Hard disk testing method and system

Abstract

The invention relates to the technical field of computers, and discloses a hard disk testing method and system. The hard disk testing method comprises the steps that a plurality of solid state disks are obtained, and the solid state disks are high-speed solid state disks of fourth-generation nonvolatile memories interconnected with enterprise-level high-speed peripheral devices; performing eye pattern detection on the plurality of solid state disks to obtain eye pattern information corresponding to each solid state disk; obtaining a score of an eye pattern state corresponding to the solid state disk according to the eye pattern information; screening the solid state disk based on the score to obtain a to-be-tested solid state disk; performing consistency test on the performance of the to-be-tested solid state disk to obtain performance data of the to-be-tested solid state disk; and performing consistency deviation analysis based on the performance data, and determining a test result of the to-be-tested solid state disk. According to the scheme provided by the embodiment of the invention, the consistency test on the performance of the high-speed solid state disk of the fourth-generation nonvolatile memory interconnected with enterprise-level high-speed peripheral equipment can be realized.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a hard disk testing method and a system thereof. Background Art

[0002] The enterprise environment is complex and changeable, and the rapidly growing business needs have made enterprises put forward higher and higher requirements in terms of data storage scale, storage performance and reliability. The fourth-generation non-volatile memory high-speed solid-state drives (SSDs) with high-speed peripheral device interconnection at the enterprise level have become the preferred solution for more and more enterprises with their extremely high read and write performance and extremely low latency, and play an important role in databases, virtualization, application acceleration, big data, cloud computing and even artificial intelligence. The solid-state drive often needs to operate in a harsh environment with high concurrency, high pressure and 24-hour all-weather operation. Therefore, the performance consistency of the solid-state drive is one of the key concerns of enterprise-level users. It can be seen that in the scenario of high concurrency, high pressure and 24-hour all-weather operation, there is a need to perform consistency testing on the performance of the fourth-generation non-volatile memory high-speed solid-state drives with high-speed peripheral device interconnection at the enterprise level. Therefore, how to achieve consistency testing on the performance of the fourth-generation non-volatile memory high-speed solid-state drives with high-speed peripheral device interconnection at the enterprise level is a problem that needs to be solved urgently. Summary of the invention

[0003] In view of this, the present invention provides a hard disk testing method and system thereof, which can implement consistency testing of the performance of a fourth-generation non-volatile memory high-speed solid-state hard disk interconnected with enterprise-level high-speed peripheral devices.

[0004] In a first aspect, the present invention provides a hard disk testing method, comprising:

[0005] Acquire a plurality of solid-state hard disks, wherein the solid-state hard disks are enterprise-class high-speed peripheral device interconnect fourth-generation non-volatile memory high-speed solid-state hard disks;

[0006] Performing eye diagram detection on the plurality of solid state hard disks to obtain eye diagram information corresponding to each of the solid state hard disks;

[0007] Obtaining a score of an eye diagram state corresponding to the solid state drive according to the eye diagram information;

[0008] Screening the solid state drives based on the scores to obtain solid state drives to be tested;

[0009] Performing a consistency test on the performance of the solid-state hard disk to be tested to obtain performance data of the solid-state hard disk to be tested;

[0010] A consistency deviation analysis is performed based on the performance data to determine a test result of the solid state drive to be tested.

[0011] The hard disk testing method provided by the embodiment of the present invention obtains a plurality of enterprise-level high-speed peripheral interconnected fourth-generation non-volatile memory high-speed solid-state hard disks, performs eye diagram detection on the plurality of solid-state hard disks, obtains eye diagram information corresponding to each solid-state hard disk, and obtains the score of the eye diagram state corresponding to the solid-state hard disk according to the eye diagram information, can screen the solid-state hard disks based on the scores to obtain the solid-state hard disks to be tested, can perform consistency test on the performance of the solid-state hard disks to be tested, and obtain the performance data of the solid-state hard disks to be tested, thereby realizing consistency test on the performance of the fourth-generation non-volatile memory high-speed solid-state hard disks interconnected with enterprise-level high-speed peripheral devices, finally, performs consistency deviation analysis based on the performance data, determines the test results of the solid-state hard disks to be tested, and can ensure the consistency of the performance of the fourth-generation non-volatile memory high-speed solid-state hard disks interconnected with enterprise-level high-speed peripheral devices.

[0012] In some optional implementations, performing consistency deviation analysis based on the performance data to determine a test result of the solid state drive to be tested includes:

[0013] For various types of performance data, remove the maximum value and the minimum value in the performance data to obtain the performance data to be processed;

[0014] Calculating an average value of the performance data to be processed;

[0015] Based on the average value and the performance data, a test result of the solid state drive to be tested is determined.

