CN114840364A - Method and device for backing up storage data in memory and electronic equipment - Google Patents

Abstract

The present disclosure provides a method, an apparatus and an electronic device for backing up stored data in a memory, which relate to a data storage technology, wherein a backup space is preset in the memory; when data are read from a storage unit of a memory, if the risk that uncorrectable errors occur in the storage unit is determined, writing the data in the storage unit into a backup space, and determining that a memory address where the storage unit is located corresponds to a backup address of the data in the backup space; and if uncorrectable errors occur when the data are read from the storage unit again, reading the data from the backup space according to the backup address corresponding to the memory address of the storage unit. According to the scheme provided by the disclosure, the data of the storage unit is backed up by using the space of the memory, and the size of the storage unit can be set according to requirements, so that the electronic equipment can back up the data of the storage unit which is dispersed in the memory and possibly has the UE error, and the problem of system downtime caused by the UE error of the memory is solved.

Description

Method, device and electronic device for backing up stored data in memory

technical field

The present disclosure relates to data storage technologies, and in particular, to a method, an apparatus and an electronic device for backing up stored data in a memory.

Background technique

At present, many business functions are provided by the server, and the stability of the server has a significant impact on the quality of the business.

If there is an error in the memory of the server, the server is prone to downtime. Especially when a UE (Uncorrectable Errors, uncorrectable error) type error occurs in the memory, the server is likely to be automatically restarted, resulting in business interruption.

Therefore, how to still process services normally when a UE-type error occurs in the memory of the server is a technical problem that needs to be solved.

SUMMARY OF THE INVENTION

The present disclosure provides a method, an apparatus and an electronic device for backing up stored data in a memory, so as to solve the problem in the prior art that a server cannot normally process services due to UE-type errors.

A first aspect of the present disclosure is to provide a method for backing up stored data in a memory, and a backup space is preset in the memory;

The method includes: when reading data from a storage unit of the memory, if it is determined that the storage unit has a risk of uncorrectable errors, writing the data in the storage unit into the backup space, and determining the storage unit. The memory address where the storage unit is located corresponds to the backup address of the data in the backup space;

If an uncorrectable error occurs when data is read from the storage unit again, the data is read from the backup space according to the backup address corresponding to the memory address where the storage unit is located.

A second aspect of the present disclosure is to provide a device for backing up stored data in a memory, and a backup space is preset in the memory;

The device includes:

The backup unit is configured to write the data in the storage unit into the backup space if it is determined that there is a risk of uncorrectable errors occurring in the storage unit when reading data from the storage unit of the memory, and determine the storage unit. The memory address where the storage unit is located corresponds to the backup address of the data in the backup space;

A backup data reading unit, configured to read from the backup space according to the backup address corresponding to the memory address where the storage unit is located if an uncorrectable error occurs when reading data from the storage unit again fetch data.

A third aspect of the present disclosure is to provide an electronic device, comprising:

memory;

processor; and

Computer program;

Wherein, the computer program is stored in the memory and configured to be executed by the processor to implement the method for backing up stored data in the memory as described in the first aspect above.

A fourth aspect of the present disclosure is to provide a computer-readable storage medium on which a computer program is stored, the computer program being executed by a processor to implement the method of backing up stored data in a memory as described in the first aspect above method.

The technical effects of the method, device and electronic device for backing up stored data in memory provided by the present disclosure are:

The method, device and electronic device for backing up stored data in a memory provided by this embodiment, a backup space is preset in the memory; including: when reading data from a storage unit in the memory, if it is determined that there is an uncorrectable error in the storage unit If there is a risk of errors, write the data in the storage unit into the backup space, and determine the correspondence between the memory address where the storage unit is located and the backup address of the data in the backup space; if an uncorrectable error occurs when reading data from the storage unit again error, read data from the backup space according to the backup address corresponding to the memory address where the storage unit is located. In the solution provided by the present disclosure, the data of the storage unit can be backed up by using the space of the memory, and the size of the storage unit can be set according to the requirements, so that the electronic device can perform data backup on the data in the storage units scattered in the memory that may have UE errors. Backup to fix system downtime due to memory UE error.

