Performance test method and device, electronic equipment and medium

Abstract

The embodiment of the invention provides a performance test method and device, electronic equipment and a medium. According to the method and the device, performance analysis is carried out on the server to obtain BIOS attributes and OS attributes which influence the performance of the server, on the basis, one attribute group under one attribute (namely, a first attribute) is fixed, an optimal attribute group under the other attribute (namely, a second attribute) is sought, and the optimal attribute group of the second attribute is fixed, so that the performance of the server is improved. Compared with a mode of manually determining the optimal parameter of the server performance, the mode of automatically obtaining the optimal parameter group enabling the server performance to be the optimal performance is faster, the real-time performance of obtaining the optimal parameter group can be ensured, and the efficiency of obtaining the optimal parameter group is improved. The problem that the real-time performance is poor due to the fact that the optimal server performance parameters are manually determined is solved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to performance testing methods, devices, electronic equipment and media. Background Art

[0002] In recent years, with the development of computer technology, the overall performance requirements of servers have become increasingly higher. Therefore, it is very important to test the performance of the entire server and determine the parameters that make the server performance in the optimal state based on the test results.

[0003] Currently, after obtaining the test results, engineers are required to determine the optimal parameters of server performance based on experience. However, this manual determination of the optimal parameters of server performance is slow and has the problem of poor real-time performance. Summary of the invention

[0004] In view of this, the embodiments of the present application provide a performance testing method, device, electronic device and medium to enable rapid acquisition of optimal server performance parameters, so as to solve the problem of poor real-time performance in manually determining the optimal server performance parameters.

[0005] The present application provides a performance testing method, which includes:

[0006] By analyzing the performance of the server, two types of attributes that affect the server performance are obtained: basic input and output system BIOS attributes and operating system OS attributes; BIOS attributes have corresponding BIOS attribute group sets; BIOS attribute group combinations include multiple attribute groups; any BIOS attribute group includes multiple different types of parameters that affect server performance; OS attributes have corresponding OS attribute group sets; OS attribute group sets include multiple OS attribute groups; any OS attribute group includes multiple different types of parameters that affect server performance;

[0007] The server is configured based on the first attribute group under the fixed first attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test respectively under the configuration of the server based on each attribute group under the second attribute; and the optimal attribute group under the second attribute is determined based on the test performance results corresponding to each attribute group; one attribute of the BIOS attribute and the OS attribute is used as the first attribute, and the other attribute is used as the second attribute; the first attribute group is an attribute group selected from the attribute group set corresponding to the first attribute;

[0008] The server is configured based on the optimal attribute group under the fixed second attribute to obtain the test performance results corresponding to each attribute group by performing SPECCPU test on each attribute group under the configuration; and the optimal attribute group under the first attribute is determined based on the test performance results corresponding to each attribute group; the optimal attribute group under the first attribute and the optimal attribute group under the second attribute constitute the optimal parameter group corresponding to the server.

[0009] The present application also provides a server performance optimization device, the device comprising:

[0010] A performance testing device, characterized in that the device comprises:

[0011] The acquisition module is used to obtain two types of attributes that affect the performance of the server by performing performance analysis on the server: basic input and output system BIOS attributes and operating system OS attributes; the BIOS attributes have corresponding BIOS attribute group sets; the BIOS attribute group combination includes multiple attribute groups; any BIOS attribute group includes multiple different types of parameters that affect the performance of the server; the OS attributes have corresponding OS attribute group sets; the OS attribute group set includes multiple OS attribute groups; any OS attribute group includes multiple different types of parameters that affect the performance of the server;

[0012] The first optimization module is used to configure the server based on the first attribute group under the fixed first attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test respectively under the configuration of the server based on each attribute group under the second attribute; and determine the optimal attribute group under the second attribute based on the test performance results corresponding to each attribute group; one attribute of the BIOS attribute and the OS attribute is used as the first attribute, and the other attribute is used as the second attribute; the first attribute group is an attribute group selected from the attribute group set corresponding to the first attribute;

[0013] The second optimization module is used to configure the server based on the optimal attribute group under the fixed second attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test on each attribute group under the configuration; and determine the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group; the optimal attribute group under the first attribute and the optimal attribute group under the second attribute constitute the optimal parameter group corresponding to the server.