[0016] The hard disk testing method provided by the embodiment of the present invention can obtain the performance data to be processed by removing the maximum value and the minimum value in the performance data of various types, and calculate the average value of the performance data to be processed. It can then determine the average level of the performance of the solid-state hard disk to be tested based on the average value and the performance data, and can determine the test result of the solid-state hard disk to be tested. It can also determine the consistency test result of the solid-state hard disk to be tested, and then effectively block the solid-state hard disks with large performance fluctuations in this batch from entering the market based on the test result, which is beneficial to improving user experience and the quality of solid-state hard disks.

[0017] In some optional implementations, determining the test result of the solid state drive to be tested based on the average value and the performance data includes:

[0018] obtaining a first deviation limit and a second deviation limit based on the average value, wherein the first deviation limit is greater than the second deviation limit;

[0019] Comparing the performance data with the first deviation limit and the second deviation limit respectively to obtain a comparison result;

[0020] The test result of the solid state drive to be tested is determined according to the comparison result.

[0021] The hard disk testing method provided by the embodiment of the present invention obtains the first deviation limit and the second deviation limit based on the average value, and compares the performance data with the first deviation limit, and compares the performance data with the second deviation limit to obtain a comparison result, which can determine the degree of deviation of the performance of the solid-state hard disk to be tested, and then determine the test result of the solid-state hard disk to be tested based on the comparison result, so as to determine whether the performance of the solid-state hard disk to be tested meets the consistency standard.

[0022] In some optional implementations, determining the test result of the solid state drive to be tested according to the comparison result includes:

[0023] When the comparison result indicates that the performance data is less than or equal to the first deviation limit and greater than or equal to the second deviation limit, it is determined that the corresponding test result of the solid state drive to be tested meets the consistency standard.

[0024] The hard disk testing method provided in the embodiment of the present invention determines the test results of the corresponding solid-state hard disk to be tested by comparing the results. Only when the test results meet the consistency standards can the solid-state hard disk to be tested be put on the market. In this way, the requirements of high concurrency, high pressure, and 24-hour all-weather operation scenarios can be met.

[0025] In some optional implementations, determining the test result of the solid state drive to be tested according to the comparison result includes:

[0026] When the comparison result indicates that the performance data is greater than the first deviation limit, or less than the second deviation limit, it is determined that the corresponding test result of the solid state drive to be tested does not meet the consistency standard.

[0027] The hard disk testing method provided in the embodiment of the present invention determines the test result of the corresponding solid-state hard disk to be tested by comparing the results. When the test result is that the consistency does not meet the standard, the solid-state hard disk to be tested that does not meet the consistency standard will be intercepted to prevent the solid-state hard disk to be tested from entering the market.

[0028] In some optional implementations, performing consistency testing on the performance of the solid-state drive to be tested to obtain performance data of the solid-state drive to be tested includes:

[0029] Get the current test type;

[0030] Determining configuration parameters of the solid state drive to be tested based on the current test type;

[0031] Configuring the solid state drive to be tested based on the configuration parameters;

[0032] The configured solid state drive to be tested is tested according to the current test type to obtain the performance data.

[0033] The hard disk testing method provided by the embodiment of the present invention obtains the current test type, determines the configuration parameters of the solid-state hard disk to be tested based on the current test type, and configures the solid-state hard disk to be tested based on the configuration parameters. It can perform the current test type test on the configured solid-state hard disk to be tested and obtain performance data, so that subsequent steps can make consistency judgments on the performance of the solid-state hard disk to be tested based on the performance data.

[0034] In some optional implementations, the current test type includes at least one of a sequential write test, a sequential read test, a random read test, and a random write test.

[0035] In some optional implementations, the calculation formula of the first deviation limit is as follows:

[0036] Hmax=μ*c1

[0037] Wherein, Hmax is the first deviation limit, μ is the average value, and c1 is a constant.

[0038] In some optional implementations, the calculation formula for the second deviation limit is as follows:

[0039] Hmin=μ*c2

[0040] Wherein, Hmin is the second deviation limit, μ is the average value, and c2 is a constant.

[0041] In a second aspect, the present invention provides a hard disk testing system, comprising:

[0042] Multiple solid-state drives;

[0043] A testing device, wherein the testing device is communicatively connected to the multiple solid-state hard disks, and the testing device performs consistency testing on the multiple solid-state hard disks by executing the hard disk testing method described in the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A schematic diagram of an optional hard disk testing process provided in an embodiment of the present application;

[0046] Figure 2 is a flow chart of another optional hard disk testing process according to an embodiment of the present invention;

[0047] Figure 3 is a flow chart of a hard disk testing method according to an embodiment of the present invention;

[0048] Figure 4 is a schematic diagram of eye diagram information according to an embodiment of the present invention;

[0049] Figure 5 is a flow chart of another hard disk testing method according to an embodiment of the present invention;

[0050] Figure 6 is a schematic diagram of a test result of a solid state drive to be tested according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0052] It is worth noting that with the widespread application of NVME (Non-Volatile Memory express) hard disks in the server field, the NVMe standard can bring about performance improvements in many aspects. As an SSD of the PCIE (Peripheral Component Interconnect Express) channel, the NVME hard disk specification is bound to gradually replace the current AHCI (Serial ATA Advanced Host Controller Interface) standard SSD in the future. Among them, NVME is a protocol similar to AHCI based on the M.2 interface, which is a protocol specially designed for flash storage. PCIE, namely PCI-Express, is a high-speed serial computer expansion bus standard.

