WO2023206968A1

WO2023206968A1 - Data storage method and system, and computer readable storage medium

Info

Publication number: WO2023206968A1
Application number: PCT/CN2022/122872
Authority: WO
Inventors: 张子奇
Original assignee: 浪潮电子信息产业股份有限公司
Priority date: 2022-04-29
Filing date: 2022-09-29
Publication date: 2023-11-02
Also published as: CN114756178A

Abstract

A data storage method and system, and a computer readable storage medium. The method comprises: a solid-state storage device receiving a data writing request sent by a data writing apparatus, and extracting from the data writing request data to be written and a logic address; searching for an original physical address corresponding to the logic address, adding to the original physical address a "to be recovered" label, and setting a default retention time for the original physical address; writing into a new physical address the data to be written, and mapping the new physical address to the logic address; and receiving a data recovery request for the logic address sent by the data writing apparatus, and, in response to determining that the default retention time has not ended, returning to the data writing apparatus original data corresponding to the original physical address.

Description

A data storage method, system and computer-readable storage medium

Cross-references to related applications

This application requires the priority of the Chinese patent application submitted to the China Patent Office on April 29, 2022, with the application number 202210467206.X, and the application name is "A data storage method, system and computer-readable storage medium", all of which The contents are incorporated into this application by reference.

Technical field

The present application relates to a data storage method, system and computer-readable storage medium.

Background technique

Solid-state storage devices (SSD, Solid-state memory) have fast read and write speeds, so they are often installed in storage systems to improve the read and write performance of storage systems. However, the inventor realized that traditional solid-state disks can only protect the latest data when protecting the reliability of data, and directly give up protection for old data in overwritten logical addresses. This means that if the storage system needs to protect old data, it needs to add complex redundant protection mechanisms to the upper-layer software, which is not only detrimental to system maintenance but also affects system performance.

Contents of the invention

According to various embodiments disclosed in this application, a data storage method, system and computer-readable storage medium in this application are provided.

A data storage method that includes:

The solid-state storage device receives the data writing request sent by the data writing device, and extracts the data to be written and the logical address from the data writing request;

Find the original physical address corresponding to the logical address, add a pending recycling mark to the original physical address, and set a preset retention time for the original physical address;

Write the data to be written to the new physical address and map the new physical address to the logical address; and

Receive the data recovery request for the logical address sent by the data writing device, and return the original data corresponding to the original physical address to the data writing device in response to determining that the preset retention time has not expired.

A data storage system, including: a solid-state storage device and a data writing device, wherein,

The solid-state storage device is used to receive the data writing request sent by the data writing device, and extract the data to be written and the logical address from the data writing request; to find the original physical address corresponding to the logical address, and to add the data to be written to the original physical address. Recycle the mark and set the preset retention time for the original physical address; write the data to be written into the new physical address, and map the new physical address to the logical address; receive the data recovery request for the logical address sent by the data writing device, and respond When the preset retention time has not expired, the original data corresponding to the original physical address is returned to the data writing device;

A data writing device is used to send data writing requests and data recovery requests.

A computer device, including a memory and one or more processors. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the processor, they cause the one or more processors to execute any one of the above. Data storage method steps.

A non-volatile computer-readable storage medium. Computer-readable instructions are stored in the non-volatile computer-readable storage medium. When the computer-readable instructions are loaded and executed by a processor, the data described in any of the above items is realized. Store the steps of the method.

Description of the drawings

In order to explain the embodiments of the present application or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only This is an embodiment of the present application. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without exerting creative efforts.

Figure 1 is a flow chart of a data storage method provided by one or more embodiments of the present application;

Figure 2 is a schematic diagram of a traditional data consistency protection method provided by one or more embodiments of the present application;

Figure 3 is a schematic diagram of the data rollback process provided by one or more embodiments of the present application;

Figure 4 is a structural block diagram of a data storage system provided by one or more embodiments of the present application;

Figure 5 is a structural block diagram of a computer device provided by one or more embodiments of the present application;

Figure 6 is a structural block diagram of a non-volatile computer-readable storage medium provided by one or more embodiments of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments 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 These are part of the embodiments of this application, but not all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

When protecting the reliability of data, traditional solid-state disks can only protect the latest data, and directly give up protection for overwritten old data. This means that if the storage system needs to protect old data, it needs to add complex redundant protection mechanisms to the upper-layer software, which is not only detrimental to system maintenance but also affects system performance. In view of this, the present application provides a data storage method, in which a solid-state storage device can temporarily store old data that needs to be overwritten, thereby achieving protection of the old data. Please refer to Figure 1, which is a flow chart of a data storage method provided by one or more embodiments of the present application. The method may include:

S101. The solid-state storage device receives the data writing request sent by the data writing device, and extracts the data to be written and the logical address from the data writing request.

