Storage performance test method and device and medium

Abstract

The invention relates to a storage performance test method, equipment and a medium, which are used for testing storage performance data of a distributed storage system in various test scenes. The method comprises the following steps: in response to a test mode selected by a user, when it is determined that the selected test mode is a specified time delay mode, obtaining a target time delay and an adjustment coefficient input by the user; executing at least one pre-test in a test object at an initial queue depth, and determining a target queue depth according to the test delay of the pre-test, the target delay and the adjustment coefficient; and executing a storage performance test in the test object by using the target queue depth to obtain storage performance data of the test object.

Description

Technical Field

[0001] The present application relates to the field of automated testing technology, and in particular to a storage performance testing method, device, and medium. Background Art

[0002] In recent years, distributed storage has been widely used due to its high scalability, multi-protocol, high performance, high reliability and open ecosystem. Before distributed storage is applied, it is often necessary to conduct performance tests based on the application scenario to obtain performance indicators.

[0003] In the prior art, a distributed storage performance automation test system generally includes modules such as environment deployment, test execution, log collection and data processing, wherein the test execution is generally performed according to a preset stress testing tool configuration file.

[0004] The current test system cannot meet the test requirements, and a new storage performance testing method is needed to test the storage performance data of distributed storage in various test scenarios. Summary of the invention

[0005] Embodiments of the present application provide a storage performance testing method, device, and medium for testing storage performance data of a distributed storage system in a variety of test scenarios.

[0006] In a first aspect, an embodiment of the present application provides a storage performance testing method, the method comprising:

[0007] In response to the test mode selected by the user, when it is determined that the selected test mode is a specified delay mode, obtaining the target delay and adjustment coefficient input by the user, the specified delay mode being used to test the storage performance of the test object under the target delay;

[0008] Perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, the target delay, and the adjustment coefficient, wherein the queue depth is used to indicate the number of IO requests sent to the test object;

[0009] A storage performance test is performed in the test object using the target queue depth to obtain storage performance data of the test object.

[0010] Through the above method, when performing storage performance testing on the test object, the user can select the specified latency mode. In this test mode, the user can enter the specified target latency to test the storage performance data of the test object under the specified target latency. Through the extended specified latency mode, the object can enter the specified target latency for storage performance testing, which expands the application scenarios of storage performance testing.

[0011] In a possible implementation manner, determining the target queue depth according to the pre-test test delay, the target delay, and the adjustment coefficient includes:

[0012] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is continued;

[0013] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0014] Through the above method, at least one pre-test is performed in the test object with the initial queue depth, and the current queue depth is adjusted using the adjustment coefficient to determine the optimal queue depth when the test delay is less than the target delay specified by the object. In the subsequent storage performance test, this optimal queue depth is used to perform storage performance testing to ensure that the optimal storage performance data can be obtained when the delay meets the requirements.

[0015] In a possible implementation manner, increasing the current queue depth based on the adjustment coefficient to obtain a new queue depth includes:

[0016] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain a new queue depth;

[0017] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain a new queue depth.

[0018] Through the above method, when the current queue depth is increased, the granularity of queue depth adjustment during the pre-test process can be controlled by flexibly setting the adjustment coefficient.

[0019] In a possible implementation manner, reducing the current queue depth by using the adjustment coefficient includes:

[0020] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded down to obtain the target queue depth;

[0021] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0022] Through the above method, when it is determined that the test delay of the pre-test is greater than the target delay, the test delay under the current queue depth can no longer meet the target delay specified by the user, so the adjustment coefficient is used to reduce the current queue depth to obtain the target queue depth, that is, the optimal queue depth that meets the target delay specified by the user.

[0023] In a possible implementation manner, the test duration of the pre-test is shorter than a preset duration, and / or the number of tests is shorter than a preset number.

[0024] In a possible implementation, the method further includes:

[0025] When it is determined that the selected test mode is a specified hit rate mode, obtaining the target hit rate and offset input by the user, the specified hit rate mode being used to test the storage performance of the test object under the target hit rate;

[0026] Perform at least one round of storage performance testing on the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates;

[0027] Based on the at least one test result, storage performance data of the test object when the hit rate is between the first hit rate and the second hit rate is determined as the storage performance data of the test object at the target hit rate, the first hit rate being the difference between the target hit rate and the offset, and the second hit rate being the sum of the target hit rate and the offset.

[0028] Through the above method, when performing storage performance testing on the test object, the user can choose to specify the hit rate mode. In this test mode, the user can enter the target hit rate and offset specified by the user to obtain the storage performance data of the test object within the specified hit rate range. By extending the specified hit rate mode, for large data volume reading and writing application scenarios, the user can enter a specified hit rate to perform storage performance testing, thereby expanding the application scenarios of storage performance testing.

[0029] In a possible implementation, the method further includes: obtaining a test round number n input by the user, where n is a positive integer;

[0030] The performing at least one round of storage performance test on the test object comprises:

[0031] n rounds of storage performance tests are performed in the test object to obtain n test results, wherein each round of storage performance tests is as follows: data in the cache layer of the test object is flushed to the capacity layer, a storage performance test is performed in the test object, and the correspondence between the real-time hit rate and the test time is recorded; when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, the correspondence between the real-time hit rate and the test time and the test data of this round of testing are used as test results.

[0032] By using the above method, n rounds of storage performance tests are performed on the test object to obtain n test results, so as to improve the accuracy of the test results.

[0033] In a possible implementation, determining, based on the at least one test result, performance data when the hit rate of the test object is between the first hit rate and the second hit rate includes:

[0034] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time described in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0035] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data whose hit rate of the test object is between the first hit rate and the second hit rate;

[0036] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0037] Through the above method, storage performance data with a hit rate between the first hit rate and the second hit rate in each test result can be obtained, and then storage performance data with a hit rate between the first hit rate and the second hit rate in the entire test process (including one test result or multiple test results) can be obtained.

[0038] In a possible implementation manner, obtaining the target delay and adjustment coefficient input by the user includes:

[0039] Presenting a test parameter input option to the user, wherein the test parameters include a target delay and an adjustment factor;

[0040] Receive the target delay and adjustment coefficient input by the user through the test parameter input option.

[0041] In a possible implementation, presenting the test parameter input options to the user includes:

[0042] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0043] In a second aspect, an embodiment of the present application provides a storage performance testing method, the method comprising:

[0044] Obtain the target delay and adjustment coefficient input by the user;

[0045] Perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, the target delay, and the adjustment coefficient, wherein the queue depth is used to indicate the number of IO requests sent to the test object;

[0046] A storage performance test is performed in the test object using the target queue depth to obtain storage performance data of the test object.

[0047] In a possible implementation manner, determining the target queue depth according to the pre-test test delay, the target delay, and the adjustment coefficient includes:

[0048] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is continued;

[0049] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0050] In a possible implementation manner, increasing the current queue depth based on the adjustment coefficient to obtain a new queue depth includes:

[0051] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain a new queue depth;

[0052] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain a new queue depth.