[0014] An embodiment of the present application also provides an electronic device, comprising: a processor and a computer-readable storage medium for storing computer program instructions, wherein the computer program instructions, when executed by the computer-readable storage medium, enable the processor to execute the steps of the above method.

[0015] An embodiment of the present application also provides a machine-readable storage medium, which stores computer program instructions. When the computer program instructions are executed, the steps of the above method can be implemented.

[0016] It can be seen from the above technical solutions that in this embodiment, by performing performance analysis on the server, two types of attributes, namely, BIOS attributes and operating system OS attributes, which affect the performance of the server, are obtained. The BIOS attributes have multiple BIOS attribute groups, and any BIOS attribute group includes multiple different types of parameters. The OS attributes have multiple OS attribute groups, and any OS attribute group includes multiple different types of parameters. This provides a strong basis for automatically obtaining the optimal parameter group that makes the performance of the server optimal.

[0017] On this basis, one of the BIOS attribute and the OS attribute is used as the first attribute, and the other is used as the second attribute. The server is configured through the first attribute group under the fixed first attribute, so that the server is configured, and the SPECCPU test is performed based on each attribute group under the second attribute to obtain the test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the second attribute is determined. The server is configured based on the optimal attribute group under the fixed second attribute, so that the server is configured, and the SPECCPU test is performed based on each attribute group under the first attribute to obtain the test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the first attribute is determined. In this way, the optimal parameter group that makes the server performance in the optimal state can be automatically obtained. This method of automatically making the server performance in the optimal state is faster than the method of manually determining the optimal parameters of server performance, and can ensure the real-time performance of obtaining the optimal parameter group, which solves the problem of poor real-time performance of manually determining the optimal parameters of server performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of a process for a method provided in an embodiment of the present application;

[0019] Figure 2 A flowchart for determining an optimal attribute group under a second attribute provided in an embodiment of the present application;

[0020] Figure 3 Another schematic diagram of the method provided in the embodiment of the present application;

[0021] Figure 4 A schematic diagram of the structure of the device provided in the embodiment of the present application;

[0022] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solutions provided by the embodiments of the present application and to make the above-mentioned purposes, features and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings.

[0024] Figure 1 , Figure 1 A flow chart of the method provided in an embodiment of the present application.

[0025] like Figure 1 As shown, the process may include the following steps:

[0026] S101, by performing performance analysis on the server, two types of attributes that affect the server performance are obtained: basic input and output system BIOS attributes and operating system OS attributes; BIOS attributes have corresponding BIOS attribute group sets; BIOS attribute group combinations include multiple attribute groups; any BIOS attribute group includes multiple different types of parameters that affect server performance; OS attributes have corresponding OS attribute group sets; OS attribute group sets include multiple OS attribute groups; any OS attribute group includes multiple different types of parameters that affect server performance.

[0027] In this embodiment, the specific implementation method of obtaining two types of attributes that affect server performance by analyzing the performance of the server in step S101 will be described later and will not be repeated here. The specific forms of the BIOS attributes and OS attributes will also be described later and will not be repeated here.

[0028] S102, configuring the server based on the first attribute group under the fixed first attribute to obtain the server under the configuration, performing SPECCPU test on each attribute group under the second attribute to obtain the test performance results corresponding to each attribute group; and determining the optimal attribute group under the second attribute based on the test performance results corresponding to each attribute group, one attribute of the BIOS attribute and the OS attribute as the first attribute, and the other attribute as the second attribute; the first attribute group is an attribute group selected from the attribute group set corresponding to the first attribute.

[0029] In this embodiment, the specific implementation method of the above step S102 will be explained later and will not be repeated here.

[0030] In this embodiment, optionally, the test performance result may be represented by a test performance score obtained by a SPECCPU test.

[0031] S103, configuring the server based on the optimal attribute group under the fixed second attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test on each attribute group under the configuration; and determining the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group; the optimal attribute group under the first attribute and the optimal attribute group under the second attribute constitute the optimal parameter group corresponding to the server.

[0032] In this embodiment, the specific implementation method of the above step S103 will be described later and will not be repeated here.

[0033] So far, completed Figure 1 The process shown.

