CN107680632B

CN107680632B - Method and device for testing service life of solid state disk

Info

Publication number: CN107680632B
Application number: CN201610622202.9A
Authority: CN
Inventors: 张艾艾; 丁金德
Original assignee: Datang Mobile Communications Equipment Co Ltd
Current assignee: Datang Mobile Communications Equipment Co Ltd
Priority date: 2016-08-01
Filing date: 2016-08-01
Publication date: 2020-10-16
Anticipated expiration: 2036-08-01
Also published as: CN107680632A

Abstract

The application discloses a method and a device for testing the service life of a solid state disk, which are used for improving the testing efficiency and verifying whether the master control of the solid state disk supports a static Wear-Leveling algorithm or not, so that the testing result is more accurate. The application provides a method for testing the service life of a solid state disk, which comprises the following steps: respectively writing the cold data and the hot data into a solid state disk to be tested, and storing the cold data into another device for backup; continuously writing the hot data into the solid state disk, and recording the writing times of the hot data; and comparing whether the cold data in the solid state disk is consistent with the backup cold data according to the writing times of the hot data and a preset period, if so, continuously writing the hot data into the solid state disk, and otherwise, finishing the test.

Description

Method and device for testing service life of solid state disk

Technical Field

The present application relates to the field of electronic technologies, and in particular, to a method and an apparatus for testing a lifetime of a solid state disk.

Background

At present, the technology of Solid State Drives (SSD) has leap forward, and especially in terms of life, has reached the acceptable range for consumer users. However, for enterprise users, although SSD has no mechanical parts and is more stable, the problem of lifetime still causes concern to many enterprises. Therefore, it is necessary to have a corresponding life test technology to determine whether the life of the SSD can meet the enterprise requirement.

Compared with a traditional mechanical Hard Disk (HDD), a Flash memory (NAND Flash) -based SSD solid state Disk has the advantages of fast speed, small volume, low power consumption, no noise, and the like, and is popular among large enterprises and consumers. However, NAND Flash is essentially a Floating Gate metal oxide semiconductor field effect transistor (Floating Gate MOSFET), and one or more oxide layers are broken down in each erasing/writing (P/E) process, which may cause a certain degree of physical damage to NAND Flash, so that the number of times of reading and writing is limited, which leads to a limited lifespan of SSD based on NAND Flash, which is also the largest defect.

The types of NAND Flash are divided into two types: Multi-Level cells (MLC) and Single-Level cells (SLC). Theoretically, the life of the MLC is 1 ten thousand times and the life of the SLC is 10 ten thousand times, but through practical tests, the life of the MLC is only 3 thousand times and the life of the SLC is only 3 ten thousand times. The defects of the NAND Flash need to be complemented by an SSD main control chip or an operating system algorithm, and a Wear-Leveling (WL) technology is produced at the same time.

The Wear-Leveling algorithm is divided into two types: dynamic (Dynamic) WL and Static (Static) WL. The Dynamic WL ensures that the writing/erasing circulation of data is uniformly distributed in all free blocks (blocks) of the NAND Flash, and avoids permanent damage of a storage unit caused by the fact that an application program repeatedly and continuously performs a P/E process on the same storage area. But using only the Dynamic WL does not guarantee that all blocks can use the WL algorithm with the same probability because the Dynamic WL does not take into account wear leveling when there is a cold data storage state. Wherein the cold data refers to data that is infrequent, infrequently accessed, or never accessed. Static WL takes cold data storage Block into account, and makes all blocks of Flash wear evenly by adjusting the physical address of cold data, so that SSD service life maximization is realized. When the SSD master control supports the Static WL algorithm and the SSD has cold data storage, the continuous write/erase process of the hot data introduces write amplification, and the write amplification may have a certain effect on the lifetime of the SSD. The hot data, i.e., data that is frequently active and frequently accessed.

At present, a commonly used SSD life test method is large-scale data writing, that is, a data writing process is continuously performed on an SSD which does not store any data by using some software or writing corresponding codes until the SSD is damaged. As shown in fig. 1, the specific writing process is as follows:

the method comprises the following steps: the data is written full in a completely new SSD.

Step two: and continuing to write data, and sequentially covering the Logical (Logical) Block 0, the Logical Block1, the Logical Block 2 …, the corresponding Physical (Physical) Block 0, the Physical Block1 and the Physical Block 2 …, wherein each resource Block is subjected to the erasing/writing process.

Step three: and repeating the second step until the SSD is damaged, and recording the writing times M, namely the service life of the SSD.

The process has no storage of cold data, does not cause write amplification, and is a wear-leveling process for NAND Flash, so that the test result is consistent no matter whether the SSD master control has the WL algorithm or not.