It should be noted that in this embodiment of the present application, the data writing device is upper-layer software that manages and controls the solid-state storage device. Generally, the data writing request sent by the device usually includes the data to be written and the corresponding logical address of the data in the solid-state storage device. In other words, the meaning of the data write request is to overwrite the original data originally stored in the logical address with the data to be written. It should be noted that the logical address refers to the address of the memory unit, storage unit or network host seen from the perspective of the application program in the computer architecture, which is often different from the physical address. The relationship between logical address and physical address is: the logical address can be converted into a physical address through an address translator or mapping function.

S102. Find the original physical address corresponding to the logical address, add a to-be-recycled mark to the original physical address, and set a preset retention time for the original physical address.

The original physical address refers to the physical address that was originally mapped to the above-mentioned logical address, in which the above-mentioned original data is stored. In traditional technology, after the data to be written overwrites the logical address, the original data in the original physical address becomes invalid immediately, and the address is automatically added to the garbage collection space of the solid-state storage device. In this embodiment of the present application, the original physical address and the data therein will not become invalid immediately, but will be marked with a to-be-recycled mark and set with a preset retention time. The mark indicates that the address and the data therein will be recycled. Temporary storage, and the preset retention time represents the temporary storage time, which can be implemented using a countdown. In other words, in this embodiment of the present application, the original data corresponding to the overwritten logical address will not be deleted immediately, but will continue to be saved for a period of time. In this way, when a writing error occurs in new data or an abnormality occurs in the storage system, the upper-layer software can directly export the original data from the solid-state storage device and perform data rollback. Obviously, compared with traditional solid-state storage devices, the solid-state storage devices in the embodiments of the present application can protect old data, thereby avoiding the storage system from adding complex redundant protection mechanisms in upper-layer software, and can improve storage The processing performance of the system and the convenience of system maintenance.

It should be noted that the embodiment of the present application does not limit the specific form of the mark to be recycled, as long as the mark can represent the original data temporarily stored in the marked physical address, and the garbage collection mechanism of the solid-state storage device reads the mark. , automatically skip the marked physical address. The embodiments of the present application do not limit the specific value of the preset retention time, which can be set according to actual application requirements. Furthermore, it can be understood that the physical address marked with the to-be-reclaimed mark can only be read and cannot be modified (such as written and deleted), and the to-be-reclaimed space composed of this part of the physical address reduces the available storage in the solid-state storage device. space, the larger the space to be reclaimed, the more serious the shrinkage impact will be. In other words, there should be an upper limit on the space to be reclaimed. In this embodiment of the present application, the upper limit of the space to be reclaimed can be set according to the maximum bandwidth of the solid-state storage device. For example, it can be set to a preset multiple of the maximum bandwidth. For example, if the bandwidth is 3GB/s, then the size of the space to be reclaimed can be 2GB×3=6GB to ensure that when the solid-state storage device writes data at the maximum bandwidth, the original data overwritten within 3 seconds will not be lost. That is, solid-state storage devices can ensure that old data that needs to be overwritten is protected as much as possible under the worst data writing conditions.

In one or more embodiments, all physical addresses added with a to-be-reclaimed mark constitute a space to be reclaimed, and the size of the space to be reclaimed does not exceed a preset multiple of the maximum bandwidth of the solid-state storage device.

It should be noted that the embodiments of the present application do not limit the specific value of the preset multiple, which can be set according to actual application requirements.

S103. Write the data to be written into a new physical address, and map the new physical address to a logical address.

In this embodiment of the present application, the data to be written will be written to a new physical address, and the new physical address will also be mapped to a logical address. In this way, when the data at the logical address is read normally, new data can be read in the new physical address according to the new mapping relationship corresponding to the logical address.

S104. Receive the data recovery request for the logical address sent by the data writing device, and return the original data corresponding to the original physical address to the data writing device in response to the fact that the preset retention time has not expired.