Description of drawings

FIG. 1 is a schematic flowchart of a method for backing up stored data in a memory according to an exemplary embodiment of the present disclosure;

FIG. 2 is a flowchart of a method for backing up stored data in a memory according to another exemplary embodiment of the present disclosure;

3 is a schematic diagram of data backup according to an exemplary embodiment of the present disclosure;

4 is a schematic diagram of reading data according to another exemplary embodiment of the present disclosure;

5 is a structural diagram of an apparatus for backing up stored data in a memory according to an exemplary embodiment of the present disclosure;

6 is a structural diagram of an apparatus for backing up stored data in a memory according to another exemplary embodiment of the present disclosure;

FIG. 7 is a structural diagram of an electronic device according to an exemplary embodiment of the present disclosure.

Detailed ways

The memory component is a key component in the computer architecture that directly provides data cache to the central processing unit. The failure of the memory system directly causes the computer's processor to stop responding, resulting in system downtime.

There are UE (Uncorrectable Errors, uncorrectable errors) errors in memory failures. Such errors cannot be corrected based on the existing ECC ("Error Correcting Code, Error Checking and Correction) technology. In order to ensure system reliability and The accuracy of the data, the system is very likely to restart.

If the memory set in the server is restarted due to a UE error, services will be interrupted, resulting in a bad user experience.

In order to solve this technical problem, data backup can be performed based on memory mirroring. When memory access fails, the content in the corresponding backup block is used. However, the disadvantage of this method is that the backup space is required to be continuous, resulting in the smallest unit of backup. It is relatively large, which reduces the memory space that the user can actually use.

In another solution, the wrong bit (bit) in the memory can be replaced by the bit stored in the memory controller. However, the storage space in the memory controller is limited, resulting in the high cost of this single-bit backup method, which can only be stored Fewer backup bits.

In another solution, the redundant circuit in the memory can be used to replace the bad row of the content. Specifically, the bad row in the memory can be remapped to the row with built-in redundancy. However, this approach has a limited number of rows for backup, and this increases hardware costs.

In order to solve the above technical problems, in the solution provided by the present disclosure, a backup space is pre-divided in the memory. When it is determined that a UE error may occur at the memory address where the storage unit is located, the data of the storage unit can be written into the backup space, and the determination of The mapping relationship between the memory address where the storage unit is located and the backup address, so as to realize the purpose of backing up the data in the storage unit. This implementation method does not increase additional hardware costs, and the size of the storage unit can be set according to requirements, and the implementation method is more flexible.

FIG. 1 is a schematic flowchart of a method for backing up stored data in a memory according to an exemplary embodiment of the present disclosure.

The method provided by the present disclosure can be applied to an electronic device provided with a memory, and when the electronic device is powered on, the electronic device can divide a backup space in the memory. For example, the size of the backup space can be preset, so that the electronic device can divide the backup space of a corresponding size in the memory. The address of the backup space in the memory may be a continuous address or a non-continuous address.

As shown in FIG. 1 , the method for backing up stored data in memory provided by the present disclosure includes:

Step 101: When reading data from the storage unit of the memory, if it is determined that there is a risk of uncorrectable errors occurring in the storage unit, the data in the storage unit is written into the backup space, and it is determined that the memory address of the storage unit and the data are in the backup space. The backup address in the space corresponds.

Among them, the storage components in the electronic device may include hard disks, memory, cache, etc., wherein, when the electronic device reads data, it can read data from the cache first, if there is no cache, read it from the memory, if there is no more data in the memory. read from the hard drive.

Specifically, the electronic device may store the more commonly used data in the cache or the memory, so as to improve the data reading speed. Therefore, electronic devices typically have data stored in memory during runtime.

Further, when the data to be read is stored in the memory, the electronic device can read the data from the memory. For example, when a user operates a picture and the picture is stored in the memory, the electronic device can read the data of the picture from the memory.

In practical applications, the memory can be divided into multiple storage units, the data to be read can be scattered in multiple storage units, and the electronic device can read data from these storage units. The size of the storage unit can be set according to requirements, for example, it can be 32B, 64B, 72B and so on.

In an optional implementation manner, the size of the storage unit may be the size of the smallest amount of data when the electronic device reads data. For example, when the electronic device writes the data in the memory into the cache, at least 64B (byte, byte) is written each time, and the size of the storage unit is also 64B.

In another optional implementation manner, the size of the storage unit may also be set based on requirements, for example, it may be 64B*2, 64B*3, and so on.

Wherein, when the electronic device reads data from the storage unit, if it is determined that there is a risk of UE error occurring in the storage unit, the data in the storage unit can be written into the backup space.

Specifically, the electronic device can use the solutions in the prior art to determine whether the storage unit has the risk of UE errors. For example, if a storage unit has multiple CE errors, it can be determined that the storage unit has the risk of UE errors.