In summary, in the prior art, the method for large-scale data writing is slow, and data in all blocks needs to be erased and then written in each P/E process; the conventional method for testing the service life of the SSD by writing large-scale data cannot effectively test whether the SSD master control supports a static Wear-Leveling algorithm; the existing method for testing the service life of the SSD does not consider the state of cold data storage, neglects the influence of write amplification on the service life of the SSD in the actual use process, and cannot truly and effectively reflect the service life of the SSD in the actual use process.

Disclosure of Invention

The embodiment of the application provides a method and a device for testing the service life of a solid state disk, which are used for improving the testing efficiency and verifying whether the master control of the solid state disk supports a static Wear-Leveling algorithm or not, so that the testing result is more accurate.

The method for testing the service life of the solid state disk provided by the embodiment of the application comprises the following steps:

respectively writing the cold data and the hot data into a solid state disk to be tested, and storing the cold data into another device for backup;

continuously writing the hot data into the solid state disk, and recording the writing times of the hot data;

and comparing whether the cold data in the solid state disk is consistent with the backup cold data according to the writing times of the hot data and a preset period, if so, continuously writing the hot data into the solid state disk, and otherwise, finishing the test.

By the method, SSD life test is realized based on a mode of combining cold data storage state and hot data continuous writing, compared with the prior art that only hot data is used for writing/erasing, the test efficiency is improved, whether the solid state disk master control supports a static Wear-Leveling algorithm or not can be verified, the influence of writing amplification on the SSD life is fully considered, the life of the actual service condition of the SSD is truly reflected, and the test result is more accurate.

Preferably, the size of the hot data is smaller than or equal to the size of the storage space of one block in the solid state disk to be tested.

Preferably, the writing the cold data and the hot data into the solid state disk to be tested respectively specifically includes:

writing cold data into a block in the solid state disk to be tested, and reserving a blank block;

writing hot data into the reserved blank block.

Preferably, the continuously writing the thermal data into the solid state disk specifically includes:

and continuously carrying out erasing and writing operations on the solid state disk by utilizing the hot data.

Preferably, the comparing, according to the number of writing times of the hot data and according to a preset period, whether the cold data in the solid state disk is consistent with the backed-up cold data includes:

when the write-in times of the hot data are smaller than or equal to a preset threshold value, comparing whether the cold data in the solid state disk is consistent with the backup cold data or not according to a preset first period; and comparing whether the cold data in the solid state disk is consistent with the backup cold data according to a preset second period at the stage that the writing times of the hot data are larger than the threshold value, wherein the second period is larger than the first period.

Preferably, for a multi-layer unit flash memory type solid state disk, the threshold value is 5 ten thousand times, the first period is 1 ten thousand times, and the second period is 100 ten thousand times;

for a single-layer unit flash memory type solid state disk, the threshold value is 50 ten thousand times, the first period is 10 ten thousand times, and the second period is 100 ten thousand times.

The life-span testing arrangement of solid state hard drives that this application embodiment provided includes:

the first unit is used for respectively writing the cold data and the hot data into the solid state disk to be tested and storing the cold data into another device for backup;

the second unit is used for continuously writing the hot data into the solid state disk and recording the writing times of the hot data;

and the third unit is used for comparing whether the cold data in the solid state disk is consistent with the backup cold data according to the writing times of the hot data and a preset period, if so, continuously writing the hot data into the solid state disk, and otherwise, finishing the test.

Preferably, the writing of the cold data and the hot data into the solid state disk to be tested by the first unit includes:

writing hot data into the reserved blank block.

Preferably, the continuously writing the thermal data into the solid state disk by the second unit specifically includes:

Preferably, the third unit compares, according to the number of times of writing the hot data and according to a preset period, whether the cold data in the solid state disk is consistent with the backed-up cold data, and specifically includes:

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a diagram illustrating a prior art SSD being full of data for the first time;

fig. 2 is a schematic overall flow chart of a method for testing a lifetime of a solid state disk according to an embodiment of the present application;

fig. 3 is a schematic flowchart illustrating a specific flowchart of a method for testing a lifetime of a solid state disk according to an embodiment of the present application;

FIG. 4 is a schematic diagram illustrating a mapping relationship between a logical address and a physical address of a NAND Flash after cold data is stored in an embodiment of the present application;

FIG. 5 is a schematic diagram illustrating a mapping relationship between a NAND Flash logical address and a physical address after hot data is written for the first time according to an embodiment of the present disclosure;

FIG. 6 is a schematic diagram illustrating a mapping relationship between a NAND Flash logical address and a physical address after hot data is written in for the 2 nd time according to an embodiment of the present application;

FIG. 7 is a schematic diagram illustrating a mapping relationship between a NAND Flash logical address and a physical address after 3 rd write of hot data according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a device for testing a lifetime of a solid state disk according to an embodiment of the present application.