Since the embodiment of the present application temporarily stores the data in the original physical address, the solid-state storage device can provide external data recovery functions, that is, the data writing device can send the data to the solid-state storage device for a recently overwritten logical address. Data recovery request, and when the latter determines that the preset retention time of the original physical address corresponding to the logical address has not expired, the corresponding original data can be sent back to the data writing device for the data writing device to perform data rollback. recover. It can be understood that the solid-state storage device can provide a data recovery interface, and the data writing device only needs to call this interface to achieve the corresponding effect of step S104.

Of course, it is worth pointing out that if the solid-state storage device encounters a power outage or restarts, the data stored in the physical address marked with the recovery flag will still not be deleted. When the solid-state storage device returns to normal working status again, if these If the preset retention time corresponding to the data has not been exhausted, saving will continue.

Furthermore, when the preset retention time corresponding to the original physical address expires, the original physical address can be added to the garbage collection space of the solid-state storage device and restored to a writable state.

In one or more embodiments, after adding a to-be-recycled mark to the original physical address and setting a preset retention time for the original physical address, it may also include:

Step 11: In response to the preset retention time having expired, add the original physical address to the garbage collection space.

Of course, after completing the original physical address setting and new data writing, the solid-state storage device can also send a writing success message to the data writing device to inform the latter that the data writing is completed.

In one or more embodiments, after mapping the new physical address to the logical address, it may also include:

Step 21: Send writing success information to the data writing device.

Based on the above embodiments, when the solid-state storage device of the present application receives a data writing request issued by the data writing device, it will first extract the logical address from the request, and search the device for the original physical address corresponding to the address. ;Further, the solid-state storage device will set a pending recovery mark for the original physical address and set a preset retention time. The pending recovery mark indicates that the original data in the original physical address can be temporarily retained without deletion, and the preset retention time indicates that the original The length of time that the data can be temporarily retained; while the new data in the data write request is written to the new physical address, and the new physical address is corresponding to the above logical address. In other words, this application can save the new data that needs to cover the logical address in the new physical address in the solid-state storage device, and at the same time, continue to store the original data in the original physical address corresponding to the logical address before the preset retention time expires. to save. In this way, when the data writing device needs to recover the original data, it only needs to send a data recovery request for the logical address to the solid-state storage device before the preset retention time expires to control the latter to re-extract the data in the original physical address. It can realize the protection of old data by solid-state storage devices and avoid the storage system adding complex redundant protection mechanisms in the upper-layer software, thereby improving the processing performance of the storage system and the convenience of system maintenance.

Based on the above embodiments, one or more implementations of introducing the solid-state storage device provided by this application into a storage system will be introduced in detail below. In one or more embodiments, before the solid-state storage device receives the data writing request sent by the data writing device, it may also include:

S201. The data writing device receives the file coverage request, and reads the new file and the logical address to be overwritten corresponding to the new file from the file coverage request.

First of all, it should be noted that in this embodiment of the present application, multiple solid-state storage devices will be used simultaneously for distributed data storage. In other words, the data writing device will interface with multiple solid-state storage devices. It should be noted that the embodiments of the present application do not limit the number of solid-state storage devices in the storage system, which can be set according to actual application requirements. It should also be pointed out that the logical address to be overwritten in this step is a logical address set based on considering these solid-state storage devices as a whole (i.e., system logical address), which is different from the internal logical address of each solid-state storage device. .

S202. Query in the database the old file corresponding to the logical address to be overwritten and the logical addresses corresponding to the old files in multiple solid-state storage devices.

In this embodiment of the present application, the mapping relationship between each file in the distributed storage system and the system logical address is stored in the database. Therefore, you can use the logical address to be overwritten in the file overwrite request to query the database for the old file that establishes a mapping relationship with the address. In addition, it can be understood that a single file is usually divided into several data blocks and stored in different solid-state storage devices. Each data block also has a corresponding logical address in the corresponding solid-state storage device. Therefore, the mapping relationship between the system logical address and the logical address of each storage device is also recorded in the database. It should be noted that the embodiments of the present application do not limit the mapping relationship between each file and the system logical address, as well as the storage form of the mapping relationship between the system logical address and the logical address of each solid-state storage device in the database, and querying the above-mentioned For the specific method of mapping relationship, please refer to the related technologies of distributed storage systems.

S203. Perform block processing on the new file to obtain data blocks, generate a data write request using the data blocks and corresponding logical addresses, and send the data write request to the corresponding solid-state storage device.