Further, the storage unit has a corresponding memory address. For example, a space with a memory address of 0-63B is the first storage unit, and a space with a memory address of 64-127B is the second storage unit. If the space with memory address 0-63B has CE errors many times, such as one CE error in the 4th byte, and 8 CE errors in the 10th byte, it can be determined that the memory address is 0- There is a risk of UE error occurring in the space of 63B, and it is further determined that there is a risk of UE error occurring in the storage unit corresponding to the memory address 0-63B.

In practical application, if it is determined that there is a risk of UE error in the storage unit, the data in the storage unit is backed up to the backup space, and specifically, the data in the storage unit may be written into the backup space.

Wherein, each time the data in the storage unit is written to the backup space, the data may be written in the backup space in sequence. For example, if it is determined that the first storage unit has a UE error risk, the data in the first storage unit may be written to After that, if it is determined again that the fourth storage unit has the risk of UE error, the data in the fourth storage unit can also be written into the backup space, and the data in the fourth storage unit can be arranged in the first storage unit. after the data.

Specifically, if there is a risk of UE errors in the storage unit, it can be considered that the data stored in the storage unit has a risk of being unreadable. At this time, backing up the data in the storage unit can prevent the UE from actually appearing in the storage unit. When there is an error, the corresponding data is read from the backup space to avoid the problem of downtime caused by UE errors in memory.

Further, after the data in the storage unit is backed up to the backup space, the correspondence between the source address of the data and the backup address can also be determined. For example, it can be determined that there is a mapping relationship between the two. Specifically, the memory address where the storage unit is located can be determined. It is determined as the source address of the data, and the address of the data in the backup space is used as the backup address.

For example, the memory address of the first storage unit is 0-63B, the data in the first storage unit is written in the backup space, and the address of the data is N-(N+63B), then determine the address (0-63B ) corresponds to the address (N-(N+63B)).

In practical application, if it is determined that there is a mapping relationship between addresses, an address mapping table can be generated, and the determined mapping relationship can be written into the address mapping table.

In this implementation, data can be backed up in units of storage units, and the granularity of the storage units can be controlled according to requirements, for example, can be controlled to 64B, so that data with the smallest read granularity can be backed up.

The memory address where the storage unit is located and the backup address of the data refer to physical addresses.

Step 102: If an uncorrectable error occurs when data is read from the storage unit again, read data from the backup space according to the backup address corresponding to the memory address where the storage unit is located.

Among them, if a UE error occurs when reading data from the storage unit again, it can be determined whether the data in the storage unit has been backed up, and if it has been backed up, the corresponding data can be read from the backup space to avoid system downtime resulting in business interruption.

Specifically, a backup address corresponding to the memory address where the storage unit is located can be determined, and corresponding data can be read from the backup space based on the backup address. For example, if the first source address corresponds to the first backup address, and a UE error occurs when data is read from the storage unit, and the memory address where the storage unit is located is the first source address, it can be read from the first backup address in the backup space. Take backup data.

In an optional implementation manner, after the data in the storage unit is backed up to the backup space, a mark may be added to the storage unit, for example, a backup mark may be added.

If a UE error occurs when reading data from the storage unit again, and the electronic device determines that the storage unit has a backup mark, the electronic device can read from the backup address of the backup space according to the backup address corresponding to the memory address where the storage unit is located. To obtain data, for example, a backup address corresponding to the memory address where the storage unit is located can be determined according to the mapping relationship, and then data is obtained from the backup address.

In another optional implementation manner, if a UE error occurs when data is read from the storage unit again, the electronic device can determine the memory address where the storage unit is located, and according to the backup address corresponding to the memory address where the storage unit is located , and then get the data from the backup address.

In this implementation, even if a UE error occurs in the storage unit, the system will not be down.

In the solution provided by the present disclosure, the storage unit can be pre-judged to determine the storage unit with the risk of UE error, so that the data of the storage unit with the risk of UE error can be backed up. When the UE error occurs in the storage unit, The corresponding data can be read from the backup space, so as to avoid the problem that the data cannot be obtained due to the UE error of the storage unit.