Detailed Description

According to the technical scheme provided by the embodiment of the application, in order to enable the SSD life test result to reflect the actual use condition more accurately and verify whether the SSD main control chip supports the Static Wear-leveling algorithm, the SSD life test method and the SSD life test device based on cold data storage and hot data continuous writing are provided.

The technical solutions provided by the embodiments of the present application are described below with reference to the accompanying drawings.

Referring to fig. 2, a method for testing a lifetime of a solid state disk provided in an embodiment of the present application includes:

s101, respectively writing the cold data and the hot data into the solid state disk to be tested, and storing the cold data into another device for backup.

writing hot data into the reserved blank block.

And S102, continuously writing the hot data into the solid state disk, and recording the writing times of the hot data.

The solid state disk is continuously subjected to erasing and writing operations by using the hot data, that is, the hot data in the SSD is overwritten by performing the overwriting operation of the hot data once per second according to a preset period.

S103, comparing whether cold data in the solid state disk is consistent with backup cold data according to a preset period according to the writing times of the hot data, if so, continuously writing the hot data into the solid state disk, and otherwise, ending the test.

Referring to fig. 3, a detailed flow of the technical solution provided in the embodiment of the present application is as follows.

The method comprises the following steps: and storing the cold data.

Firstly, determining the SSD capacity to be tested and the capacity of each Block, wherein the size of the Block is 128K, 256K or 512K generally; then, cold data with a certain size is stored in the SSD, and the cold data is backed up to another device for subsequent comparison and use. Note that a blank Block is reserved as a physical address to which hot data is continuously written. The size of the reserved space can be adjusted according to actual conditions.

It should be noted that, in the embodiment of the present application, one blank block may be reserved for hot data, and multiple blank blocks may also be reserved for hot data, and it is determined according to the actual use condition of the user how much cold data exists in the actual use process, the cold data with the corresponding size should be stored in the test, and the rest are reserved as the reserved blocks for storing the hot data.

Step two: and checking cold data.

In order to ensure that no error occurs during cold data writing, the second step may be performed, that is, whether the cold data written into the SSD is consistent with the backed-up cold data is checked, if so, the following third step is performed, otherwise, the cold data is written into the SSD again, and the cold data is backed-up again.

Step three: the hot data is written continuously.

Writing corresponding codes, continuously writing the tested SSD, wherein the process is realized through programming, connecting an SSD hard disk with a computer through a USB interface, and continuously writing and covering through a written program. The hot data continuous writing comprises the following steps: the hot data is used for covering the file with the same file name in the SSD, for example, the file A is the hot data, and the file A is continuously covered on the file A in the SSD. If file B is cold data, then file B in the SSD would not be overwritten. The write file size does not exceed one Block. Since the minimum erase unit of NAND Flash is one Block, writing once is equivalent to one Block undergoing a P/E process regardless of the size of the written file. It is recommended that the written file should not be too large, otherwise the writing speed is affected. The size of the written file can be adjusted according to actual use conditions.

Step four: and (5) comparing the staged cold data.

For the SSD based on the MLC NAND Flash type, 1 ten thousand times are taken as steps in the initial stage of testing (the writing times of hot data are less than 5 ten thousand), cold data in the SSD is compared with backup cold data, and whether data errors occur or not is checked. After the initial stage test is finished, the cold data in the SSD is compared with the backup cold data by taking 100 ten thousand times as stepping until a data error phenomenon occurs, the test is finished, and the writing frequency of the hot data is recorded, namely the service life of the SSD is obtained.

For the SSD based on the SLC NAND Flash type, the comparison process is similar to that described above, but 10 ten thousand steps are taken in the initial stage (the write-in times are less than 50 ten thousand) and 100 ten thousand steps are taken in the later stage until the data has an error phenomenon, the test is finished, and the write-in times of the hot data are recorded, namely the service life of the SSD.

The test principle is as follows:

first, after storing the cold data, the NAND Flash only has a space of one Block, i.e. a blank Block (Free Block) as shown in fig. 4.

Secondly, hot data is written for the first time, at this time, Free Block is occupied, and it is assumed that the Physical address is Physical Block N and the corresponding Logical address is Logical Block N, as shown in fig. 5.

Thirdly, when the hot data is continuously written in, for the SSD which does not support the WL algorithm or only supports the Dynamic near-level algorithm by the main control, the Physical Block N in the NAND Flash is continuously erased/written until the Block N is damaged. The process is relatively short, and depends on the type of NAND Flash, and theoretically, the life of MLC is 1 ten thousand times, and the life of SLC is 10 ten thousand times.