[0034] pass Figure 1 It can be seen from the process shown that in this embodiment,

[0035] By analyzing the performance of the server, we can obtain the two types of attributes that affect the server performance, namely, BIOS attributes and operating system OS attributes. The BIOS attributes have multiple BIOS attribute groups, and any BIOS attribute group includes multiple different types of parameters. The OS attributes have multiple OS attribute groups, and any OS attribute group includes multiple different types of parameters. This provides a strong basis for automatically obtaining the optimal parameter group that makes the server performance in the optimal state.

[0036] On this basis, one of the BIOS attribute and the OS attribute is used as the first attribute, and the other is used as the second attribute. The server is configured through the first attribute group under the fixed first attribute, so that the server is configured, and the SPECCPU test is performed based on each attribute group under the second attribute to obtain the test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the second attribute is determined. The server is configured based on the optimal attribute group under the fixed second attribute, so that the server is configured, and the SPECCPU test is performed based on each attribute group under the first attribute to obtain the test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the first attribute is determined. In this way, the optimal parameter group that makes the server performance in the optimal state can be automatically obtained. This method of automatically making the server performance in the optimal state is faster than the method of manually determining the optimal parameters of server performance, and can ensure the real-time performance of obtaining the optimal parameter group, which solves the problem of poor real-time performance of manually determining the optimal parameters of server performance.

[0037] The following is a detailed description of the two types of attributes that affect server performance, as well as BIOS attributes and OS attributes, obtained by performing performance analysis on the server.

[0038] In the above step S101, there are many ways to obtain the two types of attributes that affect the performance of the server by analyzing the performance of the server. For example, as an embodiment, the Standard Performance Evaluation Corporation CPU test (SPECCPU) is a standardized benchmark tool used to evaluate the performance of the server, which is crucial throughout the life cycle of server development. The SPECCPU test suite contains a series of test programs for measuring the performance of the server. This test allows users to understand the performance differences of servers of different models.

[0039] Through the testing process of the SPECCPU test server and the performance test results of different servers using the SPECCPU test server, the server performance analysis problem is modeled. During the analysis and modeling process of this problem, two types of attributes that affect server performance can be determined: BIOS attributes and operating system OS attributes.

[0040] As an embodiment, the at least two parameter types corresponding to the BIOS attribute are at least two of the following parameter types:

[0041] CPU Core Feature Set parameter type, which includes multiple parameters;

[0042] Data Fabric Feature Sets parameter type, which includes multiple parameters;

[0043] Non-blocking input feature set BIO Feature Sets parameter type; BIO Feature Sets parameter type includes multiple parameters;

[0044] The virtualization function is combined with the Virtualization parameter type, and the Virtualization parameter type includes multiple parameters.

[0045] Any BIOS attribute group includes a parameter under each parameter type under at least two parameter types corresponding to the BIOS attribute. A parameter under each of the above classes constitutes a BIOS attribute group, denoted as b, and each BIOS attribute group constitutes a BIOS attribute group set, denoted as {B}, b∈{B}.

[0046] The at least two parameter types corresponding to the OS attribute are at least two of the following parameter types:

[0047] CPU Scheduler Policy parameter type, which includes multiple parameters;

[0048] Huge Pages parameter type, which includes multiple parameters;

[0049] CPU Frequency Scaling parameter type, which includes multiple parameters;

[0050] Read / write scheduling strategy I / O Scheduler parameter type, which includes multiple parameters;

[0051] CPU Affinity parameter type, which includes multiple parameters;

[0052] The swapping frequency swappiness parameter type includes multiple parameters.

[0053] High Resolution timer parameter type High Resolution parameter type, which includes multiple parameters.

[0054] Any OS attribute group includes a parameter under each parameter type under at least two parameter types corresponding to the OS attribute, and a parameter under each of the above classes, forming an OS attribute group, denoted as o, and each OS attribute group constitutes an OS attribute group set, denoted as {O}, o∈{O}.