It should be noted that the embodiments of the present application do not limit the specific method of block processing, which can be set according to actual application requirements. For example, new files can be simply divided into blocks, or new files can be processed using erasure coding mechanisms and other technologies. Blocking is performed. The erasure coding mechanism is a redundant protection mechanism that ensures that when part of a data block is lost, the remaining data blocks can be used to restore the original file. Further, after the data is divided into blocks, the data writing device can use the obtained data blocks and the previously queried data writing requests to generate data writing requests for each solid-state storage device and issue the requests.

Correspondingly, it can also include:

S204. The data writing device determines whether the solid-state storage device returns writing success information within a preset time. If yes, go to step S205; if not, go to step S206.

It can be understood that solid-state storage devices can usually complete data writing within a predictable time without encountering abnormalities. Therefore, in this embodiment of the present application, a data writing time, that is, a preset time, will be set for each solid-state storage device. When the solid-state storage device can return writing success information within a preset time, it is determined that the device does not have a writing error; otherwise, the device is considered to have a writing error. On this basis, if the data writing device determines that all solid-state storage devices can return writing success information within a preset time, it can be determined that the new file has been written correctly, and then the relevant information in the database can be processed. Update; conversely, if the data writing device determines that some solid-state storage devices do not return writing success information within a preset time, it can be determined that the new file encounters a writing problem. In the latter case, since it can be determined that the solid-state storage device that has returned the writing success information has completed the data writing, that is, data recovery is required. At this time, the data writing device can call the solid-state storage device that has returned the writing success information. Data recovery interface to obtain original data and perform data rollback.

Of course, since writing a new file involves writing sub-operations with multiple solid-state storage devices, in order to ensure data consistency, the data writing device can create a writing transaction and set the corresponding transaction before writing the data. Expiration time, in which all write sub-operations in the same transaction either succeed or fail. When some write sub-operations encounter write failures, the remaining successful sub-operations will also be judged to have failed when the transaction expiration time expires.

In one or more embodiments, before performing block processing on the new file to obtain the data blocks, it may also include:

Step 31: The data writing device creates a writing transaction and sets the transaction expiration time; the transaction expiration time is greater than the preset time.

S205. In the database, use the new mapping relationship between the new file and the logical address to be overwritten to overwrite the old mapping relationship between the old file and the logical address to be overwritten.

It should be noted that since traditional solid-state storage devices cannot protect overwritten old data, existing data storage systems need to set up additional redundancy protection mechanisms to protect old data through upper-layer software. Specifically, please refer to Figure 2, which is a schematic diagram of the traditional data consistency protection method, in which disk0~disk5 represent 6 solid-state storage devices, 0, 1, 2, 3, P, and Q represent data blocks, DB ( Database) means database. The workflow of traditional data consistency protection is as follows:

1) When data is not written, the stripe data is marked in light gray. It is assumed that the system logical address corresponding to these data is A;

2) When updating data, the updated data is marked in dark gray. These data will not be written directly to system logical address A, but will be written to system logical address B;

3) Record the mapping relationship between the new address B and the data in the database;

4) Delete the mapping relationship between address A and data recorded in the database;

5) Delete the data corresponding to address A;

6) The data is updated.

It can be seen that when completing data writing, the data writing device needs to perform an additional data writing and deletion operation in the database, and this is repeated with the internal operation mode of the solid-state storage device. When large-scale data writing is performed, additional deletion operations are required. Operation will cause additional system performance loss. This is because when traditional solid-state storage devices protect the reliability of data through space redundancy and various algorithms, they can only protect the latest data and directly give up the protection of overwritten old data. Therefore, they can only use append writing through upper-layer software. way to prevent data loss. In this application, since the solid-state storage device can continue to protect the overwritten old data for a certain period of time, the data writing device can use the new mapping relationship between the new file and the logical address to be overwritten to protect the old file and the same system logic. The old mapping relationship between addresses is directly overwritten, which can effectively avoid additional deletion operations, thereby effectively improving the write performance of the storage system.

S206. Send a data recovery request to the solid-state storage device that has sent the write success information and receive the original data, and rewrite the original data to the corresponding logical address in the corresponding solid-state storage device.

When some solid-state storage devices have writing exceptions, the data writing device will perform data rollback on the solid-state storage devices that have completed data writing. Specifically, the device will send a data recovery request to the solid-state storage device that has sent write success information, receive the original data returned by these devices, and then rewrite the original data back to the original location in the solid-state storage device.