[0053] The enterprise environment is complex and ever-changing, and the rapidly growing business needs have led to higher and higher requirements for data storage scale, storage performance, reliability, and other aspects. Solid-state drives have become the preferred solution for more and more enterprises with their extremely high read and write performance and extremely low latency, and play an important role in databases, virtualization, application acceleration, big data, cloud computing, and even artificial intelligence. Enterprise-level SSDs often need to operate in harsh environments with high concurrency, high pressure, and 24-hour operation. Their consistency and reliability are one of the key concerns of enterprise-level users.

[0054] Reliability refers to the ability of a component or system to continue to perform its intended function within a specific period of time under specified operating conditions. For enterprise-level SSDs, it is a very important indicator that not only directly determines the core indicators such as the yield and failure rate of product shipments, but also plays a key role in protecting data availability and consistency.

[0055] Generally, the reliability of SSD is measured by MTBF (Mean Time Between Failure). MTBF is the ratio of the cumulative working time of the product to the number of failures during the total use period. It reflects the time quality of the product. That is, the fewer product failures, the higher the MTBF and the higher the product reliability.

[0056] Compared with consumer-grade SSD products, enterprise-grade SSDs face higher challenges in reliability and consistency.

[0057] For NVMe SSDs that do not meet the monitoring standards or have large consistency fluctuations in consistency testing, they need to be intercepted and replaced in a timely manner at the production end, and the reasons for the consistency fluctuations need to be analyzed to see if it is a batch problem to prevent them from entering the market.

[0058] In the related technology, after the NVMe Gen4 (i.e., the fourth generation NVME specification) self-developed solid-state drive SMT (Surface Mounted Technology) and assembly, the enterprise-level high-performance NVMe solid-state drive must pass FT1 and FT2 before it can be shipped. Specifically, the factory production process of enterprise-level high-performance NVMe solid-state drives includes FT1 (Factory Test 1, the first stage test in the factory process), BIST (Built-in Self-Test, aging test in the factory process) and FT2 (Factory Test 2, the second stage test in the factory process), among which the FT1 process includes burning firmware and detecting the hardware link status of the solid-state drive, the BIST process includes aging the hard disk components nand (i.e., flash memory chip material) and DDR (Double Data Rate), and the FT2 process includes system testing, performance testing and business simulation testing. Therefore, the performance of the solid-state drive is not tested for consistency in the entire factory testing process, and the consistency of the performance of the solid-state drive cannot be guaranteed.

[0059] Therefore, in order to prevent SSDs with large fluctuations in hard disk performance consistency or SSDs with problems in the same batch from entering the market and affecting the user experience who have high requirements for SSD consistency, a consistency test station, namely DCT (Disk Consistency Test), is added to the factory testing process. In this way, the consistency test of SSD performance can be automated, which can reduce manpower consumption and improve product quality.

[0060] Reference Figure 1 , Figure 1 A flow chart of an optional hard disk testing process is provided for an embodiment of the present application, wherein the hard disk testing process includes a first-stage test in a factory process, an aging test in a factory process, a second-stage test in a factory process, and a hard disk consistency test.

[0061] Among them, the first stage test in the factory process and the aging test in the factory process are carried out in the first stage test station in the factory process, the second stage test in the factory process is carried out in the second stage test station in the factory process, and the hard disk consistency test is carried out in the hard disk consistency test station. Figure 2First, the SSD is assembled using surface mount technology, where surface mount technology includes offline burning and assembly of the SSD. Offline burning means that before the disk is mounted on the surface, the firmware is burned into EEPROM (Electrically Erasable Programmable read only memory) or NAND through a burning fixture, and then the burned SSD is assembled. In this way, the SSD can directly run the corresponding program.

[0062] Then, enter the factory test, i.e., the diagnostic test. Specifically, the first stage test station in the factory process includes the first stage test in the factory process and the aging test in the factory process, wherein the first stage test in the factory process includes burning BIST firmware, PCIE link monitoring, state monitoring of each sensor of the SSD, and writing the basic information of the SSD, the basic information of the SSD includes the serial number (i.e., SN) of the SSD, the model (i.e., MN) of the SSD, and the configuration parameters required for the aging test in the factory process (such as the number of aging rounds, etc.); the aging test in the factory process includes DDR testing, NAND aging, polling NAND particles, screening out NAND bad blocks, and SSDs with hardware assembly problems such as high temperature or cold soldering. It is understandable that when NAND leaves the factory, it contains bad blocks. When performing SMT, NAND is welded to the PCB (Printed circuit board) of the SSD, and then when the SSD leaves the factory, a diagnostic test is performed to screen out bad particles and SSDs with hardware assembly problems such as cold soldering during the assembly process.