[0055] Based on the above description, the server performance analysis problem is modeled and represented by Problem PO (Performance Optimization). Refer to the following table:

[0056] Table 1

[0057] Problem PO: <![CDATA[P max =maxSPEC(b,o)]]> subject to: b∈{B} (1) o∈{O} (2) variables: b,o

[0058] Wherein, b represents a BIOS attribute group;

[0059] Constraint (1) means that BIOS attribute group b is an element of the BIOS attribute group set {B};

[0060] o indicates an OS attribute group;

[0061] Constraint (2) means that OS attribute group o is an element of the OS attribute group set {O};

[0062] The function SPEC(b,o) represents the server performance score P returned by the server running the SPECCPU test under BIOS attribute group b and OS attribute group o.

[0063] Through the above method, it is determined that the two types of attributes that affect server performance are BIOS attributes and operating system OS attributes. Then, by subsequently finding the optimal BIOS attribute group under BIOS attributes and the optimal BIOS attribute group under OS attributes, the optimal parameter group corresponding to the server can be obtained.

[0064] The following is a detailed description of the two types of attributes that affect server performance, as well as BIOS attributes and OS attributes, obtained through the performance analysis of the server.

[0065] The following is a detailed description of the above test performance results corresponding to each attribute group to determine the optimal attribute group under the second attribute:

[0066] Please refer to Figure 2 , Figure 2 A flowchart of determining an optimal attribute group under a second attribute provided in an embodiment of the present application.

[0067] like Figure 2 As shown, the process may include the following steps:

[0068] S201, for each attribute group, if the attribute group is the first attribute group to be tested by SPECCPU, compare the test performance result corresponding to the attribute group with the set test performance result to obtain a better test performance result under the attribute group.

[0069] In this embodiment, if the test performance score obtained by the first attribute group subjected to the SPECCPU test is higher than the set test performance score, the test performance score returned by the SPECCPU test is retained to be compared with the test performance score obtained by the next attribute group subjected to the SPECCPU test.

[0070] If the test performance score obtained by the first attribute group subjected to the SPECCPU test is lower than the set test performance score, the set test performance score is retained to be compared with the test performance score obtained by the next attribute group subjected to the SPECCPU test.

[0071] If the test performance score obtained by the first attribute group subjected to the SPECCPU test is equal to the set test performance score, either one of the two can be retained.

[0072] S202: If the attribute group is not the first attribute group to be tested by SPECCPU, compare the test performance result corresponding to the attribute group with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group.

[0073] In this embodiment, if the attribute group is not the first attribute group to be tested with the SPECCPU, and if the test performance score corresponding to the attribute group is greater than the better test performance score under the previous attribute group, the test performance score corresponding to the attribute group is determined as the better test performance score under the current attribute group.

[0074] If the attribute group is not the first attribute group to be tested with SPECCPU, and if the test performance score corresponding to the attribute group is less than the better test performance score under the previous attribute group, the better test performance score under the previous attribute group will be determined as the better test performance score under the current attribute group.

[0075] If the attribute group is not the first attribute group to be tested by SPECCPU, and if the test performance score corresponding to the attribute group is equal to the better test performance score under the previous attribute group, then any one of the two test performance scores is determined as the better test performance score under this attribute group.

[0076] Similarly, after the above steps are performed on multiple attribute groups under the second attribute, the maximum test performance score can be determined.

[0077] S203: Determine the attribute group corresponding to the optimal test performance result as the optimal attribute group under the second attribute.

[0078] In this embodiment, the attribute group corresponding to the maximum test performance score is determined as the optimal attribute group under the second attribute.

[0079] The following describes in detail how to determine the optimal attribute group under the second attribute based on the test performance results corresponding to each attribute group.

[0080] For example, the first attribute is an OS attribute, and the second attribute is a BIOS attribute. Determining the optimal BIOS attribute group under the BIOS attribute can be performed by the script shown in the following table:

[0081] Table 2

[0082]

[0083]

[0084] Wherein, o is any OS attribute group o in the OS attribute group set {O}, which is actually a fixed OS attribute group o;

[0085] The set test performance score is 0;

[0086] P curThe server performance score P returned by running the SPECCPU test on the server under the current BIOS attribute group b and the fixed OS attribute group o;

[0087] P max Score the best server performance under the current BIOS attribute group b;

[0088] P m-max Score the maximum test performance, b * It is the BIOS attribute group corresponding to the maximum test performance score (that is, the optimal BIOS attribute group under the BIOS attributes).