[0053] In a possible implementation manner, reducing the current queue depth by using the adjustment coefficient to obtain the target queue depth includes:

[0054] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded down to obtain the target queue depth;

[0055] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0056] In a possible implementation manner, the test duration of the pre-test is shorter than a preset duration, and / or the number of tests is shorter than a preset number.

[0057] In a possible implementation manner, obtaining the target delay and adjustment coefficient input by the user includes:

[0058] Presenting a test parameter input option to the user, wherein the test parameters include a target delay and an adjustment factor;

[0059] Receive the target delay and adjustment coefficient input by the user through the test parameter input option.

[0060] In a possible implementation, presenting the test parameter input options to the user includes:

[0061] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0062] In a third aspect, an embodiment of the present application provides a storage performance testing method, the method comprising:

[0063] Get the target hit rate and offset entered by the user;

[0064] Perform at least one round of storage performance testing on the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates;

[0065] Based on the at least one test result, storage performance data of the test object when the hit rate is between the first hit rate and the second hit rate is determined as the storage performance data of the test object at the target hit rate, the first hit rate being the difference between the target hit rate and the offset, and the second hit rate being the sum of the target hit rate and the offset.

[0066] In a possible implementation, the method further includes: obtaining a test round number n input by the user, where n is a positive integer;

[0067] The performing at least one round of storage performance test on the test object comprises:

[0068] n rounds of storage performance tests are performed in the test object to obtain n test results, wherein each round of storage performance tests is as follows: data in the cache layer of the test object is flushed to the capacity layer, a storage performance test is performed in the test object, and the correspondence between the real-time hit rate and the test time is recorded; when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, the correspondence between the real-time hit rate and the test time and the test data of this round of testing are used as test results.

[0069] In a possible implementation, determining, based on the at least one test result, storage performance data when the hit rate of the test object is between the first hit rate and the second hit rate includes:

[0070] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time described in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0071] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data whose hit rate of the test object is between the first hit rate and the second hit rate;

[0072] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0073] In a possible implementation manner, obtaining the target hit rate and offset input by the user includes:

[0074] Presenting a test parameter input option to the user, wherein the test parameters include a target hit rate and an offset;

[0075] Receive the target hit rate and offset input by the user through the test parameter input option.

[0076] In a possible implementation, presenting the test parameter input options to the user includes:

[0077] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0078] In a fourth aspect, an embodiment of the present application provides a storage performance testing device, the device comprising:

[0079] An interaction unit, configured to respond to a test mode selected by a user, and when determining that the selected test mode is a specified delay mode, obtain a target delay and an adjustment coefficient input by the user;

[0080] A pre-test unit, configured to perform at least one pre-test with an initial queue depth in a test object, and determine a target queue depth according to a test delay of the pre-test, the target delay, and the adjustment coefficient;

[0081] The testing unit is used to perform a storage performance test in the test object using the target queue depth to obtain storage performance data of the test object.

[0082] In a possible implementation manner, the pre-test unit is specifically used to:

[0083] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is continued;

[0084] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0085] In a possible implementation manner, the pre-test unit is specifically used to:

[0086] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain a new queue depth;

[0087] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain a new queue depth.

[0088] In a possible implementation manner, the pre-test unit is specifically used to:

[0089] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded down to obtain the target queue depth;

[0090] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0091] In a possible implementation manner, the test duration of the pre-test is shorter than a preset duration, and / or the number of tests is shorter than a preset number.

[0092] In a possible implementation manner, the interaction unit is further configured to:

[0093] When it is determined that the selected test mode is the specified hit rate mode, obtaining the target hit rate and offset input by the user;

[0094] The test unit is further used to: perform at least one round of storage performance test in the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates;

[0095] The device also includes: a processing unit, for determining, based on the at least one test result, storage performance data of the test object when the hit rate is between a first hit rate and a second hit rate as storage performance data of the test object at the target hit rate, wherein the first hit rate is a difference between the target hit rate and the offset, and the second hit rate is a sum of the target hit rate and the offset.

[0096] In a possible implementation manner, the interaction unit is further used to: obtain the number of test rounds n input by the user, where n is a positive integer;

[0097] The testing unit is specifically used for:

[0098] n rounds of storage performance tests are performed in the test object to obtain n test results, wherein each round of storage performance tests is as follows: data in the cache layer of the test object is flushed to the capacity layer, a storage performance test is performed in the test object, and the correspondence between the real-time hit rate and the test time is recorded; when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, the correspondence between the real-time hit rate and the test time and the test data of this round of testing are used as test results.

[0099] In a possible implementation manner, the processing unit is specifically configured to:

[0100] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time described in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0101] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data whose hit rate of the test object is between the first hit rate and the second hit rate;

[0102] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0103] In a possible implementation manner, the interaction unit is specifically configured to:

[0104] Presenting a test parameter input option to the user, wherein the test parameters include a target delay and an adjustment factor;

[0105] Receive the target delay and adjustment coefficient input by the user through the test parameter input option.

[0106] In a possible implementation manner, the interaction unit is specifically configured to:

[0107] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0108] In a fifth aspect, an embodiment of the present application provides a storage performance testing device, the device comprising:

[0109] An acquisition unit, used to acquire a target delay and an adjustment coefficient input by a user;

[0110] A pre-test unit, configured to perform at least one pre-test with an initial queue depth in a test object, and determine a target queue depth according to a test delay of the pre-test, the target delay, and the adjustment coefficient;

[0111] The testing unit is used to perform a storage performance test in the test object using the target queue depth to obtain storage performance data of the test object.

[0112] In a possible implementation manner, the pre-test unit is specifically used to:

[0113] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is continued;

[0114] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0115] In a possible implementation manner, the pre-test unit is specifically used to:

[0116] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain a new queue depth;

[0117] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain a new queue depth.

[0118] In a possible implementation manner, the pre-test unit is specifically used to:

[0119] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded down to obtain the target queue depth;

[0120] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0121] In a possible implementation manner, the test duration of the pre-test is shorter than a preset duration, and / or the number of tests is shorter than a preset number.

[0122] In a possible implementation manner, the acquisition unit is specifically configured to:

[0123] Presenting a test parameter input option to the user, wherein the test parameters include a target delay and an adjustment factor;

[0124] Receive the target delay and adjustment coefficient input by the user through the test parameter input option.

[0125] In a possible implementation manner, the acquisition unit is specifically configured to:

[0126] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0127] In a sixth aspect, an embodiment of the present application provides a storage performance testing device, the device comprising:

[0128] An acquisition unit, used to acquire a target hit rate and an offset input by a user;

[0129] A testing unit, configured to perform at least one round of storage performance testing on a test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates;

[0130] A processing unit is used to determine, based on the at least one test result, storage performance data of the test object when the hit rate is between a first hit rate and a second hit rate as the storage performance data of the test object at the target hit rate, wherein the first hit rate is the difference between the target hit rate and the offset, and the second hit rate is the sum of the target hit rate and the offset.