[0063] Then, the power is cut off and the plug is removed. Currently, the machine is controlled by BMC to realize automatic power off and power on. In the first phase test station in the factory process, the basic information of the SSD needs to be tested. For the SSDs that pass the test, subsequent test items can be carried out, and the SSDs that fail the test will be intercepted and analyzed to avoid the shipment of defective products. Specifically, FT2 includes PCIE link monitoring, burning user firmware, detecting smart logs, vendor logs, and the bad block table detects whether the NAND bad blocks exceed the standard, and then performs business scenario simulation tests, and finally performs factory formatting to restore the factory state. Among them, the bad block table is used to record the bad block status, and the business scenario refers to the scenario that simulates user use, including but not limited to small / large block data reading and writing, random reading and writing, etc.

[0064] Then, the power is turned off and the plug is removed again until the hard disk consistency test station is entered. Since the test time of the hard disk consistency test station is very long, considering the impact of test efficiency, test resources and production capacity, a random test is adopted, for example, the random test ratio is 10 / 2000. The hard disk consistency test in the hard disk consistency test station is described in detail in the following method steps.

[0065] Finally, the equipment server exchanges information with the first-stage test station in the factory process, the second-stage test station in the factory process, and the hard disk consistency test station through the network. At the same time, the equipment server also exchanges information with the Manufacturing Execution System (MES) and the preset information system through the network. The Manufacturing Execution System is used to monitor, track and control the production process in real time, and the preset information system is used to synchronously transmit information back to the preset user.

[0066] Furthermore, the embodiment of the present invention provides a hard disk testing method and system thereof, which can implement consistency testing of the performance of a fourth-generation non-volatile memory high-speed solid-state hard disk interconnected with enterprise-level high-speed peripheral devices.

[0067] According to an embodiment of the present invention, a hard disk testing method embodiment is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0068] In this embodiment, a hard disk testing method is provided, which can be used in the hard disk testing process described above. Specifically, a testing device is used to perform consistency testing on a solid state hard disk. Figure 3 is a flow chart of a hard disk testing method according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0069] Step S301, obtaining multiple solid state drives.

[0070] Among them, the solid-state drive is an enterprise-level high-speed peripheral device interconnected fourth-generation non-volatile memory high-speed solid-state drive.

[0071] In addition, the solid state drive is an nvme solid state drive that has passed the detection of the basic information of the solid state drive in the FT2 workstation. Specifically, in the test environment, a fixed station for the solid state drive is set, and multiple solid state drives are distributed and set on the corresponding fixed stations, and then the test equipment is connected to the solid state drive for communication, and the consistency test of multiple solid state drives is realized through the communication connection between the two.

[0072] Step S302: perform eye diagram detection on multiple solid state drives to obtain eye diagram information corresponding to each solid state drive.

[0073] Eye diagram detection is performed on multiple solid-state drives. Specifically, the eye diagram information corresponding to each solid-state drive can be obtained by passing the parameter pcie-cross-hair-info through a crawler tool. The eye diagram information is as follows: Figure 4 In addition, the eye diagram detection can be an electronic eye diagram detection.

[0074] Step S303, obtaining a score of the eye diagram state corresponding to the solid state drive according to the eye diagram information.

[0075] The eye diagram information includes the W2H Score, which is the score of the eye diagram state corresponding to the solid-state drive obtained by the W2H analysis method. For example, Figure 4 The W2H Scores in are 16590, 15960, 14035 and 12648 respectively. Therefore, the score of the eye diagram state corresponding to the solid state drive can be obtained according to the eye diagram information.

[0076] Step S304, screening the solid state drives based on the scores to obtain the solid state drives to be tested.

[0077] SSDs can be screened based on the score, which can be to screen out SSDs with a score W2H Score greater than a preset score, where the preset score can be 2000, 2500 or other scores. Figure 4 As shown, solid-state drives with a score of W2HScore greater than 2000 are screened out.

[0078] Step S305 , performing consistency test on the performance of the solid state drive to be tested, and obtaining performance data of the solid state drive to be tested.

[0079] Performance data can include iops and bandwidth, where iops stands for iopersecond, that is, the number of io operations per second.

[0080] The performance of the SSD to be tested is tested for consistency, specifically, a serial test is performed on the performance of the SSD to be tested for consistency, for example, a sequential write test, a sequential read test, a random read test, and a random write test are performed on the performance of the SSD to be tested in sequence. The reason for performing a serial test on the performance of the SSD to be tested for consistency, rather than a parallel test, is that a parallel test will affect the performance data of the SSD to be tested, that is, the performance data obtained by the parallel test is inaccurate.