The method for backing up stored data in a memory provided by the present disclosure includes presetting a backup space in the memory; the method includes: when reading data from a storage unit in the memory, if it is determined that there is a risk of uncorrectable errors occurring in the storage unit, then Write the data in the storage unit into the backup space, and determine the correspondence between the memory address where the storage unit is located and the backup address of the data in the backup space; if an uncorrectable error occurs when reading the data from the storage unit again, according to The backup address corresponding to the memory address where the storage unit is located, reads data from the backup space. In the solution provided by the present disclosure, the space of the memory can be used to back up the data of the storage unit, and the size of the storage unit can be set according to the requirements, so that the electronic device can perform the data backup of the data in the storage units scattered in the memory and possibly having UE errors. Backup to fix system downtime due to memory UE error.

FIG. 2 is a flowchart of a method for backing up stored data in a memory according to another exemplary embodiment of the present disclosure.

As shown in FIG. 2 , the method for backing up stored data in the memory provided by this embodiment includes:

Step 201, when reading data from the storage unit of the memory, if it is determined that the storage unit has the risk of a correctable error, then determine that the memory can correct the error information; risk of error.

The method provided by the present disclosure can be performed by an electronic device provided with a memory, and the electronic device can divide a backup space in the memory when the electronic device is powered on. The size of the backup space can be set based on demand to meet the business needs of various application scenarios.

The memory can be divided into multiple storage units, the data to be read can be scattered in multiple storage units, and the electronic device can read data from these storage units.

Specifically, when the electronic device reads data from the storage unit, if it is determined that the storage unit has the risk of CE error, the electronic device can determine the correctable error information in the memory, and the correctable error information in the memory may include the location and cause of the error. Wait.

This memory correctable error information is used to determine whether the storage unit is at risk of UE errors.

In an optional implementation manner, a management program may be set in the electronic device, and the management program in the electronic device may determine whether the risk of UE error occurs in the storage unit according to the memory correctable error information.

In another optional implementation manner, the electronic device may also send memory-correctable error information to the cloud, and the cloud determines whether a risk of UE error occurs in the storage unit according to the memory-correctable error information.

Further, a prediction strategy can also be set, and when the memory correctable error information is used to determine whether there is a risk of UE errors occurring in the storage unit, the prediction strategy and the memory correctable error information can be combined to determine whether there is a UE error in the storage unit. risk.

For example, if it is determined that CE errors have occurred in multiple bytes in the storage unit according to the correctable error information in the memory, it can be determined whether there is a risk of UE errors occurring in the storage unit. For example, according to the historically received memory correctable error information and the currently received memory correctable error information, it can be determined whether there are multiple bytes in the storage unit corresponding to the currently received memory correctable error information and CE errors have occurred , if yes, it can be determined that there is a risk of UE error occurring in the storage unit.

Step 202: Obtain information for characterizing the risk of uncorrectable errors occurring in the storage unit.

Wherein, if the management program of the electronic device determines that the storage unit has the risk of sending UE errors, the management program may also generate information for representing the risk of uncorrectable errors occurring in the storage unit, so that the electronic device can obtain the information .

The management program can be set in the electronic device or in the cloud. If it is set in the electronic device, it can be set in the BIOS (Basic Input Output System), or in the BMC (Baseboard Management Controller). , baseboard management controller), or set in an OS (operating system, operating system), which is not limited in the present disclosure.

Step 203: Write the data in the storage unit into the first backup address and the second backup address of the backup space.

Further, when data is written into the backup space, the data can be backed up twice, for example, the data can be written in the backup space twice continuously, thereby making the data more secure.

In practical application, the data in the storage unit can be written in the first backup address and the second backup address of the backup space, and the data stability can be further ensured by means of multiple backups. For example, the data in the backup space may include data A, data A', data B, and data B', wherein data A and data A' are the same, and data B and data B' are the same.

Step 204, adding a backup mark to the storage unit; the backup mark is used to represent the data of the storage unit to be written into the backup space.

In the method provided by the present disclosure, after the data of the storage unit is backed up to the backup space, the electronic device may further add a backup mark to the storage unit, thereby indicating that the data in the storage unit is backed up in the backup space.

For example, if there is a first storage unit, and the electronic device stores data in the first storage unit in a backup space, a backup mark may be added to the first storage unit. For another example, if there is a fourth storage unit, and the electronic device stores data in the fourth storage unit in the backup space, a backup mark may be added to the fourth storage unit.

FIG. 3 is a schematic diagram of data backup according to an exemplary embodiment of the present disclosure.

As shown in FIG. 3 , there is a backup space 31 in the memory, and the size of the backup space 31 is larger than the size of the storage unit.