[0089] The above method determines the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group. Except for configuring the server with the optimal attribute group under the fixed second attribute, and performing SPECCPU tests on each attribute group under the first attribute to obtain the test performance results corresponding to each attribute group and determine the optimal attribute group under the first attribute, the specific implementation process is the same as the above method. Figure 3 The process shown is similar and will not be repeated here.

[0090] For example, the first attribute is an OS attribute, and the second attribute is a BIOS attribute. Determining the optimal OS attribute group under the OS attribute may be performed by the script shown in the following table: Determining the optimal OS attribute group under the OS attribute may be performed by the script shown in the following table:

[0091] Table 3

[0092]

[0093]

[0094] Among them, b * It is the optimal BIOS attribute group under BIOS attributes;

[0095] The set test performance score is 0;

[0096] P cur Set the optimal BIOS attribute group bb for the server * The server performance score P returned by running the SPECCPU test under the current OS attribute group o;

[0097] P max Score the best server performance under the current OS attribute group o;

[0098] P m-max Score for maximum test performance, O * It is the OS attribute group corresponding to the maximum test performance score (that is, the optimal OS attribute group under the OS attributes).

[0099] b * ,o * is the optimal parameter group corresponding to the server.

[0100] The above describes in detail how to determine the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group.

[0101] In order to explain the method provided by the present application in more detail, the following Figure 3 The solution provided in this application is described in more detail by way of specific embodiments.

[0102] In this embodiment, for the convenience of description, OS is the first attribute and BIOS is the second attribute.

[0103] An OS attribute group is denoted as o, and each OS attribute group constitutes an OS attribute group set, denoted as {O}, o∈{O}

[0104] A BIOS attribute group is denoted as b, and each BIOS attribute group constitutes a BIOS attribute group set, denoted as {B}, b∈{B}

[0105] Combination Figure 3 As shown, the process includes the following steps:

[0106] 1. Modeling the performance optimization problem of the server system can produce the following model, expressed as Problem PO (Performance Optimization).

[0107] Problem PO: <![CDATA[P max =maxSPEC(b,o)]]> subject to: b∈{B} (1) o∈{O} (2) variables: b,o

[0108] Wherein, b represents a BIOS attribute group;

[0109] Constraint (1) means that BIOS attribute group b is an element of the BIOS attribute group set {B};

[0110] o indicates an OS attribute group;

[0111] Constraint (1) means that OS attribute group o is an element of the OS attribute group set {O};

[0112] The function SPEC(b,o) represents the server performance score P returned by the server running the SPECCPU test under BIOS attribute group b and OS attribute group o.

[0113] 2. For the vertical decomposition of the above Problem PO, given the OS attribute group o, we can get the lower-level problem Problem PO-Sub, which is expressed as:

[0114] Problem PO-Sub: <![CDATA[P max =maxSPEC(b,o)]]> subject to: b∈{B} (3) variables: b

[0115] Wherein, the constraint (2) indicates that the BIOS attribute group b is an element of the BIOS attribute group set {B}, and the BIOS attribute group is changed;

[0116] maxSPEC(b,o) is used to represent the BIOS attribute group corresponding to the maximum server performance score P under a given OS attribute group o.

[0117] Through the linear search method, the optimal BIOS attribute group b can be obtained * , the specific solution method refers to the above Table 2.

[0118] 3. Control b in maxSPEC(b,o) to b * , we further solve the upper-level problem Problem PO-Top, which can be expressed as:

[0119] Problem PO-Top: <![CDATA[P max =maxSPEC(b * ,o)]]> subject to: o∈{O} (4) variables: o

[0120] Among them, the constraint (4) indicates that the OS attribute group o is an element of the OS attribute group set {O}, and the BIOS attribute group is changed;

[0121] maxSPEC(b,o) is used to indicate the optimal BIOS attribute group b * Next, we seek the OS attribute group corresponding to the maximum server performance score P.

[0122] Through the linear search method, the optimal OS attribute group can be obtained. * , the specific solution method refers to the above Table 3.

[0123] 3. Based on b * and * , obtain the optimal parameters of the server to be tested.