[0131] In a possible implementation, the acquisition unit is further used to: acquire the number of test rounds n input by the user, where n is a positive integer;

[0132] The testing unit is specifically used for:

[0133] n rounds of storage performance tests are performed in the test object to obtain n test results, wherein each round of storage performance tests is as follows: data in the cache layer of the test object is flushed to the capacity layer, a storage performance test is performed in the test object, and the correspondence between the real-time hit rate and the test time is recorded; when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, the correspondence between the real-time hit rate and the test time and the test data of this round of testing are used as test results.

[0134] In a possible implementation manner, the processing unit is specifically configured to:

[0135] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time described in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0136] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data whose hit rate of the test object is between the first hit rate and the second hit rate;

[0137] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0138] In a possible implementation manner, the acquisition unit is specifically configured to:

[0139] Presenting a test parameter input option to the user, wherein the test parameters include a target delay and an adjustment factor;

[0140] Receive the target delay and adjustment coefficient input by the user through the test parameter input option.

[0141] In a possible implementation manner, the acquisition unit is specifically configured to:

[0142] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0143] In the seventh aspect, an embodiment of the present application provides a computer device, comprising a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes any one of the storage performance testing methods in the first aspect, the second aspect, and the third aspect.

[0144] In an eighth aspect, an embodiment of the present application provides a computer-readable storage medium, which includes a computer program. When the computer program is run on a computer device, the computer program is used to enable the computer device to execute any one of the storage performance testing methods in the first aspect, the second aspect, and the third aspect.

[0145] In the ninth aspect, an embodiment of the present application provides a computer program product, which includes a computer program, and the computer program is stored in a computer-readable storage medium; when the processor of a computer device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the computer device executes any one of the storage performance testing methods in the first aspect, the second aspect, and the third aspect above.

[0146] In the tenth aspect, the present application also provides a computer chip, which is connected to a memory, and is used to read and execute a software program stored in the memory, and to execute any one of the storage performance testing methods in the first aspect, the second aspect, and the third aspect.

[0147] For the technical effects that may be achieved by each of the above-mentioned second aspect to the tenth aspect, please refer to the above-mentioned description of the technical effects that can be achieved by various possible schemes in the first aspect, and no further details will be given here. BRIEF DESCRIPTION OF THE DRAWINGS

[0148] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0149] Figure 1 A schematic diagram of an application scenario in an embodiment of the present application;

[0150] Figure 2 A schematic flow chart of the overall test process of the storage performance test method provided in an embodiment of the present application;

[0151] Figure 3 A schematic flow chart of a storage performance testing method provided in an embodiment of the present application;

[0152] Figure 4 A schematic flow chart of a specific test process for a specified delay mode provided in an embodiment of the present application;

[0153] Figure 5 A schematic flow chart of another storage performance testing method provided in an embodiment of the present application;

[0154] Figure 6A schematic diagram of a real-time hit rate variation curve provided in an embodiment of the present application;

[0155] Figure 7 A schematic flow chart of a specific test process for a specified hit rate mode provided in an embodiment of the present application;

[0156] Figure 8 A schematic flow chart of another storage performance testing method provided in an embodiment of the present application;

[0157] Fig. 9 A schematic flow chart of another storage performance testing method provided in an embodiment of the present application;

[0158] Fig.10 A schematic diagram of the structure of a storage performance testing device provided in an embodiment of the present application;

[0159] Fig.11 A schematic diagram of the structure of another storage performance testing device provided in an embodiment of the present application;

[0160] Fig.12 A schematic diagram of the structure of another storage performance testing device provided in an embodiment of the present application;

[0161] Fig.13 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0162] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the technical solution of the present application, rather than all of the embodiments. Based on the embodiments recorded in the application documents, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the technical solution of the present application.

[0163] The following is an introduction to some concepts involved in the embodiments of the present application.

[0164] 1. Cache layer: In a distributed storage system, the cache layer is an important part of the storage hierarchy. The cache layer is used to increase the speed of data access. The system usually sets up a cache layer on the memory or high-speed storage device. The cache layer can cache frequently accessed data and reduce the reading of the basic storage layer.

[0165] 2. Capacity layer: In a distributed storage system, the capacity layer is an important part of the storage hierarchy. The capacity layer is mainly responsible for data storage and belongs to the basic storage layer. It has a large storage capacity and can meet the needs of large-scale data storage.

[0166] 3. Disk flushing: When a certain percentage of the cache layer capacity is used, in order to ensure overall performance, the cache layer data will be triggered to migrate to the capacity layer, which is called disk flushing.

[0167] 4. Pull: When reading data from the capacity layer, in order to improve the reading performance, the read data will be migrated from the capacity layer to the cache layer, which is called pull.

[0168] 5. Hit rate: The ratio of the amount of data read from the cache layer to the total amount of data. Generally, the higher the hit rate, the better the performance. The read and write process of distributed storage is accompanied by frequent disk flushing and pulling. Therefore, the hit rate of distributed storage changes in real time.

[0169] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application may be combined with each other if there is no conflict.

[0170] like Figure 1 As shown, it is a schematic diagram of an application scenario of an embodiment of the present application. The application scenario diagram includes any terminal device 110 among multiple terminal devices and any server 120 among multiple servers.

[0171] In the embodiment of the present application, the terminal device 110 is a computing device for managing the test, which may include but is not limited to mobile phones, tablet computers, laptops, desktop computers and other devices; the terminal device may be installed with a client such as storage performance test software, which may be software, web pages, applets, etc., and the server 120 is the background service end of the storage performance test. The server 120 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0172] It should be noted that the storage performance testing method in the embodiment of the present application can be executed by the server 120, and the overall testing process is as follows: Figure 2 As shown, including:

[0173] Step 201 : the server 120 deploys a test environment, such as allocating test volumes, building a distributed storage system model, creating test objects, and the like.

[0174] Step 202: The server 120 displays a test mode selection interface to the user through the terminal device 110. The test mode selection interface may provide a variety of test modes for the user to choose from, such as a normal test mode, a specified delay mode, and a specified hit rate mode. The server 120 performs corresponding storage performance tests in response to the test mode selected by the user through the terminal device 110.

[0175] Step 203: during the storage performance test, the server 120 collects storage performance data of the test object and peripheral data recorded in the log.

[0176] Step 204: After the storage performance test is completed, the server 120 verifies whether the data is valid and processes invalid data.

[0177] Step 205 : The server 120 outputs valid data to the user via the terminal device 110 as a storage performance test result.

[0178] In an optional implementation, the terminal device 110 and the server 120 may communicate with each other via a communication network, which may be a wired network or a wireless network.

[0179] It should be noted that Figure 1 What is shown is just an example. In fact, the number and communication method of terminal devices and servers are not limited. When there are multiple servers, multiple servers can form a blockchain, and the server is a node on the blockchain, which is not specifically limited in the embodiments of the present application.

[0180] In practical applications, the storage performance testing method provided in the embodiments of the present application can be applied to storage performance testing of various storage systems, such as distributed file storage systems and distributed object storage systems.

[0181] The above introduces the application scenarios and overall test process of the storage performance testing method provided in the embodiment of the present application in combination with specific application scenarios. The following introduces the specific test processes of the various test modes provided in the embodiment of the present application in combination with specific embodiments.