[0081] In some optional implementations, before the SSD to be tested is subjected to consistency testing, the SSD to be tested may be preprocessed, that is, preset configuration parameters of the SSD to be tested are determined, the SSD to be tested is configured based on the preset configuration parameters, and then the configured SSD to be tested is subjected to 2 rounds of 128K sequential write tests, wherein the preset configuration parameters may be ioengine=libaio (i.e., the io engine adopts the libaio mode), iodepth=128 (i.e., the queue depth is 128), bs=128k (i.e., the block size is 128k), rw=write (i.e., sequential write operations are performed), and loops=2 (i.e., 2 rounds of loops). In addition, the preprocessing of the SSD to be tested may be parallel preprocessing of the SSD to be tested, that is, preprocessing of multiple SSDs to be tested simultaneously to improve the efficiency of SSD preprocessing.

[0082] It should be noted that if the SSD to be tested is not preprocessed, the entire test may fall into the "behavior trap" of the SSD, that is, if the SSD to be tested is not preprocessed, and the SSD to be tested is directly tested or the place where no writing is done is tested, the SSD's FTL (Flash Translation Layer) may still not mark this space on the particle as written, then the test result is that the test is not written to the disk at all (that is, no reading or writing to the storage medium itself), and the SSD master controller directly feedbacks 00 or FF, even if the preprocessing is not done well, the SSD may still find a pattern that makes the test result wrong. It can be seen that it is particularly important to preprocess the SSD to be tested before performing consistency testing on the SSD to be tested.

[0083] Step S306: Perform consistency deviation analysis based on the performance data to determine the test result of the solid state drive to be tested.

[0084] The consistency deviation analysis based on performance data is to determine the degree of deviation of the performance data, and to determine whether the corresponding SSD to be tested meets the consistency standard through the degree of deviation of the performance data of each SSD to be tested. The test results of the SSD to be tested are used to indicate whether the SSD to be tested meets the consistency standard.

[0085] The hard disk testing method provided in this embodiment obtains a plurality of enterprise-level high-speed peripheral interconnected fourth-generation non-volatile memory high-speed solid-state hard disks, performs eye diagram detection on the plurality of solid-state hard disks, obtains eye diagram information corresponding to each solid-state hard disk, and obtains the score of the eye diagram state corresponding to the solid-state hard disk according to the eye diagram information, can screen the solid-state hard disks based on the scores to obtain the solid-state hard disks to be tested, can obtain the solid-state hard disks whose signal quality meets the requirements, and then performs consistency testing on the performance of the solid-state hard disks to be tested to obtain performance data of the solid-state hard disks to be tested, and performs consistency deviation analysis based on the performance data to determine the test results of the solid-state hard disks to be tested, thereby realizing consistency testing on the performance of the fourth-generation non-volatile memory high-speed solid-state hard disks interconnected with enterprise-level high-speed peripheral devices, and ensuring the consistency of the performance of the fourth-generation non-volatile memory high-speed solid-state hard disks interconnected with enterprise-level high-speed peripheral devices.

[0086] In this embodiment, a hard disk testing method is provided, which can be used in the above-mentioned hard disk testing process. Figure 5 is a flow chart of a hard disk testing method according to an embodiment of the present invention. Figure 5 As shown, the process includes the following steps:

[0087] Step S501, obtaining multiple solid state drives.

[0088] Among them, the solid state drive is an enterprise-level high-speed peripheral device interconnection fourth-generation non-volatile memory high-speed solid state drive. For details, please refer to Figure 3 Step S301 of the illustrated embodiment will not be described in detail here.

[0089] Step S502 , performing eye diagram detection on a plurality of solid state drives to obtain eye diagram information corresponding to each solid state drive.

[0090] For details, please see Figure 3 Step S302 of the illustrated embodiment will not be described in detail here.

[0091] Step S503, obtaining a score of the eye diagram state corresponding to the solid state drive according to the eye diagram information.

[0092] For details, please see Figure 3 Step S303 of the illustrated embodiment will not be described in detail here.

[0093] Step S504, screening the solid state drives based on the scores to obtain the solid state drives to be tested.

[0094] For details, please see Figure 3 Step S304 of the illustrated embodiment will not be described in detail here.

[0095] Step S505, performing consistency test on the performance of the solid state drive to be tested to obtain performance data of the solid state drive to be tested;

[0096] Specifically, the above step S505 includes:

[0097] Step S5051, obtain the current test type.

[0098] The current test type includes at least one of a sequential write test, a sequential read test, a random read test, and a random write test.

[0099] Step S5052: Determine the configuration parameters of the solid state drive to be tested based on the current test type.

[0100] Different current test types have different configuration parameters. For example, if the current test type is a 1-hour 128K sequential write test, the corresponding configuration parameters of the SSD to be tested can be ioengine=libaio (i.e. the io engine uses libaio), iodepth=128 (i.e. queue depth 128), bs=128k (i.e. block size 128k), rw=write, time_based=1, runtime=1h (i.e. test time is 1 hour), ramp_time=120. Ramp_time is used to run the device for a while without recording. time_based can prevent the task from ending early.

[0101] For another example, the current test type is a 0.5-hour 128K sequential read test, and the corresponding configuration parameters of the solid-state drive to be tested may be ioengine=libaio, iodepth=128, bs=128k, rw=read, time_based=1, runtime=30m, ramp_time=120.