[0124] In this embodiment, based on the SPECCPU test, the performance issues are modeled to analyze the impact on server performance, obtain two types of attributes, BIOS attributes and operating system OS attributes, that affect server performance, and determine that these two types of performance are decoupled issues in the optimization process. Therefore, a property group under one of the attributes (i.e., the first attribute) is fixed, and the optimal property group under another attribute (i.e., the second attribute) is sought. The optimal property group of the second attribute is fixed, and the optimal property group under the first attribute is sought. This can automatically obtain the optimal parameter group that makes the server performance at the optimal performance.

[0125] It should be noted that the above-mentioned optimal parameter group for obtaining the optimal performance of the server ignores the influence of hardware factors, and the default server hardware environment maintains the optimal configuration.

[0126] The method provided in the embodiment of the present application is described above. The device provided in the embodiment of the present application is described below:

[0127] See also Figure 4 , Figure 4 This is a diagram of the device structure provided in the embodiment of the present application. Figure 4 As shown, the device includes: an acquisition module 401, a first optimization module 402, and a second optimization module 403.

[0128] The obtaining module 401 is used to obtain two types of attributes that affect the performance of the server by performing performance analysis on the server: basic input and output system BIOS attributes and operating system OS attributes; the BIOS attributes have corresponding BIOS attribute group sets; the BIOS attribute group combination includes multiple attribute groups; any BIOS attribute group includes multiple different types of parameters that affect the performance of the server; the OS attributes have corresponding OS attribute group sets; the OS attribute group set includes multiple OS attribute groups; any OS attribute group includes multiple different types of parameters that affect the performance of the server;

[0129] The first optimization module 402 is used to configure the server based on the first attribute group under the fixed first attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test respectively under the configuration of the server based on each attribute group under the second attribute; and determine the optimal attribute group under the second attribute based on the test performance results corresponding to each attribute group; one attribute of the BIOS attribute and the OS attribute is used as the first attribute, and the other attribute is used as the second attribute; the first attribute group is an attribute group selected from the attribute group set corresponding to the first attribute;

[0130] The second optimization module 403 is used to configure the server based on the optimal attribute group under the fixed second attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test on each attribute group under the configuration; and determine the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group; the optimal attribute group under the first attribute and the optimal attribute group under the second attribute constitute the optimal parameter group corresponding to the server.

[0131] In this embodiment, one of the BIOS attribute and the OS attribute is used as the first attribute, and the other is used as the second attribute. The server is configured through the first attribute group under the fixed first attribute, so that the server is configured, and the SPECCPU test is performed based on each attribute group under the second attribute to obtain the test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the second attribute is determined. The server is configured based on the optimal attribute group under the fixed second attribute, so that the server is configured, and the SPECCPU test is performed based on each attribute group under the first attribute to obtain the test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the first attribute is determined. In this way, the optimal parameter group that makes the server performance in the optimal state can be automatically obtained. This method of automatically making the server performance in the optimal state is faster than the method of manually determining the parameters that make the server performance optimal, and can ensure the real-time performance of obtaining the optimal parameter group, solving the problem of poor real-time performance of manually determining the parameters that make the server performance optimal.

[0132] As an embodiment, based on the test performance results corresponding to each attribute group, determining the optimal attribute group under the second attribute includes:

[0133] For each attribute group, if the attribute group is the first attribute group to be tested by SPECCPU, the test performance result corresponding to the attribute group is compared with the set test performance result to obtain a better test performance result under the attribute group;

[0134] If the attribute group is not the first attribute group to be tested by SPECCPU, the test performance result corresponding to the attribute group is compared with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group;

[0135] The attribute group corresponding to the optimal test performance result is determined as the optimal attribute group under the second attribute.

[0136] As an embodiment, determining the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group includes:

[0137] For each attribute group, if the attribute group is the first attribute group to be tested by SPECCPU, the test performance result corresponding to the attribute group is compared with the set test performance result to obtain a better test performance result under the attribute group;

[0138] If the attribute group is not the first attribute group to be tested by SPECCPU, the test performance result corresponding to the attribute group is compared with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group;

[0139] The attribute group corresponding to the optimal test performance result is determined as the optimal attribute group under the first attribute.