[0182] In the storage performance test method provided in the embodiment of the present application, in order to meet the test requirements of more application scenarios, the test modes include: normal mode, specified delay mode and specified hit rate mode. In some test scenarios, users are more concerned about the latency and hit rate of the storage system, such as latency-sensitive application scenarios, where users are more concerned about the latency of the storage system, and in application scenarios with large amounts of data read and written, where users are more concerned about the hit rate of the storage system. In view of this, in the storage performance test of the storage system, the embodiment of the present application sets a specified latency mode and a specified hit rate mode to adapt to more diverse test scenarios.

[0183] Test mode 1: Normal test mode.

[0184] When it is determined that the test mode selected by the user is the normal test mode, the user is requested to input parameters such as data block size, read-write ratio, random ratio, queue depth, etc., and then based on the parameters entered by the user, the pre-created test object is reconfigured or a new test object is created to perform storage performance testing in the test object.

[0185] The storage performance test data in the normal test mode may include, but is not limited to: the number of read and write operations per second (Input / Output Operations Per Second, IOPS), latency, throughput, hit rate, response time, and bandwidth.

[0186] I / O, or Input / Output, refers to writing data to the storage system, and output refers to reading data from the storage system. I / O can be simply understood as disk reading and writing. IOPS refers to the number of I / O requests that the system can handle per unit time. Usually, the formula for calculating IOPS is: (total number of read operations + number of write operations) / time (seconds). IOPS is usually used to measure the random access performance of the system.

[0187] Latency, or delay, refers to the speed at which the storage system processes a single I / O request. Bandwidth refers to the average amount of data read and written per second. Throughput refers to the amount of data read and written within a period of time. It should be noted that, under the same latency, increasing concurrency can increase throughput. Response time refers to the time from when an object sends a request to when the application processes the request and returns it to the object.

[0188] Test mode 2: Specify delay mode.

[0189] When performing storage performance testing in this mode, you can dynamically modify the parameters of the stress testing tool based on the multiplication method, adjust the test latency to the target latency specified by the user, and output the storage performance data of the test object, such as IOPS, throughput, hit rate, response time, and bandwidth.

[0190] In actual applications, the specified latency mode is suitable for application scenarios that are sensitive to latency, such as storage performance evaluation in database scenarios. In such application scenarios, you can specify the target latency, that is, the maximum latency of the test object, and obtain the storage performance data of the test object under the specified latency.

[0191] Test mode three: specify hit rate mode.

[0192] When performing performance testing in this mode, you can first pre-write the test volume with large blocks of data to construct an initial condition with a hit rate of 0%, then perform a storage performance test on the test object to obtain the correspondence between the real-time hit rate and the test time. Finally, based on the correspondence between the real-time hit rate and the test time, obtain storage performance data within the specified hit rate range, such as IOPS, latency, throughput, response time, and bandwidth.

[0193] In actual applications, specifying the hit rate mode is suitable for application scenarios with large data volume reading and writing, such as storage performance evaluation in video surveillance scenarios. In such application scenarios, you can specify the target hit rate and offset, that is, specify the hit rate range of the test object, and obtain the storage performance data of the test object.

[0194] The following describes a specific implementation process of a specified delay mode provided by this embodiment. Figure 3 As shown, the specific implementation process of the specified delay mode is as follows S301-S303:

[0195] S301: In response to a test mode selected by a user, when it is determined that the selected test mode is a specified delay mode, obtaining a target delay and an adjustment coefficient input by the user.

[0196] In specific implementation, when obtaining the target delay and adjustment coefficient input by the user, the test parameter input option can be first displayed to the user, and the test parameters include the target delay and adjustment coefficient, and then the target delay and adjustment coefficient input by the user through the test parameter input option are received. Specifically, the test parameter input option can be displayed to the user through the test parameter input interface or the test parameter input dialog box. Of course, in other embodiments of the present application, other methods can also be used to display the test parameter input option to the user and obtain the test parameters entered by the user.

[0197] The target delay is the maximum test delay specified by the user, and the adjustment coefficient is the multiple of the queue depth. The queue mentioned here is used to store the IO requests generated during the test, and the queue depth indicates the number of IO requests issued by the server at the same time. Since it is necessary to determine the maximum test delay less than the target delay by multiplying the queue depth, the adjustment coefficient can be set to any other value greater than 0 and not equal to 1, and the embodiments of the present application do not limit this.

[0198] In practical applications, in addition to obtaining the target delay and adjustment coefficient input by the user, one or more of the following test parameters can also be obtained: data block size, read-write ratio, random ratio, write-through parameter, and test time.

[0199] In the specific implementation, after obtaining the test parameters input by the user, if a test object has been created in advance, the created test object can be reconfigured based on the test parameters input by the user to obtain the test object used in this test; if the test object has not been created in advance, a new test object can be created based on the test parameters input by the user to obtain the test object used in this test.

[0200] S302: Perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, a target delay, and an adjustment coefficient.

[0201] It should be noted that the initial queue depth can be initialized to a set value when configuring or creating the parameters of the test object, for example, initialized to 1, 10, or 100, or it can be input by the user as a test parameter, which is not limited in the embodiments of the present application.

[0202] In specific implementation, pre-test means that the test duration is less than the preset duration, and / or the number of tests is less than the preset number of times. The preset duration and the preset number of times can be set based on experience. For example, the preset duration is 1 minute and the preset number of times is 5 times. This embodiment of the present application does not limit this.

[0203] It should be noted that the target queue depth can be the queue depth corresponding to any test delay when the test delay is less than the target delay. However, in the actual test process, in order to obtain the optimal storage performance data of the test object when the test delay is less than the target delay, it is necessary to determine the largest possible queue depth as the target queue depth while ensuring that the test delay is less than the target delay.

[0204] In actual applications, at least one pre-test is performed with the initial queue depth in the test object. When it is determined that the test delay is less than the target queue depth of the target delay, a pre-test can be performed with the initial queue depth to determine the test delay of the pre-test.

[0205] If it is determined that the test delay of the pre-test is less than or equal to the target delay, indicating that there is still room for the current queue depth to be increased, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is performed using the new queue depth.

[0206] Specifically, when the current queue depth is increased based on the adjustment coefficient, if the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the new queue depth; if the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain the new queue depth. Of course, in actual applications, other methods can also be used to increase the current queue depth based on the adjustment coefficient, for example, the result of rounding up the sum of the current queue depth and the adjustment coefficient is used as the new queue depth.

[0207] The queue depth is increased by adjusting the coefficient, and the pre-test is continuously performed to determine the pre-test delay until the pre-test delay is greater than the target delay. At this time, the pre-test delay is greater than the target delay, indicating that the current queue depth cannot meet the target delay specified by the user. Therefore, the target queue depth can be obtained by reducing the current queue depth using the adjustment coefficient. This target queue depth is the optimal queue depth when the test delay is less than the target delay.

[0208] When specifically using the adjustment coefficient to reduce the current queue depth, a method opposite to the increasing method mentioned in the embodiment of the present application (using the adjustment coefficient to increase the current queue depth) may be used.