Furthermore, if the data writing device is set with a transaction, the data rollback recovery should wait until the transaction expiration time expires to avoid data writing chaos.

In one or more embodiments, rewriting the original data to the corresponding logical address in the corresponding solid-state storage device may include:

Step 41: In response to the transaction expiration time having expired, rewrite the original data to the corresponding logical address in the corresponding solid-state storage device.

Of course, it can be understood that if all solid-state storage devices with write exceptions can return write success information before the transaction expiration time expires, the data rollback operation may not be performed.

The following is an introduction to the implementation of introducing an erasure coding mechanism. In one or more embodiments, performing block processing on a new file to obtain data blocks may include:

Step 51: Use the erasure coding mechanism to perform block processing on the new file to obtain data blocks; when the number of lost data blocks does not exceed the preset value, the remaining data blocks can be used to restore the new file.

It should be noted that the default value here refers to the protection level of the erasure coding mechanism. In other words, as long as the number of lost data blocks of the new file does not exceed the erasure coding protection level, the mechanism can use the remaining data blocks to restore the new file. It should be noted that the embodiments of the present application do not limit specific preset values, which can be set according to actual application requirements. The embodiments of this application do not limit the specific erasure coding mechanism. For example, it can be a 4+2 erasure code, an 8+3 erasure code, a 6+2 erasure code, etc., which can be set according to actual application requirements. .

Correspondingly, before sending a data recovery request to the solid-state storage device that has sent write success information and receiving old data, it also includes:

Step 61: The data writing device determines whether the number of solid-state storage devices that have not sent writing success information is greater than the preset value; if so, proceed to step 62; if not, proceed to step S205;

Since erasure coding can redundantly protect new files, as long as the number of abnormal solid-state storage devices does not exceed the protection level of erasure coding, there is no need to perform data rollback. Otherwise, data rollback is required.

Step 62: Execute the steps of sending a data recovery request to the solid-state storage device that has sent the write success information and receiving old data.

For ease of understanding, the data rollback process provided by the embodiment of the present application will be introduced below in conjunction with a specific flow chart. Please refer to Figure 3, which is a schematic diagram of a data rollback process provided by one or more embodiments of the present application. When data update fails, all data blocks need to be restored to their original state. The process is as follows:

1) When data is not written, the stripe data is marked in light gray, and the corresponding system logical address is A;

2) When writing data, the updated data is marked in dark gray and directly overwrites the system logical address A;

3) Disk2, 3, and 4 were successfully written, but disk0, 1, and 5 suffered a power outage and the writing was unsuccessful. At this time,

disks

2, 3, and 4 mark the original data as "to be recycled" and start the 3-second timer;

4) After 1.5 seconds (can be set and adjusted), the upper-layer software found that disk0, 1, and 5 took much longer than the normal time (generally SSD writing is within 0.1 milliseconds), and at this time the 4+2 erasure code cannot recover the data. , so the old data recovery interface provided by the solid-state disk is called to read all the old data in the strip including disk2, 3, and 4 and save it.

6) After the 3-second timer expires, Disk2, 3, and 4 change the data block from the "to be recycled" state to the "deleted" state, and move the old data into the garbage collection space;

7) After the upper-layer software waits for the transaction mechanism to time out (if disk0, 1, and 5 reply successfully within the timeout period, the transaction is successful and no follow-up action is required), the old data is written to disk2, 3, and 4 address A, and the transaction rollback is successful. All data within the stripe is restored to light gray old data.

Based on the above embodiments, the embodiments of the present application can use the improved solid-state storage device to change the data deletion action from the active deletion of the upper-layer software to the internal aging of the solid-state disk, thereby reducing one deletion action and thus significantly reducing the recording time. The amount of data reduces the pressure on the distributed storage system and improves system performance.

The data storage system and computer-readable storage medium provided by the embodiments of the present application are introduced below. The data storage system and computer-readable storage medium described below and the data storage method described above can be mutually referenced.

Please refer to Figure 4. Figure 4 is a structural block diagram of a data storage system provided by one or more embodiments of the present application. The system may include: a solid-state storage device 401 and a data writing device 402, wherein,

The solid-state storage device 401 is used to receive the data writing request sent by the data writing device 402, and extract the data to be written and the logical address from the data writing request; find the original physical address corresponding to the logical address, and obtain the original physical address Add a mark to be recycled and set a preset retention time for the original physical address; write the data to be written into the new physical address, and map the new physical address to the logical address; receive data recovery for the logical address sent by the data writing device 402 Request, in response to the fact that the preset retention time has not expired, return the original data corresponding to the original physical address to the data writing device 402;

The data writing device 402 is used to send data writing requests and data recovery requests.