If the electronic device determines that the storage unit 32 may have a risk of UE error, it writes the data data32 in the storage unit 32 into the backup space 31 . Thereafter, if the electronic device further determines that the storage unit 33 may have a risk of UE error, the data data33 in the storage unit 33 is also written into the backup space 31 .

It should be noted that the addresses in the backup space for backing up data may be continuous or discontinuous. For example, data32 may be written to the third backup address, and data33 may be written to the fourth backup address, and the third backup address and the fourth backup address may be discontinuous.

Wherein, after acquiring the information representing the risk of uncorrectable errors occurring in the storage unit, the electronic device may also add a backup mark to the storage unit, for example, may add a backup mark to both the storage unit 32 and the storage unit 33 .

Step 205: Generate an address mapping item according to the memory address where the storage unit is located and the backup address used for backing up data in the backup space.

The execution sequence of steps 203-205 is not limited, that is, the execution sequence of backing up data, adding a backup mark to a storage unit, and generating an address mapping item is not limited.

In an optional implementation, when data needs to be backed up, a backup address for backing up data can be determined first, and then a mapping item between the memory address where the storage unit is located and the backup address can be generated, and then the data that needs to be backed up is written. into the backup address.

In another optional implementation manner, when data needs to be backed up, data can be written in the backup space first, and then the backup address can be determined according to the location of the written data, and an address mapping item can be generated.

Further, the electronic device can determine the memory address where the storage unit is located and the backup address used for backing up data in the backup space, and generate an address mapping item. For example, if the memory address where the storage unit is located is address 1, the data in the storage unit will be written into the backup space, and the written backup address is address 2, then an address mapping item can be generated according to address 1 and address 2, and the address mapping The entry is used to represent that address 1 corresponds to address 2.

Further, if two or more copies of the same data are backed up in the backup space, the electronic device can also determine the correspondence between the memory address where the storage unit is located and the first backup address, for example, the memory address where the storage unit is located is determined to be the same as the first backup address. The first backup addresses have a first mapping relationship, and the correspondence between the first backup address and the second backup address can also be determined, for example, it is determined that the first backup address and the second backup address have a second mapping relationship. Both the first mapping relationship and the second mapping relationship can be written into the address mapping table.

For example, if the memory address of the storage unit is address 1, the data in the storage unit is stored in the backup space, and the specific storage location is address 2, it can be determined that there is a first mapping relationship between address 1 and address 2, and at the same time, If the data is also stored at another location address 3 in the backup space, it can be determined that there is a second mapping relationship between address 2 and address 3.

In this implementation, if a UE error occurs when reading the data of the storage unit, the electronic device can read the backup data from the first backup address. If a UE error also occurs when reading data from the first backup address, then The electronic device can also read the backup data from the second backup address. Therefore, the problem of system downtime caused by UE error occurs in the address used for backing up data in the backup space is solved.

Step 206, write the address mapping entry into the address mapping table.

In practical application, an address mapping table may also be generated for recording address mapping items.

The electronic device can determine whether there is currently an address mapping table capable of writing address mapping items, and if there is no address mapping table, the electronic device can also generate a new address mapping table and write the address mapping items in it.

If there is an address mapping table in which the address mapping item can be written, the electronic device can write the currently generated address mapping item into the address mapping table.

The electronic device can also determine the hash value according to the memory address where the storage unit is located, and determine the target position in the address mapping table according to the hash value; and store the address mapping item in the target position of the address mapping table.

Specifically, the electronic device may determine the hash value according to the low-order value of the memory address, for example, the hash value may be determined by using the low-order 4-bit value of the memory address. Then use the hash value as the offset item of the mapping table to determine the target position. For example, if the hash value is hash(m), the hash(m) row of the address mapping table can be used as the target location. In this implementation, when the address mapping item of the storage unit needs to be queried, the target position in the address mapping table can be quickly located according to the memory address of the storage unit, so that the address mapping item can be quickly read so as to quickly read the backup address.

Further, if the address mapping items are recorded by means of an address mapping table, the address mapping table can be stored in the memory, and a large number of address mapping items can be stored in this manner.

In an optional implementation, two copies of the address mapping table can be stored in the memory, for example, the address mapping table is stored in the first position of the memory, and the backup table can also be stored in the second position. The contents of the mapping table are the same.

The address mapping table is a logical table, and the contents stored in the storage medium are address mapping items. The address mapping table can be regarded as a collection of address mapping items. The addresses of these address mapping items in the storage medium can be continuous or discontinuous.

In practical application, every time an address mapping item is added to the address mapping table, an address mapping item can also be added to the backup table in the same way, so that the contents of the two tables are synchronized.