[0209] Specifically: If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the comparison value is rounded down to obtain the target queue depth; if the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0210] It should be noted that when the test delay of the pre-test is greater than the target delay, the adjustment coefficient is used to reduce the current queue depth. Specifically, the reduction can be performed once. In other embodiments of the present application, multiple reductions can also be performed, and a pre-test is performed after each reduction until the test delay of the pre-test is less than the target delay, and the reduction process is stopped.

[0211] It should be noted that the smaller the adjustment coefficient is set, the more likely it is to determine the largest possible queue depth as the target queue depth, and the larger the adjustment coefficient is set, the easier it is to determine the target queue depth. Therefore, in actual situations, the adjustment coefficient can be set according to actual needs, or different adjustment coefficients can be set for multiple tests to obtain the best storage performance data when the test delay is less than the target delay.

[0212] In addition, in an embodiment of the present application, in the process of determining the target queue depth, when it is determined that the test delay of the pre-test is equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth to continue the pre-test. In other embodiments of the present application, when it is determined that the test delay of the pre-test is equal to the target delay, the queue depth used in this pre-test can be directly determined as the target queue depth, or the queue depth used in this pre-test can be reduced by using the adjustment coefficient to obtain the target queue depth.

[0213] S303: Perform a storage performance test on the test object using the target queue depth to obtain storage performance data of the test object.

[0214] In specific implementation, after determining the target queue depth, the target queue depth is used to perform storage performance testing in the test object, which can ensure that the test delay is less than the target delay. Therefore, the storage performance data of the test object when the test delay is less than the target delay can be obtained. The storage performance data may include but is not limited to: IOPS, throughput, hit rate, response time, and bandwidth.

[0215] Combine the following Figure 4 Taking the adjustment coefficient greater than 1 as an example, the specific test process of specifying the delay mode in the storage performance test method provided in the embodiment of the present application is described in detail. Figure 4 As shown, the specific test process of specifying the latency mode in the storage performance test method provided in the embodiment of the present application includes:

[0216] Step 401, obtaining test parameters, the test parameters including target delay and adjustment coefficient.

[0217] Of course, the test parameters are input by the user, and the test parameters also include the data block size, read-write ratio, random ratio, write-through size, and test time required for the test.

[0218] After obtaining the test parameters input by the user, if a test object has been created in advance, the created test object can be reconfigured based on the test parameters input by the user to obtain the test object used in this test. If the test object has not been created in advance, a new test object can be created based on the test parameters input by the user to obtain the test object used in this test.

[0219] Step 402: Initialize the queue depth in the test object. For example, initialize the queue depth to a preset queue depth, such as initializing the queue depth to 1.

[0220] Of course, in other embodiments of the present application, the initialization queue depth may also be input by the user, and the embodiments of the present application do not limit this.

[0221] Step 403: perform a pre-test to determine the current test delay.

[0222] It should be noted that, for the first execution of the pre-test, that is, the pre-test is executed with the initial queue depth, if the pre-test is continued after executing step 407 , the pre-test is executed with the new queue depth calculated in step 407 .

[0223] Step 404, determine whether the current test delay is greater than the target delay, if so, execute step 405, otherwise execute step 407.

[0224] Step 405: When it is determined that the current test delay is greater than the target delay, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded down to obtain the target queue depth.

[0225] Step 406: Perform a storage performance test on the test object using the target queue depth to obtain storage performance data of the test object.

[0226] Step 407, when it is determined that the current test delay is less than or equal to the target delay, first calculate the product of the current queue depth and the adjustment coefficient, then round up the product to obtain a new queue depth, and execute step 403 to continue the next pre-test.

[0227] In a specific example, assume that the parameters input by the user are: data block size is 8 kilobytes (KB), read-write ratio is 7:3, random ratio is 100%, test duration is 5 minutes (300 seconds), target delay t is 10 milliseconds (ms), and adjustment coefficient K is 1.5.

[0228] Initialize the queue depth to 1 in the test object, perform a 20s pretest, and obtain the current test delay tc to be 0.15ms. At this time, tc is less than t, so the queue depth is multiplied by K, the product is 1.5, and the product is rounded up to 2 as the new queue depth for this pretest. Perform a 20s pretest and obtain the current test delay tc to be 0.17ms. Repeat the doubling of the queue depth and pretest multiple times. When the current queue depth is 315, determine that the test delay tc of the pretest is 12.5ms.

[0229] At this time, tc is greater than t, so the ratio of the current queue depth 315 to K is 210, which is used as the target queue depth. The target queue depth is 210, and a storage performance test is performed on the test object for 5 minutes according to the parameters entered by the user. The test results show that the test latency is 8.4ms and the IOPS is 220K.

[0230] It should be noted that the numerical values ​​listed in this example are only used to illustrate the test process of the specified delay mode provided in the embodiment of the present application. There may be different numerical values ​​in other examples of the embodiment of the present application, and they are not used for specific limitation here.

[0231] The following describes a specific implementation process of a specified hit rate mode provided by this embodiment. Figure 5 As shown, the specific implementation process of specifying the hit rate mode is as follows S501-S503:

[0232] S501: When it is determined that the selected test mode is the specified hit rate mode, obtain the target hit rate and offset input by the user.

[0233] Among them, the target hit rate is the hit rate specified by the user, and the offset is the error value of the target hit rate. The target hit rate and the offset can be used to determine the hit rate range specified by the user. The lower limit of this range is the first hit rate, and the upper limit is the second hit rate. The first hit rate is the difference between the target hit rate and the offset, and the second hit rate is the sum of the target hit rate and the offset.

[0234] In actual applications, in addition to obtaining the target hit rate and offset input by the user, one or more of the following parameters can also be obtained: data block size, read-write ratio, random ratio, write-through parameter, queue depth, etc.

[0235] After obtaining the test parameters input by the user, if a test object has been created in advance, the created test object can be reconfigured based on the test parameters input by the user to obtain the test object used in this test. If the test object has not been created in advance, a new test object can be created based on the test parameters input by the user to obtain the test object used in this test.

[0236] S502, performing at least one round of storage performance testing on the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates.

[0237] In specific implementation, in order to obtain more accurate storage performance data, multiple rounds of storage performance tests can be performed on the test object, and the average value of the storage performance data in the multiple rounds of test results can be calculated as the final storage performance data. Therefore, the number of test rounds n entered by the user can also be obtained, where n is a positive integer.

[0238] In this way, when at least one round of storage performance testing is performed on the test object and at least one test result is obtained, n rounds of storage performance testing are performed on the test object and n test results are obtained.

[0239] Among them, each round of storage performance test is as follows: flush the data in the cache layer of the test object to the capacity layer, perform storage performance test in the test object, record the corresponding relationship between the real-time hit rate and the test time, and when it is determined that the hit rate of the test object reaches the preset value or the test duration is greater than the preset test duration, generate the hit rate curve of the test object according to the corresponding relationship between the real-time hit rate and the test time and use it as the test result.

[0240] It should be noted that flushing the data in the cache layer of the test object to the capacity layer is to construct an initial test condition with a hit rate of 0%. Specifically, flushing the data in the cache layer to the capacity layer can be achieved by pre-writing large blocks of data to the test volume, or by other existing methods, which is not limited to the embodiments of the present application.

