREADS=$NUMPROC numactl

--physcpubind=${cpuset[NUMSOCK]}

--interleave=all./xhpl_offload_intel64|

Step 209: Generate a corresponding current log file based on the cleared original log file.

During the test process, on the basis of the cleared original log file, a log file corresponding to the current test process can be generated.

Step 210: Determine whether there is error information in the current log file, and if so, output the error information.

For the generated current log file, the system can automatically judge whether there is error information in it, and output the judged error information as a prompt, so that the staff can carry out corresponding analysis and processing.

In this embodiment, after the test is completed, the log file can be automatically detected, if there is error and other related information, a prompt will be given, and the test result will be automatically and synchronously saved during the test.

The corresponding main execution code for checking log information can be as follows:

cat /var/log/messages grep -i error

cat /var/log/messages grep -i temperature

cat /var/log/messages grep -i warning

The corresponding main execution code of automatic synchronous saving of results can be as follows:

tee-a$runlog

temp = `grep 'WC06C2C4'$runlog|awk'{print $NF}'`

echo $ttime, $MACHINECONF, $NUMPROC, ${NUMSOCK}-Socket, $psize, $temp>>$resfile

In this embodiment, the shell script can be used to realize the setting up and testing of the server floating-point computing test environment, which is convenient to use and easy to operate, and can quickly and accurately complete the test without experienced staff, so it is beneficial to reduce the cost of human resources. A lot of investment, improve work efficiency.

As shown in Figure 3, an embodiment of the present invention provides a device for performing server performance testing, including:

A determining unit 301, configured to determine a test tool;

The first processing unit 302 is configured to acquire the current configuration information of the server, and generate an optimization scheme corresponding to the current configuration information; and output the current configuration information and the optimization scheme, and trigger the first communication unit 303;

The first communication unit 303 is configured to trigger the second communication unit 304 when receiving the input stop execution control instruction, and trigger the update unit 305 when receiving the input continue execution control instruction;

The second communication unit 304 is configured to trigger the first processing unit 302 when receiving an input server update completion instruction;

The updating unit 305 is configured to update the configuration file of the test tool according to the current configuration information;

The testing unit 306 is configured to use the testing tool to perform testing according to the updated configuration file.

In one embodiment of the present invention, the current configuration information includes: the total memory capacity, the total number of threads in the system, the database type, NUMA, the IO free ports of the operating system;

The update unit 305 is specifically configured to calculate and obtain the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, and the The RFACTs value corresponding to NUMA, the threshold value corresponding to the IO free port of the operating system; the obtained Ns value, the PXQ value, the NB value, the RFACTs value, and the threshold value are written into the in the configuration file of the test tool described above.

In one embodiment of the present invention, please refer to FIG. 4, the device for performing server performance testing may also include:

A clearing unit 401, configured to clear the original log file corresponding to the server;

The second processing unit 402 is configured to generate a corresponding test result and save the test result; and generate a corresponding current log file based on the cleared original log file;

The third processing unit 403 is configured to determine whether there is error information in the current log file, and if so, output the error information.

In one embodiment of the present invention, the determination unit 301 is further configured to determine the ./fpo_qhl.sh test script corresponding to the floating-point operation;

The first processing unit 302 is specifically configured to use a terminal to run the test script, obtain current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information.

The information exchange and execution process among the units in the above-mentioned device are based on the same concept as the method embodiment of the present invention, and the specific content can refer to the description in the method embodiment of the present invention, and will not be repeated here.

In summary, each embodiment of the present invention has at least the following beneficial effects:

1. In the embodiment of the present invention, first obtain the current configuration information of the server, generate a corresponding tuning plan, and output the current configuration information and tuning plan, so that the staff can check it; if the staff chooses not to change it after checking For the current configuration environment, update the configuration file of the test tool according to the current configuration information, and then use the test tool for testing; if the staff chooses to change the current configuration environment, re-acquire the changed configuration information, and perform the same follow-up steps. Since the implementation process of determining the test environment parameters, building the test environment and conducting the test can be carried out automatically without the participation of staff, thereby avoiding human error. Therefore, the present invention can improve the accuracy of server performance test results.

2. In the embodiment of the present invention, the system can recommend a corresponding optimization solution according to the current configuration information of the server, so that the staff can choose according to actual needs. This implementation method does not require staff to manually obtain server configuration information, and does not need to manually calculate the optimal test environment parameters, thereby saving human resources and time investment, and avoiding errors that may be caused by human operations, so it can be improved. The accuracy and authority of floating-point performance test results.