[0140] As an embodiment, any BIOS attribute group includes a parameter of each parameter type of at least two parameter types corresponding to the BIOS attribute; the at least two parameter types corresponding to the BIOS attribute are at least two of the following parameter types:

[0141] CPU Core Feature Set parameter type;

[0142] Data Fabric Feature Sets parameter type;

[0143] Non-blocking input feature set BIO Feature Sets parameter type;

[0144] The virtualization function is combined with the Virtualization parameter type.

[0145] As an embodiment, any OS attribute group includes a parameter of each parameter type of at least two parameter types corresponding to the OS attribute; the at least two parameter types corresponding to the OS attribute are at least two of the following parameter types:

[0146] CPU scheduling policy CPU Scheduler Policy parameter type, large dirty pages Huge Pages parameter type, CPU frequency scaling CPU Frequency Scaling parameter type, read and write scheduling policy I / OScheduler parameter type, CPU affinity CPU Affinity parameter type, swappiness parameter type, high resolution timer HighResolution Timers parameter type.

[0147] So far, completed Figure 4 Structural description of the device shown.

[0148] See also Figure 5 , Figure 5 This is a structural diagram of an electronic device provided in an embodiment of the present application. Figure 5As shown, the hardware structure may include: a processor and a machine-readable storage medium, the machine-readable storage medium storing machine-executable instructions that can be executed by the processor; the processor is used to execute the machine-executable instructions to implement the method disclosed in the above example of this application.

[0149] Based on the same application concept as the above method, an embodiment of the present application also provides a machine-readable storage medium, on which a number of computer instructions are stored. When the computer instructions are executed by a processor, the method disclosed in the above example of the present application can be implemented.

[0150] Exemplarily, the above-mentioned machine-readable storage medium can be any electronic, magnetic, optical or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium can be: RAM (Radom Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as CD, DVD, etc.), or similar storage medium, or a combination thereof.

[0151] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A performance testing method, characterized in that: The method comprises: By performing performance analysis on the server, two types of attributes that affect the server performance are obtained: basic input and output system BIOS attributes and operating system OS attributes; the BIOS attributes have corresponding BIOS attribute group sets; the BIOS attribute group combination includes multiple attribute groups; any BIOS attribute group includes multiple different types of parameters that affect the server performance; the OS attributes have corresponding OS attribute group sets; the OS attribute group set includes multiple OS attribute groups; any OS attribute group includes multiple different types of parameters that affect the server performance; The server is configured based on the first attribute group under the fixed first attribute, so that the server is subjected to SPECCPU test under the configuration, and the test performance results corresponding to the attribute groups are obtained based on the attribute groups under the second attribute; and the optimal attribute group under the second attribute is determined based on the test performance results corresponding to the attribute groups; one of the BIOS attribute and the OS attribute is used as the first attribute, and the other attribute is used as the second attribute; the first attribute group is an attribute group selected from the attribute group set corresponding to the first attribute; The server is configured based on the optimal attribute group under the fixed second attribute, so that the server is subjected to SPECCPU test under the configuration based on each attribute group under the first attribute to obtain test performance results corresponding to each attribute group; and based on the test performance results corresponding to each attribute group, the optimal attribute group under the first attribute is determined; the optimal attribute group under the first attribute and the optimal attribute group under the second attribute constitute the optimal parameter group corresponding to the server.

2. The method according to claim 1, characterized in that The determining the optimal attribute group under the second attribute based on the test performance results corresponding to each attribute group includes: For each attribute group, if the attribute group is the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the set test performance result to obtain a better test performance result under the attribute group; If the attribute group is not the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group; The attribute group corresponding to the optimal test performance result is determined as the optimal attribute group under the second attribute.

3. The method according to claim 1, characterized in that The determining the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group includes: For each attribute group, if the attribute group is the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the set test performance result to obtain a better test performance result under the attribute group; If the attribute group is not the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group; The attribute group corresponding to the optimal test performance result is determined as the optimal attribute group under the first attribute.

4. The method according to claim 1, characterized in that: Any BIOS attribute group includes a parameter of each parameter type of at least two parameter types corresponding to the BIOS attribute; the at least two parameter types corresponding to the BIOS attribute are at least two of the following parameter types: CPU Core Feature Set parameter type; Data Fabric Feature Sets parameter type; Non-blocking input feature set BIO Feature Sets parameter type; The virtualization function is combined with the Virtualization parameter type.