[0241] When performing storage performance testing on the test object, the corresponding relationship between the real-time hit rate and the test time is recorded, and test data is collected at the same time to facilitate analysis of the storage performance data of the test object. When it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, the test is stopped, and the corresponding relationship between the real-time hit rate and the test time and the test data of this round of testing are used as test results. Among them, the preset value can be 100% or any value greater than the second hit rate, and the preset test duration can be set according to actual conditions, for example, the preset test duration is 20 minutes or 30 minutes, etc.

[0242] S503, based on at least one test result, determine the storage performance data of the test object when the hit rate is between the first hit rate and the second hit rate, as the storage performance data of the test object at the target hit rate, the first hit rate is the difference between the target hit rate and the offset, and the second hit rate is the sum of the target hit rate and the offset.

[0243] In specific implementation, after obtaining n test results, the storage performance data of the test object when the hit rate is between the first hit rate and the second hit rate can be determined based on the n test results as the storage performance data of the test object at the target hit rate.

[0244] Specifically, for each test result, the following operations are performed to obtain the storage performance data when the hit rate of the test object in each test result is between the first hit rate and the second hit rate: according to the correspondence between the real-time hit rate and the test time in the test result, determine the first moment when the hit rate is greater than or equal to the first hit rate for the first time, and the second moment when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first moment and the second moment based on the test data.

[0245] For example, the corresponding relationship between the real-time hit rate and the test time in the test results can be recorded using the hit rate change curve, such as Figure 6 As shown in Figure 1, assume that this curve records the corresponding relationship between the real-time hit rate and the test time in a test result. If the specified hit rate is 50% and the offset is 5%, the first hit rate is 45% and the second hit rate is 55%. Figure 6 In the hit rate curve shown, the first moment t1 when the hit rate is greater than or equal to the first hit rate for the first time, and the second moment t2 when the hit rate is less than or equal to the second hit rate for the first time are determined, and the storage performance data between the first moment t1 and the second moment t2 are determined based on the test data, that is, the storage performance data corresponding to this test result.

[0246] After determining the storage performance data corresponding to n test results respectively, if n is 1, the performance data corresponding to this test result is used as the performance data whose hit rate of the test object is between the first hit rate and the second hit rate; if n is greater than 1, the average value of the storage performance data corresponding to multiple test results is used as the storage performance data whose hit rate of the test object is between the first hit rate and the second hit rate.

[0247] Combine the following Figure 7 , the specific test process of specifying the hit rate mode in the storage performance test method provided in the embodiment of the present application is described in detail. Figure 7 As shown, the specific test process of specifying the hit rate mode in the storage performance test method provided in the embodiment of the present application includes:

[0248] Step 701, obtaining test parameters, the test parameters including target hit rate, offset, and number of test rounds n.

[0249] Of course, the test parameters are input by the user, and also include the data block size, read-write ratio, random ratio, write-through size, and queue depth required for the test.

[0250] After obtaining the test parameters input by the user, if a test object has been created in advance, the created test object can be reconfigured based on the test parameters input by the user to obtain the test object used in this test. If the test object has not been created in advance, a new test object can be created based on the test parameters input by the user to obtain the test object used in this test.

[0251] Step 702: Use large blocks of data to pre-write the test volume in the test object, and flush the data in the cache layer to the capacity layer. This step is to construct an initial test condition with a hit rate of 0%.

[0252] Step 703: Continue to perform storage performance test on the test object, and record the corresponding relationship between the real-time hit rate and the test time.

[0253] Step 704, determine whether the hit rate reaches 100%, if so, execute step 705, otherwise continue to execute step 703.

[0254] Step 705, stop the test, collect test data, and determine the time t1 when the hit rate is greater than or equal to the first hit rate for the first time and the time t2 when the hit rate is less than or equal to the second hit rate for the first time based on the corresponding relationship between the real-time hit rate and the test time. The first hit rate is the difference between the target hit rate and the offset, and the second hit rate is the sum of the target hit rate and the offset.

[0255] Step 706: Based on the test data, determine the storage performance data of the test object between time t1 and time t2 in this round of testing, and save it.

[0256] Step 707: The number of test rounds is increased by a preset value. For example, if the preset value is 1, the number of test rounds is increased by 1.

[0257] Step 708, determine whether the number of test rounds is equal to n, if so, execute step 709, otherwise continue to execute step 702.

[0258] Step 709, calculate the average value of the storage performance data corresponding to the n rounds of test results, and determine the storage performance data of the test object between the first hit rate and the second hit rate as the storage performance data of the test object at the target hit rate.

[0259] In a specific example, assume that the parameters input by the user are: data block size is 8 kilobytes (KB), read-write ratio is 7:3, random ratio is 100%, queue depth is 256, target hit rate is 50%, offset is 5, and test round number is 3.

[0260] Assuming that the disk block size of the test object is 512KB, in one round of testing, first use 1024KB data blocks to perform sequential writes on the test volume, so that all data is flushed to the capacity layer, constructing the initial test condition with a hit rate of 0%, and then perform storage performance tests on the test object for an unlimited duration. During the test, the correspondence between the real-time hit rate and the test time is recorded. After the test continues for a period of time, when it is determined that the hit rate has reached 100%, stop the test and collect test data.

[0261] In the correspondence between the recorded real-time hit rate and the test time, it is assumed that the real-time hit rate is just greater than or equal to the hit rate lower limit of 45% (the difference between the target hit rate and the offset) and the hit rate is 45.20%, and the corresponding test time t1 is 10:00:00; the real-time hit rate is just less than or equal to the hit rate upper limit of 55% (the sum of the target hit rate and the offset) and the hit rate is 54.95%, and the corresponding test time t2 is 11:30:00. The test data collected in this round of testing is selected from the test data from 10:00:00 (t1) to 11:30:00 (t2), and the IOPS average value r1 is calculated as the storage performance data of this round of testing.

[0262] Repeat the second round of performance testing to obtain the IOPS average value r2, repeat the third round of performance testing to obtain the IOPS average value r3, and finally calculate the average IOPS of r1, r2, and r3, which is the storage performance data of the test object when the target hit rate is 50%.

[0263] It should be noted that the numerical values ​​listed in this example are only used to illustrate the test process of the specified delay mode provided in the embodiment of the present application. There may be different numerical values ​​in other examples of the embodiment of the present application, and they are not used for specific limitation here.

[0264] Based on the same inventive concept, the present application embodiment also provides a storage performance testing method, such as Figure 8 As shown, the specific implementation process includes S801-S803:

[0265] S801: Obtain the target delay and adjustment coefficient input by the user.

[0266] S802: Perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, a target delay, and an adjustment coefficient.

[0267] S803: Perform a storage performance test on the test object using the target queue depth to obtain storage performance data of the test object.

[0268] In a possible implementation, determining the target queue depth according to the pre-tested test delay, the target delay, and the adjustment coefficient includes:

[0269] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is performed;

[0270] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0271] In a possible implementation, increasing the current queue depth based on the adjustment coefficient to obtain a new queue depth includes:

[0272] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated and the product is rounded up to obtain the new queue depth.