In one or more embodiments, solid-state storage device 401 can also be used for:

In response to the preset retention time having expired, the original physical address is added to the garbage collection space.

In one or more embodiments, all physical addresses added with a to-be-reclaimed mark constitute a space to be reclaimed, and the size of the space to be reclaimed does not exceed a preset multiple of the maximum bandwidth of the solid-state storage device 401 .

Send writing success information to the data writing device 402.

In one or more embodiments, the data writing device 402 can also be used for:

Receive the file overwrite request, read the new file and the corresponding logical address to be overwritten from the file overwrite request; query the database for the old file corresponding to the logical address to be overwritten and the corresponding logical addresses of the old files in multiple solid-state storage devices 401; Perform block processing on the new file to obtain data blocks, use the data blocks and logical addresses to generate data write requests, and send the data write requests to the corresponding solid-state storage device 401;

Correspondingly, the data writing device 402 can also be used for:

Determine whether the storage device returns write success information within the preset time; if so, use the new mapping relationship between the new file and the logical address to be overwritten to overwrite the old mapping relationship between the old file and the logical address to be overwritten in the database ; If not, send a data recovery request and receive the original data to the solid-state storage device 401 that has sent the write success information, and rewrite the original data to the corresponding logical address in the corresponding solid-state storage device 401.

In one or more embodiments, the data writing device 402 can also be used for:

Use the erasure coding mechanism to process the new file into blocks to obtain data blocks; when the number of lost data blocks does not exceed the preset value, the remaining data blocks can be used to restore the new file;

Correspondingly, before sending a data recovery request and receiving old data to the solid-state storage device 401 that has sent the writing success information, the data writing device 402 can also be used to:

Determine whether the number of solid-state storage devices 401 that have not sent write success information is greater than a preset value; if so, perform the steps of sending a data recovery request and receiving old data to the solid-state storage devices 401 that have sent write success information.

In one or more embodiments, the data writing device 402 can also be used for:

Create a write transaction and set the transaction expiration time; the transaction expiration time is greater than the preset time;

Correspondingly, the data writing device 402 can also be used for:

In response to the transaction expiration time having expired, the original data is rewritten to the corresponding logical address in the corresponding solid-state storage device 401 .

In one or more embodiments, a computer device is also provided, as shown in FIG. 5 , which is a schematic structural diagram of the computer device in one or more embodiments. The computer device includes a memory 50 and one or more processors 51. Computer-readable instructions are stored in the memory 50. When the computer-readable instructions are executed by the one or more processors 51, the one or more processors 51 execute any of the above. Steps of the data storage method of one or more embodiments.

In one or more embodiments, as shown in FIG. 6 , a non-volatile computer-readable storage medium 60 is also provided. Computer-readable instructions 61 are stored on the non-volatile computer-readable storage medium 60 . When the readable instructions 61 are executed by the processor, the steps of the data storage method of any one or more of the above embodiments are implemented.

Since the embodiments of the computer-readable storage medium part correspond to the embodiments of the data storage method part, for the embodiments of the storage medium part, please refer to the description of the embodiments of the data storage method part and will not be described again here.

Each embodiment in the specification is described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For relevant details, please refer to the description in the method section.

Those skilled in the art may further realize that the units and algorithm steps of each example described in connection with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. In order to clearly illustrate the possible functions of hardware and software, Interchangeability, in the above description, the composition and steps of each example have been generally described according to functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each specific application, but such implementations should not be considered beyond the scope of this application.

The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly in hardware, in software modules executed by a processor, or in a combination of both. Software modules may be located in random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or anywhere in the field of technology. any other known form of storage media.