Among them, some locations may also be reserved in the cache of the electronic device for storing some address mapping items. For example, a location capable of storing N address mapping items may be reserved, so that the electronic device can read the backup address more quickly.

Specifically, the address mapping items stored in the cache can be updated, and specifically, the recently read address mapping items can be stored in the cache.

After steps 2-3, 204, and 206, step 207 may be performed, and step 209 may also be performed.

Step 207: If an uncorrectable error occurs when reading data from the storage unit, and the storage unit has a backup mark, read data from the backup address of the backup space according to the mapping relationship.

Wherein, if the UE error occurs when the electronic device reads data from the storage unit, the electronic device can determine whether the storage unit has a backup mark, and if so, it means that the data in the storage unit has been backed up to the backup space, therefore, Data can be read from the backup address of the backup space according to the mapping relationship.

Specifically, if the address mapping table is used to record the address mapping relationship, the electronic device can query the address mapping table for the backup address corresponding to the memory address where the storage unit is located, and read data from the backup address.

In this implementation manner, even if a UE error occurs in the storage unit storing the data, the electronic device can still read the corresponding data from the backup space to avoid system restart.

Optionally, if the electronic device successfully reads the data from the storage unit, the electronic device can also store the data in the cache, and record the information that the storage unit has a backup mark in the cache, so that the data is synchronized from the cache next time. When reaching the memory storage unit, it can be determined that the data has been backed up in the backup space, so the electronic device can synchronize the data to the backup address according to the address mapping table.

FIG. 4 is a schematic diagram of reading data according to another exemplary embodiment of the present disclosure.

As shown in FIG. 4 , data42 is stored in the backup space 41 , data42 is data in the storage unit 42 , and data43 is also stored in the backup space 41 , and data43 is data in the storage unit 43 .

The correspondence between the backup address of data42 in the backup space 41 and the memory address of the storage unit 42 is recorded, and the correspondence between the backup address of data43 in the backup space 41 and the memory address of the storage unit 43 is also recorded.

When the electronic device reads data from the storage unit 42, if a UE error occurs, the electronic device can query the backup address corresponding to the memory address of the storage unit 42 according to the mapping relationship of the addresses, and the backup address of the data42 in the backup space 41 is the same as the one in the backup space 41. The memory addresses of the storage units 42 correspond to each other, so the data 42 can be read from the backup address.

Step 208 , if an uncorrectable error occurs when the data is read from the backup space, output information representing that the memory has an uncorrectable error.

Among them, the backup space is a part of the memory. Therefore, UE errors may also occur in the backup space. If a UE error occurs when reading data from the backup space, the electronic device cannot obtain valid data in the end. Therefore, the electronic device can output data for Information that characterizes the existence of UE errors in memory.

Step 209: In response to the write instruction to the storage unit, if it is determined that the storage unit has a backup flag, determine the backup address corresponding to the memory address where the storage unit is located.

Specifically, the electronic device can also perform a write operation on the storage unit, and specifically can execute a write instruction. For example, if the user can change the data stored in the electronic device, the electronic device can execute the write instruction to the storage unit based on the operation. For another example, when the electronic device executes the program, the data in the storage unit will also be updated, and the data in the storage unit will also be updated during the execution process. A write operation is performed on the storage unit, and a corresponding write instruction can be specifically executed.

Further, when the electronic device executes the write instruction to the storage unit, if it is determined that the storage unit has a backup flag, the backup address corresponding to the memory address where the storage unit is located may be determined.

For example, the backup address corresponding to the memory address where the storage unit is located can be determined according to the determined mapping relationship between the addresses. If the address mapping table is used to record the address mapping relationship, the electronic device can query the address mapping table with the storage unit. The backup address corresponding to the memory address where it is located.

Step 210: Write data in the storage unit and the backup address according to the write instruction.

In practical application, the electronic device may write data in the storage unit according to the write instruction, and may also write data in the backup address corresponding to the storage unit. In this implementation manner, the data in the storage unit and the backup space can be guaranteed to be consistent.

In any of the above embodiments, the size of the storage space of the storage unit is the same as the size of the cache line of the cache. The cache is another storage component in the electronic device. The cache is divided into multiple groups, and each group has multiple cachelines (cache lines). The size of the cache line is the size of one cacheline. The electronic device writes the data in the memory. When caching, the minimum granularity of data is linesize, that is, each time data is written to the cache, the minimum amount of data is linesize.