[0273] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain the new queue depth.

[0274] In a possible implementation, reducing the current queue depth by using the adjustment coefficient to obtain a target queue depth includes:

[0275] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the comparison value is rounded down to obtain the target queue depth;

[0276] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0277] In a possible implementation manner, the test duration of the pre-test is shorter than the preset duration, and / or the number of tests is shorter than the preset number.

[0278] In a possible implementation manner, obtaining a target delay and an adjustment coefficient input by a user includes:

[0279] Display test parameter input options to the user, including target delay and adjustment factor;

[0280] Receive the target delay and adjustment factor input by the user through the test parameter input option.

[0281] In a possible implementation, the test parameter input options are presented to the user, including:

[0282] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0283] Based on the same inventive concept, the present application embodiment also provides a storage performance testing method, such as Fig. 9 As shown, the specific implementation process includes S901-S903:

[0284] S901: Obtain the target hit rate and offset input by the user.

[0285] S902: Execute at least one round of storage performance test on the test object to obtain at least one test result, where the test result includes storage performance data of the test object at different hit rates.

[0286] S903: Based on at least one test result, determine the storage performance data of the test object when the hit rate is between the first hit rate and the second hit rate as the storage performance data of the test object at the target hit rate, the first hit rate is the difference between the target hit rate and the offset, and the second hit rate is the sum of the target hit rate and the offset.

[0287] In a possible implementation, the storage performance testing method further includes: obtaining a test round number n input by a user, where n is a positive integer;

[0288] Perform at least one round of storage performance testing on the test object, including:

[0289] Perform n rounds of storage performance tests on the test object and obtain n test results, where each round of storage performance tests is as follows: flush the data in the cache layer of the test object to the capacity layer, perform storage performance tests on the test object, record the correspondence between the real-time hit rate and the test time, and when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, use the correspondence between the real-time hit rate and the test time and the test data of this round of tests as test results.

[0290] In a possible implementation, determining storage performance data when a hit rate of a test object is between a first hit rate and a second hit rate based on at least one test result includes:

[0291] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0292] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate;

[0293] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0294] In a possible implementation, obtaining a target hit rate and an offset input by a user includes:

[0295] Presenting the user with test parameter input options, the test parameters including target hit rate and offset;

[0296] Receives target hit rate and offset input from the user via the test parameter input option.

[0297] In a possible implementation, the test parameter input options are presented to the user, including:

[0298] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0299] Based on the same inventive concept, the present application also provides a storage performance testing device. Fig.10 As shown, it is a schematic diagram of the structure of a storage performance testing device 1000, which may include:

[0300] The interaction unit 1001 is used to respond to the test mode selected by the user in the test mode selection interface, and when it is determined that the selected test mode is the specified delay mode, obtain the target delay and adjustment coefficient input by the user;

[0301] A pre-test unit 1002 is used to perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, a target delay, and an adjustment coefficient;

[0302] The testing unit 1003 is used to perform a storage performance test in the test object using the target queue depth to obtain storage performance data of the test object.

[0303] In a possible implementation manner, the pre-test unit 1002 is specifically configured to:

[0304] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is performed;

[0305] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0306] In a possible implementation manner, the pre-test unit 1002 is specifically configured to:

[0307] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated and the product is rounded up to obtain the new queue depth.

[0308] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain the new queue depth.

[0309] In a possible implementation manner, the pre-test unit 1002 is specifically configured to:

[0310] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the comparison value is rounded down to obtain the target queue depth;

[0311] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0312] In a possible implementation manner, the test duration of the pre-test is shorter than the preset duration, and / or the number of tests is shorter than the preset number.

[0313] In a possible implementation, the interaction unit 1001 is further configured to:

[0314] When it is determined that the selected test mode is the specified hit rate mode, obtaining the target hit rate and offset input by the user;

[0315] The testing unit 1003 is further used to: perform at least one round of storage performance testing on the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates;

[0316] The device also includes: a processing unit 1004, which is used to determine, based on at least one test result, storage performance data of the test object when the hit rate is between a first hit rate and a second hit rate, as storage performance data of the test object at a target hit rate, the first hit rate being the difference between the target hit rate and an offset, and the second hit rate being the sum of the target hit rate and the offset.

[0317] In a possible implementation, the interaction unit 1001 is further used to: obtain a test round number n input by a user, where n is a positive integer;

[0318] The testing unit 1003 is specifically used for:

[0319] Perform n rounds of storage performance tests on the test object and obtain n test results, where each round of storage performance tests is as follows: flush the data in the cache layer of the test object to the capacity layer, perform storage performance tests on the test object, record the correspondence between the real-time hit rate and the test time, and when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, use the correspondence between the real-time hit rate and the test time and the test data of this round of tests as test results.

[0320] In a possible implementation, the processing unit 1004 is specifically configured to:

[0321] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0322] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate;

[0323] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0324] In a possible implementation manner, the interaction unit 1001 is specifically configured to:

[0325] Display test parameter input options to the user, including target delay and adjustment factor;

[0326] Receive the target delay and adjustment factor input by the user through the test parameter input option.

[0327] In a possible implementation manner, the interaction unit 1001 is specifically configured to:

[0328] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0329] Based on the same inventive concept, the present application also provides a storage performance testing device. Fig.11 As shown, it is a schematic diagram of the structure of a storage performance testing device 1100, which may include:

[0330] An acquisition unit 1101 is used to acquire a target delay and an adjustment coefficient input by a user;

[0331] A pre-test unit 1102 is used to perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, a target delay, and an adjustment coefficient;

[0332] The testing unit 1103 is used to perform a storage performance test in the test object using the target queue depth, and obtain storage performance data of the test object.

[0333] In a possible implementation manner, the pre-test unit 1102 is specifically configured to:

[0334] When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is performed;

[0335] When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

[0336] In a possible implementation manner, the pre-test unit 1102 is specifically configured to:

[0337] If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated and the product is rounded up to obtain the new queue depth.

[0338] If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain the new queue depth.

[0339] In a possible implementation manner, the pre-test unit 1102 is specifically configured to:

[0340] If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the comparison value is rounded down to obtain the target queue depth;

[0341] If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

[0342] In a possible implementation manner, the test duration of the pre-test is shorter than the preset duration, and / or the number of tests is shorter than the preset number.

[0343] In a possible implementation, the acquiring unit 1101 is specifically configured to:

[0344] Display test parameter input options to the user, including target delay and adjustment factor;

[0345] Receive the target delay and adjustment factor input by the user through the test parameter input option.

[0346] In a possible implementation, the acquiring unit 1101 is specifically configured to:

[0347] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0348] Based on the same inventive concept, the present application also provides a storage performance testing device. Fig.12 As shown, it is a schematic diagram of the structure of a storage performance testing device 1200, which may include:

[0349] An acquisition unit 1201 is used to acquire a target hit rate and an offset input by a user;

[0350] The testing unit 1202 is used to perform at least one round of storage performance testing on the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates;

[0351] Processing unit 1203 is used to determine, based on at least one test result, storage performance data of the test object when the hit rate is between a first hit rate and a second hit rate, as the storage performance data of the test object at a target hit rate, the first hit rate being the difference between the target hit rate and an offset, and the second hit rate being the sum of the target hit rate and the offset.