The above describes in detail a data storage method, system and computer-readable storage medium provided by this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application. It should be noted that for those of ordinary skill in the art, several improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

A data storage method, characterized by including:

The solid-state storage device receives the data writing request sent by the data writing device, and extracts the data to be written and the logical address from the data writing request;

Find the original physical address corresponding to the logical address, add a to-be-recycled mark to the original physical address, and set a preset retention time for the original physical address;

Write the data to be written to a new physical address, and map the new physical address to the logical address; and

Receive the data recovery request for the logical address sent by the data writing device, and return the original data corresponding to the original physical address to the data writing device in response to the fact that the preset retention time has not expired.
The method of claim 1, further comprising:

In response to the preset retention time having expired, the original physical address is added to the garbage collection space.
The method of claim 1, further comprising:

The space to be reclaimed is formed according to the physical address added with the to-be-reclaimed mark.
The method of claim 1, wherein the size of the space to be reclaimed does not exceed a preset multiple of the maximum bandwidth of the solid-state storage device.
The method according to any one of claims 1 to 4, characterized in that after mapping the new physical address to the logical address, it further includes:

Send writing success information to the data writing device.
The method according to claim 5, characterized in that before the solid-state storage device receives the data writing request sent by the data writing device, it further includes:

The data writing device receives a file coverage request and reads a new file and the logical address to be overwritten corresponding to the new file from the file coverage request;

Query in the database the old file corresponding to the logical address to be overwritten and the logical addresses corresponding to the old file in multiple solid-state storage devices; and

The new file is divided into blocks to obtain data blocks, and the data write request is generated using the data blocks and the logical address.
The method of claim 6, further comprising:

Send the generated data write request to the corresponding solid-state storage device.
The method of claim 7, further comprising:

The data writing device determines whether the corresponding solid-state storage device returns the writing success information within a preset time; and

In response to the corresponding solid-state storage device returning the writing success information within a preset time, in the database, the new mapping relationship between the new file and the logical address to be overwritten is used to overwrite the old file. The old mapping relationship between the file and the logical address to be overwritten.
The method of claim 8, further comprising:

In response to the corresponding solid-state storage device not returning the write success information within a preset time, sending the data recovery request to the solid-state storage device that has sent the write success information and receiving the original data , and rewrite the original data to the corresponding logical address in the corresponding solid-state storage device.
The method according to claim 6, characterized in that said performing block processing on the new file to obtain data blocks includes:

The new file is divided into blocks using an erasure coding mechanism to obtain the data blocks.
The method according to claim 10, characterized in that the number of lost data blocks does not exceed a preset value, and the remaining data blocks that are not lost are used to restore the new file.
The method according to claim 11, characterized in that, before sending the data recovery request to the solid-state storage device that has sent the writing success information and receiving the old data, it further includes:

The data writing device determines whether the number of solid-state storage devices that have not sent the writing success information is greater than the preset value; and

In response to the number of the solid-state storage devices that have not sent the write success information being greater than the preset value, sending the data recovery request to the solid-state storage devices that have sent the write success information and receiving all Describe the steps for old data.
The method of claim 12, further comprising:

In response to the number of solid-state storage devices that have not sent the write success information being no greater than the preset value, executing in the database overwriting using a new mapping relationship between the new file and the logical address to be overwritten. The step of establishing an old mapping relationship between the old file and the logical address to be overwritten.
The method according to claim 11, characterized in that before performing block processing on the new file to obtain data blocks, it further includes:

The data writing device creates a writing transaction and sets the transaction expiration time.
The method according to claim 14, characterized in that the transaction expiration time is greater than the preset time.
The method of claim 14, wherein rewriting the original data to the corresponding logical address in the corresponding solid-state storage device includes:

In response to the transaction expiration time having expired, the original data is rewritten to the corresponding logical address in the corresponding solid-state storage device.
A data storage system, characterized by including: a solid-state storage device and a data writing device, wherein,

The solid-state storage device is configured to receive a data writing request sent by the data writing device, and extract the data to be written and a logical address from the data writing request; and search for the original physical address corresponding to the logical address. , and add a to-be-recycled mark to the original physical address and set a preset retention time for the original physical address; write the data to be written into a new physical address, and map the new physical address to the logical Address; receive a data recovery request for the logical address sent by the data writing device, and return the original data corresponding to the original physical address to the data writing device in response to the fact that the preset retention time has not expired;

The data writing device is used to send the data writing request and the data recovery request.
The system of claim 17, wherein the solid-state storage device is further configured to add the original physical address to the garbage collection space in response to the expiration of the preset retention time.
A computer device includes a memory and one or more processors. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the one or more processors, the one or more processors cause the one or more processors to A processor executes the data storage method according to any one of claims 1 to 16.
A non-volatile computer-readable storage medium, characterized in that computer-readable instructions are stored in the computer-readable storage medium, and when the computer-readable instructions are loaded and executed by one or more processors, The data storage method according to any one of claims 1 to 16.