[0102] For another example, the current test type is a 1-hour 4K random read test, and the corresponding configuration parameters of the solid-state drive to be tested may be ioengine=libaio, iodepth=128, bs=4k, rw=randread, numjobs=4, time_based=1, runtime=1h, ramp_time=120, group_reporting=1.

[0103] For another example, the current test type is a 2-hour 4K random write test, and the corresponding configuration parameters of the solid-state drive to be tested may be ioengine=libaio, iodepth=64, bs=4k, rw=randwrite, numjobs=4, time_based=1, runtime=2h, ramp_time=120, group_reporting=1.

[0104] For another example, the current test type is a 1-hour 4K random write test, and the corresponding configuration parameters of the solid-state drive to be tested may be ioengine=libaio, iodepth=64, bs=4k, rw=randwrite, numjobs=4, time_based=1, runtime=1h, ramp_time=120, group_reporting=1.

[0105] In addition, for different solid state drives to be tested, if the corresponding current test types are the same, then their configuration parameters may be the same.

[0106] Step S5053: configure the solid state drive to be tested based on the configuration parameters.

[0107] The test tool of the solid state drive to be tested is configured using the configuration parameters, so that the solid state drive to be tested can be tested using the test tool with the configuration parameters in subsequent steps.

[0108] Step S5054, performing a test of the current test type on the configured solid state drive to be tested to obtain performance data.

[0109] It is understandable that different current test types correspond to different performance data. In addition, for some test types, testing the configured SSD to be tested may not necessarily obtain performance data. For example, when a 2-hour 4K random write test is performed on the configured SSD to be tested, and the corresponding configuration parameters of the SSD to be tested are ioengine=libaio,iodepth=64,bs=4k,rw=randwrite,numjobs=4,time_based=1,runtime=2h,ramp_time=120,group_reporting=1, no performance data is obtained.

[0110] In an embodiment of the present application, by obtaining the current test type, determining the configuration parameters of the solid-state drive to be tested based on the current test type, and configuring the solid-state drive to be tested based on the configuration parameters, the configured solid-state drive to be tested can be tested for the current test type to obtain performance data, so that subsequent steps can make consistency judgments on the performance of the solid-state drive to be tested based on the performance data.

[0111] Step S506: Perform consistency deviation analysis based on the performance data to determine the test result of the solid state drive to be tested.

[0112] Specifically, the above step S506 includes:

[0113] Step S5061: for various types of performance data, remove the maximum value and the minimum value in the performance data to obtain the performance data to be processed.

[0114] The types of performance data include IOPS and bandwidth. When there are multiple solid state drives to be tested, the performance data in the step is a collection of performance data of all solid state drives to be tested.

[0115] For example, assuming that the performance data is IOPS and there are 10 SSDs to be tested, the performance data is 10 IOPS values. Alternatively, assuming that the performance data is bandwidth and there are 8 SSDs to be tested, the performance data is 8 bandwidth values.

[0116] Step S5062, calculating the average value of the performance data to be processed.

[0117] The average value of the performance data to be processed refers to the average value of a certain type of performance data, for example, the average value of IOPS or the average value of bandwidth.

[0118] Step S5063: Determine the test result of the solid state drive to be tested based on the average value and the performance data.

[0119] The average value can reflect the general level of the overall performance of the solid state drives to be tested. Based on the average value and the performance data, the degree of deviation of the performance of the solid state drives to be tested can be determined.

[0120] In an embodiment of the present application, by removing the maximum and minimum values ​​in various types of performance data to obtain the performance data to be processed, and calculating the average value of the performance data to be processed, the average level of the performance of the solid-state hard disk to be tested can be determined, and then based on the average value and the performance data, the test result of the solid-state hard disk to be tested can be determined, and the consistency test result of the solid-state hard disk to be tested can be determined, and then according to the test result, the solid-state hard disks with large performance fluctuations in this batch can be effectively blocked from entering the market, which is beneficial to improving user experience and the quality of solid-state hard disks.

[0121] In some optional implementations, the above step S5063 includes:

[0122] Step a1, obtaining a first deviation limit and a second deviation limit based on the average value.

[0123] Among them, the first deviation limit is greater than the second deviation limit. The calculation formula of the first deviation limit is as follows:

[0124] Hmax=μ*c1

[0125] Wherein, Hmax is the first deviation limit, μ is the average value, and c1 is a constant. For example, c1 can be 1.05 or other values, which can be adjusted according to actual test conditions.

[0126] The second deviation limit is calculated as follows:

[0127] Hmin=μ*c2

[0128] Wherein, Hmin is the second deviation limit, μ is the average value, and c2 is a constant. For example, c2 can be 0.95 or other values, which can be adjusted according to actual test conditions.

[0129] Step a2: Compare the performance data with the first deviation limit and the second deviation limit respectively to obtain a comparison result.

[0130] The performance data is compared with the first deviation limit, and the performance data is compared with the second deviation limit to obtain a comparison result.