[0352] In a possible implementation, the acquisition unit 1201 is further configured to: acquire a test round number n input by a user, where n is a positive integer;

[0353] The testing unit 1202 is specifically used for:

[0354] Perform n rounds of storage performance tests on the test object and obtain n test results, where each round of storage performance tests is as follows: flush the data in the cache layer of the test object to the capacity layer, perform storage performance tests on the test object, record the correspondence between the real-time hit rate and the test time, and when it is determined that the hit rate of the test object reaches a preset value or the test duration is greater than the preset test duration, use the correspondence between the real-time hit rate and the test time and the test data of this round of tests as test results.

[0355] In a possible implementation, the processing unit 1203 is specifically configured to:

[0356] For each test result, the following operations are performed to determine the storage performance data corresponding to each test result: according to the corresponding relationship between the real-time hit rate and the test time in the test result, determine the first time when the hit rate is greater than or equal to the first hit rate for the first time, and the second time when the hit rate is less than or equal to the second hit rate for the first time, and determine the storage performance data between the first time and the second time based on the test data;

[0357] If n is 1, the storage performance data corresponding to the test result is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate;

[0358] If n is greater than 1, an average value of the storage performance data corresponding to the multiple test results is used as the storage performance data of the test object whose hit rate is between the first hit rate and the second hit rate.

[0359] In a possible implementation, the acquiring unit 1201 is specifically configured to:

[0360] Display test parameter input options to the user, including target delay and adjustment factor;

[0361] Receive the target delay and adjustment factor input by the user through the test parameter input option.

[0362] In a possible implementation, the acquiring unit 1201 is specifically configured to:

[0363] The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

[0364] It should be noted that the division of units in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. Each functional unit in the embodiments of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0365] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, the process or function according to the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid state drive (SSD).

[0366] In a simple embodiment, those skilled in the art can imagine that the electronic device in the above embodiment can be used Fig.13 The form shown.

[0367] like Fig.13 The electronic device 1300 shown includes at least one processor 1301 , a memory 1303 , and optionally, a communication interface 1302 .

[0368] The memory 1303 may be a volatile memory, such as a random access memory; the memory may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), or the memory 1303 may be any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 1303 may be a combination of the above memories.

[0369] The specific connection medium between the processor 1301 and the memory 1303 is not limited in the embodiment of the present application.

[0370] The processor 1301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, artificial intelligence chips, chips on chips, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. When the processor 1301 communicates with other devices, data transmission may be performed through the communication interface 1302, such as receiving parameters input by an object.

[0371] When electronic devices use Fig.13 When the form shown is Fig.13 The processor 1301 in the electronic device can call the computer execution instructions stored in the memory 1303 so that the electronic device can execute the storage performance testing method in any of the above method embodiments.

[0372] Specifically, Fig.10 The functions / implementation processes of the interaction unit 1001, the pre-test unit 1002, the test unit 1003, and the processing unit 1004 can be realized by Fig.13 The processor 1301 in the embodiment calls the computer execution instruction stored in the memory 1303 to implement. Or, Fig.11 The functions / implementation processes of the acquisition unit 1101, the pre-test unit 1102, and the test unit 1103 can be Fig.13 The processor 1301 in the embodiment calls the computer execution instruction stored in the memory 1303 to implement. Or, Fig.12 The functions / implementation processes of the acquisition unit 1201, the test unit 1202, and the processing unit 1203 can be Fig.13 The processor 1301 in the embodiment calls the computer execution instructions stored in the memory 1303 to implement.

[0373] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0374] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0375] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0376] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0377] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A storage performance testing method, characterized in that: The method comprises: In response to a test mode selected by a user, when it is determined that the selected test mode is a specified delay mode, obtaining a target delay and an adjustment coefficient input by the user, wherein the specified delay mode is used to test the storage performance of the test object under the target delay; Perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, the target delay, and the adjustment coefficient, wherein the queue depth is used to indicate the number of IO requests sent to the test object; A storage performance test is performed in the test object using the target queue depth to obtain storage performance data of the test object.

2. The method according to claim 1, characterized in that The determining the target queue depth according to the pre-test test delay, the target delay, and the adjustment coefficient includes: When it is determined that the test delay of the pre-test is less than or equal to the target delay, the current queue depth is increased based on the adjustment coefficient to obtain a new queue depth, and the next pre-test is continued; When it is determined that the test delay of the pre-test is greater than the target delay, the current queue depth is reduced by using the adjustment coefficient to obtain the target queue depth.

3. The method according to claim 2, characterized in that The increasing the current queue depth based on the adjustment coefficient to obtain a new queue depth includes: If the adjustment coefficient is greater than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain a new queue depth; If the adjustment coefficient is less than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded up to obtain a new queue depth.

4. The method according to claim 2, characterized in that: The reducing the current queue depth by using the adjustment coefficient to obtain the target queue depth includes: If the adjustment coefficient is greater than 1, the ratio of the current queue depth to the adjustment coefficient is calculated, and the ratio is rounded down to obtain the target queue depth; If the adjustment coefficient is less than 1, the product of the current queue depth and the adjustment coefficient is calculated, and the product is rounded up to obtain the target queue depth.

5. The method according to any one of claims 1 to 4, characterized in that The test duration of the pre-test is less than the preset duration, and / or the number of tests is less than the preset number.

6. The method according to claim 1, characterized in that The method further comprises: When it is determined that the selected test mode is a specified hit rate mode, obtaining the target hit rate and offset input by the user, wherein the specified hit rate mode is used to test the storage performance of the test object under the target hit rate; Perform at least one round of storage performance testing on the test object to obtain at least one test result, wherein the test result includes storage performance data of the test object at different hit rates; Based on the at least one test result, storage performance data of the test object when the hit rate is between the first hit rate and the second hit rate is determined as the storage performance data of the test object at the target hit rate, the first hit rate being the difference between the target hit rate and the offset, and the second hit rate being the sum of the target hit rate and the offset.

7. The method according to claim 1, characterized in that The obtaining the target delay and adjustment coefficient input by the user includes: Presenting a test parameter input option to the user, wherein the test parameters include a target delay and an adjustment factor; Receive the target delay and adjustment coefficient input by the user through the test parameter input option.

8. The method according to claim 7, characterized in that The step of presenting the test parameter input options to the user includes: The test parameter input options are presented to the user through the test parameter input interface or the test parameter input dialog box.

9. A storage performance testing method, characterized in that: The method comprises: Obtain the target delay and adjustment coefficient input by the user; Perform at least one pre-test with an initial queue depth in the test object, and determine a target queue depth according to a test delay of the pre-test, the target delay, and the adjustment coefficient, wherein the queue depth is used to indicate the number of IO requests sent to the test object; A storage performance test is performed in the test object using the target queue depth to obtain storage performance data of the test object.

10. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of any one of the methods of claims 1 to 9.

Images