3. In the embodiment of the present invention, the shell script can be used to realize the setting up and testing of the server floating-point computing test environment, which is convenient to use and simple to operate, and the test can be completed quickly and accurately without the need for experienced staff, so it is beneficial to reduce manpower A large amount of resources are invested to improve work efficiency.

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is a relationship between these entities or operations. There is no such actual relationship or sequence. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional same elements in the process, method, article or apparatus comprising said element.

Those of ordinary skill in the art can understand that all or part of the steps to realize the above method embodiments can be completed by program instructions related hardware, and the aforementioned programs can be stored in a computer-readable storage medium. When the program is executed, the It includes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Finally, it should be noted that the above descriptions are only preferred embodiments of the present invention, and are only used to illustrate the technical solution of the present invention, and are not used to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present invention are included in the protection scope of the present invention.

Claims (10)

1. A method for performing server performance testing, characterized in that, pre-determining testing tools, also comprising: S1: Obtain the current configuration information of the server, and generate an optimization solution corresponding to the current configuration information; S2: Outputting the current configuration information and the optimization scheme; S3: Execute S4 when receiving the input stop execution control instruction, and execute S5 when receiving the input continue execution control instruction; S4: Execute S1 when the input server update completion instruction is received; S5: Update the configuration file of the test tool according to the current configuration information; S6: According to the updated configuration file, use the test tool to test. 2. The method of claim 1, wherein, The current configuration information includes: any one or more of: total memory capacity, total number of system threads, database type, non-uniform memory access architecture NUMA, and free input/output IO ports of the operating system. 3. The method of claim 1, wherein, The current configuration information includes: the total memory capacity, the total number of threads in the system, the database type, NUMA, the IO idle ports of the operating system; The S5, including: According to the current configuration information, calculate and obtain the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, the RFACTs value corresponding to the NUMA, the operation The threshold value corresponding to the IO idle port of the system; Write the obtained Ns value, the PXQ value, the NB value, the RFACTs value, and the threshold value into the configuration file of the test tool. 4. The method of claim 1, wherein, After S5, further comprising: clearing the original log file corresponding to the server; After S6, further includes: Generate corresponding test results, and save the test results; Generate a corresponding current log file based on the cleared original log file; Judging whether there is error information in the current log file, if so, outputting the error information. 5. The method according to claim 4, wherein the original log files include: kernel log files/var/log/messages, hardware monitoring log files/var/log/mcelog and test tool log files/opt/ intel/mp_linpack/ia64/log/xhpl.log. 6. The method according to any one of claims 1 to 5, characterized in that, It further includes: predetermining the ./fpo_qhl.sh test script corresponding to the floating-point operation; The S1 includes: using a terminal to run the test script, obtain current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information. 7. A device for performing server performance testing, comprising: Determining the unit, used to determine the test tool; The first processing unit is configured to acquire the current configuration information of the server, and generate an optimization scheme corresponding to the current configuration information; and output the current configuration information and the optimization scheme, and trigger the first communication unit; The first communication unit is configured to trigger the second communication unit when receiving the input stop execution control instruction, and trigger the update unit when receiving the input continue execution control instruction; The second communication unit is configured to trigger the first processing unit when receiving an input server update completion instruction; The update unit is configured to update the configuration file of the test tool according to the current configuration information; The testing unit is configured to use the testing tool to perform testing according to the updated configuration file. 8. The device for performing server performance testing according to claim 7, characterized in that, The current configuration information includes: the total memory capacity, the total number of threads of the system, the database type, the non-uniform memory access architecture NUMA, the input/output IO idle ports of the operating system; The update unit is specifically configured to calculate and obtain the Ns value corresponding to the total memory capacity, the PXQ value corresponding to the total number of system threads, the NB value corresponding to the database type, and the NUMA value according to the current configuration information. Corresponding RFACTs value, the threshold value corresponding to the IO idle port of the operating system; the obtained Ns value, the PXQ value, the NB value, the RFACTs value, and the threshold value are written into the in the configuration file of the testing tool. 9. The device for performing server performance testing according to claim 7, further comprising: a clearing unit, configured to clear the original log file corresponding to the server; The second processing unit is configured to generate a corresponding test result and save the test result; and generate a corresponding current log file based on the cleared original log file; The third processing unit is configured to determine whether there is error information in the current log file, and if so, output the error information. 10. The device for performing server performance testing according to any one of claims 7 to 9, characterized in that, The determination unit is further used to determine the ./fpo_qhl.sh test script corresponding to the floating-point operation; The first processing unit is specifically configured to use a terminal to run the test script, obtain current configuration information of the server by running the test script, and generate an optimization solution corresponding to the current configuration information.