[0131] Step a3: determining the test result of the solid state drive to be tested according to the comparison result.

[0132] The comparison result is used to indicate the relationship between the performance data and the first deviation limit and the second deviation limit. The comparison result can be used to determine whether the test result of the solid state drive to be tested meets the consistency standard.

[0133] In an embodiment of the present application, a first deviation limit and a second deviation limit are obtained based on the average value, and the performance data are compared with the first deviation limit, and the performance data are compared with the second deviation limit to obtain a comparison result, which can determine the degree of deviation of the performance of the solid-state hard disk to be tested, and then determine the test result of the solid-state hard disk to be tested based on the comparison result, so as to determine whether the performance of the solid-state hard disk to be tested meets the consistency standard.

[0134] In some optional implementations, step a3 specifically includes:

[0135] Step a31, when the comparison result indicates that the performance data is less than or equal to the first deviation limit and greater than or equal to the second deviation limit, it is determined that the test result of the corresponding solid state drive to be tested is up to the consistency standard.

[0136] If the test type is a sequential read test or a sequential write test, the corresponding test results only focus on whether the bandwidth consistency meets the standard; if the test type is a random write test or a random read test, the corresponding test results only focus on whether the IOPS consistency meets the standard.

[0137] When the test results of all test types of the SSD to be tested meet the consistency standard, the SSD to be tested meets the consistency standard. Figure 6 As shown, the test types corresponding to the solid-state drive to be tested include sequential read test (i.e., write), sequential write test (i.e., read), random write test (i.e., randread) and random read test (i.e., randwrite). Then, when the bandwidth (i.e., BW) consistency obtained by the sequential read test meets the standard (i.e., standard), the bandwidth consistency obtained by the sequential write test meets the standard, the IOPS consistency obtained by the random write test meets the standard, and the IOPS consistency obtained by the random read test meets the standard, the solid-state drive to be tested meets the consistency standard.

[0138] In an embodiment of the present application, the test results of the corresponding solid-state hard disk to be tested are determined by comparing the results. Only when the test results meet the consistency standards can the solid-state hard disk to be tested be put on the market. In this way, the requirements of high concurrency, high pressure, and 24-hour all-weather operation scenarios can be met.

[0139] In some optional implementations, step a3 specifically further includes:

[0140] Step a32: When the comparison result indicates that the performance data is greater than the first deviation limit, or less than the second deviation limit, it is determined that the test result of the corresponding solid state drive to be tested is not up to the consistency standard.

[0141] If the test result corresponding to the current test type is that the consistency does not meet the standard, no subsequent tests will be performed. For example, the performance of the SSD to be tested is sequentially written, sequentially read, randomly read, and randomly written. If the current test type is sequential write test, and the corresponding test result is that the consistency does not meet the standard, then the performance of the SSD to be tested will no longer be sequentially read, random read, and random write tested, which can improve the test efficiency.

[0142] In addition, SSDs to be tested that fail to meet the consistency test results will be intercepted to prevent them from entering the market.

[0143] In an embodiment of the present application, the test result of the corresponding solid-state hard disk to be tested is determined by comparing the results. When the test result is that the consistency does not meet the standard, the solid-state hard disk to be tested that does not meet the consistency standard will be intercepted to prevent the solid-state hard disk to be tested from entering the market.

[0144] In order to more clearly illustrate the process of performing consistency deviation analysis based on performance data to determine the test result of the solid state drive to be tested, a specific example is used below to illustrate.

[0145] Taking the performance data IOPS and the current test type of 1-hour 128K sequential write test as an example, assuming that the number of solid-state drives to be tested is 10, then the 10 IOPS values ​​are (X0, X1, ..., X9). At this time, remove the maximum value X3 and the minimum value X6 among the 10 IOPS values ​​to obtain the performance data to be processed (X0, X1, X2, X4, X5, X7, X8, X9).

[0146] Next, the average value of the performance data to be processed is calculated, that is, μ=(X0, X1, X2, X4, X5, X7, X8, X9) / 8.

[0147] Then, the first deviation limit and the second deviation limit are obtained based on the average value, that is, the second deviation limit Hmin=μ*0.95 and the first deviation limit Hmax=μ*1.05.

[0148] In this way, the deviation degree Xi / μ (percentage) of each SSD can be calculated, and the variance of the performance data can be calculated, so that the performance of the SSD can be analyzed later. The calculation formula of the variance of the performance data is as follows:

[0149]

[0150] Wherein, μ is the average value of the performance data to be processed, N is the total number of solid-state drives, and X is the IOPS value of the solid-state drive.

[0151] Finally, the performance data is compared with the first deviation limit and the second deviation limit respectively. If the performance data is compared with the first deviation limit and the second deviation limit respectively, that is, Hmin<=Xi<=Hmax, it is considered that the 1-hour 128K sequential write test consistency of the solid state drive meets the standard.

[0152] Furthermore, if the test results corresponding to all test types of the SSD meet the consistency standards, the basic status of the SSD, the vendor log and the smart log, and the factory recovery are tested again.