In this implementation, the data in the storage unit that may have the risk of UE error in the memory can be backed up, and the size of the storage unit is the size of the cacheline, so a small amount of backup space can be used to scatter the data in the memory that may be wrong. Each data block with the size of linesize is backed up to improve the reliability and stability of the electronic device.

FIG. 5 is a structural diagram of an apparatus for backing up stored data in a memory according to an exemplary embodiment of the present disclosure.

As shown in FIG. 5 , the device 500 for backing up stored data in the memory provided by this embodiment has a backup space preset in the memory; the device includes:

The backup unit 510 is configured to, when reading data from a storage unit in the memory, write the data in the storage unit into the backup space if it is determined that there is a risk of uncorrectable errors occurring in the storage unit, and determine the storage unit. The memory address where the storage unit is located corresponds to the backup address of the data in the backup space;

The backup data reading unit 530 is configured to read data from the backup space according to the backup address corresponding to the memory address where the storage unit is located if an uncorrectable error occurs when reading data from the storage unit again. fetch data.

In the solution provided by the present disclosure, the space of the memory can be used to back up the data of the storage unit, and the size of the storage unit can be set according to the requirements, so that the electronic device can perform the data backup of the data in the storage units scattered in the memory and possibly having UE errors. Backup to fix system downtime due to memory UE error.

FIG. 6 is a structural diagram of an apparatus for backing up stored data in a memory according to another exemplary embodiment of the present disclosure.

As shown in FIG. 6 , on the basis of the foregoing embodiment, the apparatus 600 for backing up stored data in a memory provided by this embodiment, the backup unit 510 includes:

The mapping item generating module 511 is used for generating an address mapping item according to the memory address where the storage unit is located and the backup address for backing up the data in the backup space;

The writing module 512 is configured to write the address mapping item into the address mapping table.

The writing module 512 is specifically used for:

Determine a hash value according to the memory address where the storage unit is located, and determine a target position in the address mapping table according to the hash value;

The address mapping entry is stored in the target location of the address mapping table.

In an optional implementation manner, the address mapping table is stored in a first location of the memory, and a backup table identical to the address mapping table is stored in a second location of the memory.

Optionally, the memory is provided in an electronic device, and a cache is also provided in the electronic device;

The apparatus further includes a cache unit 530 for:

N pieces of the address mapping entries in the address mapping table are stored in the cache, where N is a positive integer.

Optionally, the device further includes a marking unit 540 for:

adding a backup mark to the storage unit; the backup mark is used to represent that data of the storage unit is written into the backup space;

The backup data reading unit 520 includes a reading module 521 for:

If an uncorrectable error occurs when the data is read from the storage unit, and the storage unit has the backup flag, then according to the backup address corresponding to the memory address where the storage unit is located, from the backup space read data in.

The apparatus also includes a writing unit 550 for:

In response to the write instruction to the storage unit, if it is determined that the storage unit has the backup mark, then determine the backup address corresponding to the memory address where the storage unit is located;

Data is written in the storage unit and the backup address according to the write instruction.

Optionally, the backup data reading unit 520 further includes an output module 522 for:

If an uncorrectable error occurs when the data is read from the backup space, output information representing that the memory has an uncorrectable error.

Optionally, the memory is provided in an electronic device, and a cache is also provided in the electronic device;

The size of the storage space of the storage unit is the same as the size of the cache line of the cache.

Optionally, the backup unit 510 includes:

The correctable error determination module 513 is configured to, when reading data from the storage unit of the memory, determine that the storage unit is at risk of having a correctable error, determine that the memory can correct the error information; the memory may Correcting error information is used to determine whether the storage unit is at risk of uncorrectable errors;

The error information acquisition module 514 is configured to acquire information used to characterize the risk of uncorrectable errors occurring in the storage unit.

Optionally, the backup unit 510 is specifically used for:

writing the data in the storage unit into the first backup address and the second backup address of the backup space;

It is determined that the memory address where the storage unit is located corresponds to the first backup address, and it is determined that the first backup address corresponds to the second backup address.

FIG. 7 is a structural diagram of an electronic device according to an exemplary embodiment of the present disclosure.

As shown in FIG. 7 , the electronic device provided in this embodiment includes:

memory 71;

processor 72; and

Computer program;

Wherein, the computer program is stored in the memory 71 and is configured to be executed by the processor 72 to implement any of the methods for backing up stored data in the memory as described above.