However, for the SSD whose main control supports Static wear-leveling algorithm, the correspondence between the Logical address and the physical address of the NAND Flash is changed, and the physical addresses occupied by the cold data are alternately released, corresponding to the Logical Block N, as shown in fig. 6 and 7, respectively. Therefore, the loss of each Block is balanced, the resources of the NAND Flash are fully utilized, and the service life of the SSD is effectively prolonged.

Fourthly, write amplification is caused when the processes shown in fig. 6 and 7 are executed. Since the data in the released resource Block (e.g. Physical Block 0) needs to be rewritten into the exchanged resource Block (Block N), although only one Block of data is written, the actual SSD needs to complete the writing of two Block data.

For the SSD without the WL algorithm, the resource blocks participating in abrasion are only the number of the resource blocks occupied by the written files; for the SSD only with the Dynamic WL algorithm, the resource blocks participating in abrasion are the number of the reserved space resource blocks; for SSDs with the StaticWL algorithm, all resource blocks are involved in wear. Typically, an SSD has hundreds or thousands of resource blocks, so that SSDs with Static WL differ by orders of magnitude from SSDs without the Static WL algorithm and are easily distinguished. For example, the following steps are carried out:

assuming that a SLC type SSD (lifetime 10 ten thousand) has 2048 blocks of which 2047 are occupied, one Block is reserved for data writing:

in the case of no Static WL algorithm, only one reserved Block participates in abrasion, and the writing time is 10⁵Secondly;

with Static WL algorithm, all blocks are worn, and two blocks are worn out every time writing, with 2048 × 105/2 ≈ 10⁸Next, the process is carried out.

Therefore, it is easy to distinguish whether the SSD has Static WL.

Correspondingly to the above method, an embodiment of the present application provides a device for testing a lifetime of a solid state disk, with reference to fig. 8, including:

the first unit 11 is used for respectively writing cold data and hot data into the solid state disk to be tested, and storing the cold data into another device for backup;

the second unit 12 is configured to continuously write the thermal data into the solid state disk, and record the number of times of writing the thermal data;

and a third unit 13, configured to compare, according to the number of times of writing the hot data, whether the cold data in the solid state disk is consistent with the backup cold data according to a preset period, if so, continue to write the hot data into the solid state disk, and otherwise, end the test.

writing hot data into the reserved blank block.

To sum up, in the embodiment of the present application, the SSD life test is performed based on a method combining a cold data storage state and a hot data continuous write; the size of the cold data file and the size of the reserved space can be adjusted according to the actual use condition, and the influence of write amplification on the service life of the SSD is fully considered, so that the test result is more accurate; the test method for large-scale data writing is faster, whether the SSD master control supports the static weather-leveling algorithm or not can be verified, the condition of writing amplification in the actual use process is fully considered, and the service life of the actual use condition is truly reflected.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such 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 intended to include such modifications and variations as well.

Claims

1. A method for testing the service life of a solid state disk is characterized by comprising the following steps:

2. The method of claim 1, wherein the size of the thermal data is smaller than or equal to the size of a storage space of one block in the solid state disk to be tested.

3. The method according to claim 2, wherein the writing the cold data and the hot data into the solid state disk to be tested respectively comprises:

writing hot data into the reserved blank block.

4. The method according to claim 3, wherein continuously writing the thermal data to the solid state disk comprises:

5. The method according to claim 1, wherein the comparing, according to the number of times of writing the hot data and according to a preset period, whether the cold data in the solid state disk is consistent with the backed-up cold data includes:

6. The method of claim 5, wherein the threshold value is 5 ten thousand times, the first period is 1 ten thousand times, and the second period is 100 ten thousand times for a multi-level cell flash type solid state disk;

7. The utility model provides a life-span testing arrangement of solid state hard drives which characterized in that includes:

8. The apparatus of claim 7, wherein the size of the thermal data is smaller than or equal to the size of a storage space of one block in the solid state disk to be tested.

9. The apparatus of claim 8, wherein the first unit writes the cold data and the hot data into the solid state disk to be tested, respectively, and specifically includes:

writing hot data into the reserved blank block.

10. The apparatus of claim 9, wherein the second unit continuously writes the thermal data to the solid state disk, and specifically comprises:

11. The apparatus according to claim 7, wherein the third unit compares, according to the number of times of writing the hot data and according to a preset period, whether the cold data in the solid state disk is consistent with the backed-up cold data, specifically includes:

12. The apparatus of claim 11, wherein the threshold value is 5 ten thousand times, the first period is 1 ten thousand times, and the second period is 100 ten thousand times for a multi-level cell flash type solid state disk;