[0153] An embodiment of the present invention also provides a hard disk testing system, which includes multiple solid-state hard disks and a testing device. The testing device is communicatively connected to the multiple solid-state hard disks. The testing device performs consistency testing on the multiple solid-state hard disks by executing the hard disk testing method in any of the above embodiments.

[0154] Specifically, in the test environment, a fixed workstation for solid-state hard disks is set up, and multiple solid-state hard disks are distributed on the corresponding fixed workstations. The test equipment is then communicatively connected to the solid-state hard disks, and consistency testing of multiple solid-state hard disks is achieved through the communication connection between the two.

[0155] Therefore, the hard disk testing system has the beneficial effects brought about by the hard disk testing method in any of the above-mentioned embodiments, namely, performing eye diagram detection on multiple solid-state hard disks to obtain eye diagram information corresponding to each solid-state hard disk, and obtaining the score of the eye diagram state corresponding to the solid-state hard disk according to the eye diagram information, and being able to screen the solid-state hard disks based on the scores to obtain the solid-state hard disks to be tested, and being able to obtain the solid-state hard disks whose signal quality meets the requirements, and then performing consistency testing on the performance of the solid-state hard disks to be tested to obtain the performance data of the solid-state hard disks to be tested, and performing consistency deviation analysis based on the performance data to determine the test results of the solid-state hard disks to be tested, thereby achieving consistency testing of the performance of the fourth-generation non-volatile memory high-speed solid-state hard disks interconnected with enterprise-level high-speed peripheral devices, thereby ensuring the consistency of the performance of the fourth-generation non-volatile memory high-speed solid-state hard disks interconnected with enterprise-level high-speed peripheral devices.

[0156] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A hard disk testing method, characterized in that: The method comprises: Acquire a plurality of solid-state hard disks, wherein the solid-state hard disks are enterprise-class high-speed peripheral device interconnect fourth-generation non-volatile memory high-speed solid-state hard disks; Performing eye diagram detection on the plurality of solid state hard disks to obtain eye diagram information corresponding to each of the solid state hard disks; Obtaining a score of an eye diagram state corresponding to the solid state drive according to the eye diagram information; Screening the solid state drives based on the scores to obtain solid state drives to be tested; Performing a consistency test on the performance of the solid state drive to be tested to obtain performance data of the solid state drive to be tested; A consistency deviation analysis is performed based on the performance data to determine a test result of the solid state drive to be tested.

2. The method according to claim 1, characterized in that The performing consistency deviation analysis based on the performance data to determine the test result of the solid state drive to be tested includes: For various types of performance data, remove the maximum value and the minimum value in the performance data to obtain the performance data to be processed; Calculating an average value of the performance data to be processed; Based on the average value and the performance data, a test result of the solid state drive to be tested is determined.

3. The method according to claim 2, characterized in that The step of determining a test result of the solid state drive to be tested based on the average value and the performance data includes: obtaining a first deviation limit and a second deviation limit based on the average value, wherein the first deviation limit is greater than the second deviation limit; Comparing the performance data with the first deviation limit and the second deviation limit respectively to obtain a comparison result; The test result of the solid state drive to be tested is determined according to the comparison result.

4. The method according to claim 3, characterized in that Determining the test result of the solid state drive to be tested according to the comparison result includes: When the comparison result indicates that the performance data is less than or equal to the first deviation limit and greater than or equal to the second deviation limit, it is determined that the corresponding test result of the solid state drive to be tested meets the consistency standard.

5. The method according to claim 3, characterized in that: Determining the test result of the solid state drive to be tested according to the comparison result includes: When the comparison result indicates that the performance data is greater than the first deviation limit, or less than the second deviation limit, it is determined that the corresponding test result of the solid state drive to be tested does not meet the consistency standard.

6. The method according to claim 1, characterized in that The performing consistency test on the performance of the solid state drive to be tested to obtain performance data of the solid state drive to be tested includes: Get the current test type; Determining configuration parameters of the solid state drive to be tested based on the current test type; Configuring the solid state drive to be tested based on the configuration parameters; The configured solid state drive to be tested is tested according to the current test type to obtain the performance data.

7. The method according to claim 6, characterized in that The current test type includes at least one of a sequential write test, a sequential read test, a random read test, and a random write test.

8. The method according to claim 3, characterized in that The calculation formula of the first deviation limit is as follows: Hmax=μ*c1 Wherein, Hmax is the first deviation limit, μ is the average value, and c1 is a constant.

9. The method according to claim 3, characterized in that: The calculation formula of the second deviation limit is as follows: Hmin=μ*c2 Wherein, Hmin is the second deviation limit, μ is the average value, and c2 is a constant.

10. A hard disk testing system, characterized in that: include: Multiple solid-state drives; A testing device, wherein the testing device is communicatively connected to the multiple solid-state hard disks, and the testing device performs consistency testing on the multiple solid-state hard disks by executing the hard disk testing method described in any one of claims 1 to 9.

Images