This embodiment also provides a computer-readable storage medium on which a computer program is stored,

The computer program is executed by the processor to implement any of the methods for backing up stored data in memory as described above.

This embodiment also provides a computer program, including program code. When the computer runs the computer program, the program code executes any of the above-mentioned methods for backing up stored data in the memory.

Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by program instructions related to hardware. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps including the above method embodiments are executed; and the foregoing storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, but not to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features thereof can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present disclosure. scope.

Claims (14)

1. A method for backing up stored data in a memory is characterized in that a backup space is preset in the memory;

the method comprises the following steps:

when data are read from a storage unit of a memory, if the risk that uncorrectable errors occur in the storage unit is determined, writing the data in the storage unit into the backup space, and determining that a memory address where the storage unit is located corresponds to a backup address of the data in the backup space;

and if uncorrectable errors occur when the data are read from the storage unit again, reading the data from the backup space according to the backup address corresponding to the memory address of the storage unit.

2. The method according to claim 1, wherein the determining that the memory address of the storage unit corresponds to the backup address of the data in the backup space comprises:

generating an address mapping item according to the memory address of the storage unit and a backup address for backing up the data in the backup space;

and writing the address mapping item into an address mapping table.

3. The method of claim 2, wherein the address mapping table is stored in the memory.

4. The method of claim 2, wherein writing the address mapping entry to an address mapping table comprises:

determining a hash value according to the memory address of the storage unit, and determining a target position in the address mapping table according to the hash value;

storing the address mapping entry in the target location of the address mapping table.

5. The method of any of claims 2-4, wherein the address mapping table is stored at a first location in the memory, and wherein a backup table that is the same as the address mapping table is stored at a second location in the memory.

6. The method according to any one of claims 2-4, wherein the memory is provided in an electronic device, and the electronic device is further provided with a cache;

the method further comprises the following steps:

and storing N address mapping items in the address mapping table in the cache, wherein N is a positive integer.

7. The method according to any one of claims 1-4, further comprising:

adding a backup mark to the storage unit; the backup marks are used for representing data of the storage units and written into the backup space;

if an uncorrectable error occurs when reading data from the storage unit again, reading data from the backup space according to a backup address corresponding to a memory address where the storage unit is located, including:

and if an uncorrectable error occurs when the data is read from the storage unit, and the storage unit has the backup mark, reading the data from the backup space according to a backup address corresponding to the memory address of the storage unit.

8. The method of claim 7, further comprising:

responding to a write instruction of the storage unit, and if the storage unit is determined to have the backup mark, determining a backup address corresponding to a memory address where the storage unit is located;

and writing data in the storage unit and the backup address according to the writing instruction.

9. The method according to any one of claims 1 to 4, wherein if an uncorrectable error occurs when data is read from the backup space, outputting information for indicating that the memory has the uncorrectable error.

10. The method according to any one of claims 1-4, wherein the memory is provided in an electronic device, and the electronic device is further provided with a cache;

the size of the storage space of the storage unit is the same as the size of the cache line of the cache.

11. The method according to any one of claims 1-4, wherein determining that the storage unit of the memory is at risk of having an uncorrectable error when reading data from the storage unit comprises:

when data are read from the storage unit of the memory, if the storage unit is determined to have risk of error which can be corrected, error information which can be corrected by the memory is determined; the memory correctable error information is used for determining whether the storage unit has risk of generating uncorrectable errors;

information characterizing that the memory cell is at risk of an uncorrectable error is obtained.

12. The method according to any one of claims 1 to 4,

the writing the data in the storage unit into the backup space includes:

writing the data in the storage unit into a first backup address and a second backup address of the backup space;

the determining that the memory address of the storage unit corresponds to the backup address of the data in the backup space includes:

determining that the memory address of the storage unit corresponds to the first backup address;

the method further comprises the following steps: determining a correspondence between the first backup address and the second backup address.

13. A device for backing up stored data in a memory is characterized in that a backup space is preset in the memory;

the device comprises:

the backup unit is used for writing the data in the storage unit into the backup space and determining that the memory address of the storage unit corresponds to the backup address of the data in the backup space if the risk of uncorrectable errors of the storage unit is determined when the data is read from the storage unit of the memory;

and the backup data reading unit is used for reading the data from the backup space according to the backup address corresponding to the memory address of the storage unit if an uncorrectable error occurs when the data is read from the storage unit again.

14. An electronic device, comprising:

a memory;

a processor; and

a computer program;

wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of any of claims 1-12.

Images