5. The method according to claim 1, characterized in that Any OS attribute group includes a parameter of each parameter type of at least two parameter types corresponding to the OS attribute; the at least two parameter types corresponding to the OS attribute are at least two of the following parameter types: CPU scheduling policy CPU Scheduler Policy parameter type, large dirty pages Huge Pages parameter type, CPU frequency scaling CPU Frequency Scaling parameter type, read and write scheduling policy I / OScheduler parameter type, CPU affinity CPU Affinity parameter type, swappiness parameter type, high resolution timer HighResolution Timers parameter type.

6. A performance testing device, characterized in that: The device comprises: The obtaining module is used to obtain two types of attributes that affect the performance of the server by performing performance analysis on the server: basic input and output system BIOS attributes and operating system OS attributes; the BIOS attributes have corresponding BIOS attribute group sets; the BIOS attributes have corresponding BIOS attribute group sets; the BIOS attribute group combination includes multiple attribute groups; any BIOS attribute group includes multiple different types of parameters that affect the performance of the server; the OS attributes have corresponding OS attribute group sets; the OS attribute group set includes multiple OS attribute groups; any OS attribute group includes multiple different types of parameters that affect the performance of the server; A first optimization module is used to configure the server based on a first attribute group under a fixed first attribute, so as to obtain a test performance result corresponding to each attribute group by performing a SPECCPU test on the server under the configuration based on each attribute group under a second attribute; and to determine an optimal attribute group under the second attribute based on the test performance result corresponding to each attribute group; one of the BIOS attribute and the OS attribute is used as the first attribute, and the other attribute is used as the second attribute; the first attribute group is an attribute group selected from a set of attribute groups corresponding to the first attribute; The second optimization module is used to configure the server based on the optimal attribute group under the fixed second attribute, so as to obtain the test performance results corresponding to each attribute group by performing SPECCPU test on the server under the configuration based on each attribute group under the first attribute; and determine the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group; the optimal attribute group under the first attribute and the optimal attribute group under the second attribute constitute the optimal parameter group corresponding to the server.

7. The device according to claim 6, characterized in that The determining the optimal attribute group under the second attribute based on the test performance results corresponding to each attribute group includes: For each attribute group, if the attribute group is the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the set test performance result to obtain a better test performance result under the attribute group; If the attribute group is not the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group; Determining the attribute group corresponding to the optimal test performance result as the optimal attribute group under the second attribute; and / or, The determining the optimal attribute group under the first attribute based on the test performance results corresponding to each attribute group includes: For each attribute group, if the attribute group is the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the set test performance result to obtain a better test performance result under the attribute group; If the attribute group is not the first attribute group to be subjected to the SPECCPU test, the test performance result corresponding to the attribute group is compared with the better test performance result under the previous attribute group of the attribute group to obtain the better test performance result, so as to determine the best test performance result from the test performance results corresponding to each attribute group; The attribute group corresponding to the optimal test performance result is determined as the optimal attribute group under the first attribute.

8. The device according to claim 6, characterized in that Any BIOS attribute group includes a parameter of each parameter type of at least two parameter types corresponding to the BIOS attribute; the at least two parameter types corresponding to the BIOS attribute are at least two of the following parameter types: CPU Core Feature Set parameter type; Data Fabric Feature Sets parameter type; Non-blocking input feature set BIO Feature Sets parameter type; Virtualization function combined with Virtualization parameter type; and / or, Any OS attribute group includes a parameter of each parameter type of at least two parameter types corresponding to the OS attribute; the at least two parameter types corresponding to the OS attribute are at least two of the following parameter types: CPU scheduling policy CPU Scheduler Policy parameter type, large dirty pages Huge Pages parameter type, CPU frequency scaling CPU Frequency Scaling parameter type, read and write scheduling policy I / OScheduler parameter type, CPU affinity CPU Affinity parameter type, swappiness parameter type, high resolution timer HighResolution Timers parameter type.

9. An electronic device, characterized in that: The electronic device includes: Processor; and A computer-readable storage medium having computer program instructions stored therein, wherein the computer program instructions, when executed by the processor, cause the processor to perform the steps of the method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, which, when executed by a processor, enable the processor to perform the steps of any one of the methods of claims 1 to 5.

Images